Here are my 5 predictions for AI in 2026

1. Anthropic is the biggest winner in AI

While AI researchers and pundits debate who is ahead on the Arena Elo or Stanford HELM leader boards, Claude will not so quietly eat the model deployment war at the application layer.

With this being the year of the Agent, I predict that enterprises that are generally risk-averse will be shopping for the safest models, not the shiniest models, to deploy AI within their organizations. Leaders want fewer surprises, fewer legal headaches, and fewer random hallucinations showing up in customer emails.

Anthropic’s whole vibe from day 1 has been safety first, and they have been building AI for business use cases. This year, that will shine through, and they will extend their lead in model deployment across businesses.

If you are a small business owner exploring different vendors for AI agents this year, don’t just go for the trendy tool. Pick the ones that let you control the output, have verifiable audit trails, and reduce risk. If your AI touches customers, money, medical info, legal stuff, or anything sensitive, treat safety like a feature you pay for, not a nice-to-have.

2. “Digital Employees” becomes a buzzword

In an interview released earlier this month, McKinsey’s managing partner Bob Sternfels said the company now has 60,000 employees and 20,000 of those are AI Agents. Look for this to be a trend this year as more companies push forward to adopt agentic AI to gain an edge for growth or appease their board.

This will be the year Agents go mainstream, but companies won’t just be looking for bots that automate one task; they will be looking for Agentic systems that can collaborate with humans to complete work.

I predict we will see more and more startups like Viven AI building digital doubles—AI Agents with the tribal knowledge of their human counterparts to execute work and preserve institutional knowledge.

While this change will start at the enterprise level, we will also see small business owners for the first time adopt agents in the form of customer support agents or personal assistants that can read and send emails, book and prepare you for calls, and handle other admin tasks that would have needed a human hire in the past.

3. Customer support agents become the #1 AI use case for small businesses

Voice AI has been rapidly improving over the past few years, and this is the year customer support agents will have their moment in the spotlight, similar to what has been happening with coding agents over the past two years.

These agents will handle more than FAQs. They’ll do order status, intake, basic troubleshooting, and appointment changes, and they’ll learn from prior conversations and handle complex support scenarios on the fly. In many cases, it will become completely indistinguishable whether you are talking to a human or an AI agent.

This will become the #1 AI use case for small businesses that historically have been limited in their support capability for customers.

4. The creator economy gets a second wind buoyed by agentic tools

When you hear about the creator economy these days, the first thing that comes to mind is probably teenagers doing the latest dance challenge on TikTok. That’s not what I mean. In 2026, AI tools will make it so normal business owners and independent creators will ship content and small digital products without hiring a full team.

The gap between ideation and creation will disappear, and more content and personalized digital products will be created than at any time in the history of mankind.

5. The AI Job apocalypse doesn’t happen

I expect more job cuts to be announced by the hyperscalers and other large enterprises this year. Many jobs are going to get squeezed out, and companies will continue to hire slowly.

On the other side, businesses that deploy Agents will also see higher growth forecasts. More growth will require humans who can collaborate symbiotically with AI Agents to support that growth. So, though there will be some major cuts, I also anticipate companies will hire more in some roles to support their growth. This will net out some of the losses and create more balance in the job market this year than many anticipate.

2025 has been a whiplash of the highest highs and the lowest lows for me. I experienced the greatest joy anyone can ever wish for - bringing new life into the world with the birth of my daughter. But I was also broken by the loss of life, struggling through the grief of losing my niece.

I have felt lost, anxious, happy, sad, excited, and everything in between. Since October, I took some time off writing to care for myself, spend time with my newborn daughter, and mourn the loss of my niece.

Through it all, I have come to realize that we must embrace every season of our lives. This volatility is a feature not a bug; it is the very thing that makes us feel alive. It is what separates us from the other species, and certainly from the AI models we discuss here. If you are reading this and you have felt that same whiplash this year, know that you are not alone. And know that you aren't doing it wrong. You are just doing it human.

But I didn’t want to end the year in silence. I wanted to end it on a note of gratitude.

So, I wanted to share what we have achieved this year together. The AI For Mainstreet edition of 2025 WARP.

Here is the story of our 2025!

2025 WARP

HAPPY HOLIDAYS! SEE YOU IN 2026

Two weeks ago, we released our first research report based on interviews with dozens of senior finance leaders across the country. So far, it’s been the most-read newsletter to date. We also gained over 50 new subscribers in a week, and I’m excited to keep this momentum going. If you haven’t read the report yet, you can find it here.

Some of you reached out via WhatsApp and LinkedIn to share your thoughts. Some expressed concern about the future role of finance professionals, while others were excited about what AI could mean for financial operations.

I understand the anxiety around AI. While I’m incredibly excited about this technology, I don’t believe AI will replace human judgment or lived experience. In fact, most research shows that the people getting the most out of AI are those who already know their craft well and understand how to use these tools effectively. AI isn’t sentient, it’s built and trained on the foundation of human judgment and experience. It can serve as a powerful companion to help us achieve more in our work and lives.

In my second ever blog post, which I’ve reshared below, I describe AI as amplification intelligence, a clone of ourselves that helps us do far more than our human capacity alone would allow. What this means practically is:

• A boutique accounting firm that can’t afford to hire a new CPA can now scale their operations without adding headcount

• A solo real estate agent can provide 24/7 client communication and property research that previously required a team

• A local law firm can draft contracts and conduct legal research at speeds that once required multiple associates

• A fractional CFO can serve twice as many clients without sacrificing quality or burning out

Embracing AI doesn’t mean the end of human work, it means unlimited abundance through amplification of human output.

MY SECOND BLOG POST

Executive Summary

Fractional CFOs work on the frontlines of small and mid-sized main street businesses. They walk into companies that often have no structured processes, inherit messy data, and somehow need to deliver clarity and forward-looking guidance. To understand how AI might change this work, we spoke with 30 fractional CFOs across different industries and geographies within the US.

What we found was clear: trust, not technology, is what's holding them back. These CFOs are fine with AI handling routine bookkeeping and categorizing transactions, but they hesitate when it comes to forecasting or making strategic decisions. Excel is still their go-to tool for anything mission-critical, even though AI tools are becoming more available.

Five things stood out from these conversations:

1. Messy data is everywhere. CFOs spend way too much time just cleaning and organizing information before they can even start analyzing it.

2. Reporting eats up their time. Many spend up to 10 hours per week per client just preparing board presentations and variance reports.

3. Forecasting tools are lacking. CFOs told us that 13-week cash flow projections and scenario planning are critical gaps they need filled despite emerging tools In the market.

4. There's a real trust issue with AI. Classification tasks feel safe to them, but forecasting feels too risky.

5. QuickBooks runs everything. Nearly every small business uses it, so any AI tool has to work seamlessly with it.

The takeaway from these conversations was resounding; AI tools for CFOs need to start by earning trust. Tools that deliver immediate, accurate wins in data cleanup, transaction categorization, and variance analysis will get adopted. Many CFOs felt that the big prize for AI is in strategic insights and decision support that drives real outcomes for small businesses.

The role of the Chief Financial Officer is changing and nowhere is this more visible than in the world of fractional CFOs. Increasingly, small and mid-sized businesses are turning to part-time financial leaders to bring order to their books, accelerate reporting, and provide the kind of forward-looking insight once reserved for large enterprises.

Introduction

Fractional CFOs step into businesses where financial systems are often lacking. Accounting data lives across QuickBooks, spreadsheets, filing cabinets, shoeboxes of receipts, paper trail and bank portals. Documentation is incomplete, controls are inconsistent, and reporting is largely reactive. Before they can advise on growth or strategy, CFOs spend countless hours cleaning data, reconciling accounts, and assembling reports.

At the same time, demand for fractional CFO services is growing. industry data shows U.S. demand for fractional CFOs has more than doubled year-over-year. As businesses scale, they need financial clarity but cannot afford full-time executives. This creates both opportunity and pressure for fractional CFOs who are expected to do more with less, often supporting 7 to 15 clients at a time, with small teams or alone. The result is a constant trade-off between deep strategic work and manual, controller-level tasks.

Against this backdrop, AI has emerged as a potential lever for transformation. Vendors promise automation of bookkeeping, instant variance explanations, and predictive forecasting. Yet despite enthusiasm, adoption is uneven. CFOs told us they are intrigued but cautious.

This report explores these realities in depth. Drawing from interviews with fractional CFOs across industries, we identify the most pressing challenges, analyze where AI can credibly help today, and outline what must happen before CFOs trust AI with higher-order work.

Methodology

This report is based on in-depth interviews with 30 fractional CFOs across industries and geographies. Conversations were semi-structured. Notes and transcripts were coded for recurring themes, which were then synthesized into insights.

While the sample is not statistically representative, the findings highlight clear, recurring patterns in how Fractional CFOs view their challenges and the role of AI.

The State of the Fractional CFO Today

Fractional CFOs occupy a unique space in business finance. They are asked to deliver the strategic insight of a full-time CFO but with the operational burdens of a controller. The result is a constant tension between where they feel they add the most value and where they spend most of their time.

From our interviews, CFOs reported that the majority of their week is consumed by a combination reporting and cleanup work, leaving limited room for strategic guidance. In practice, close to two-thirds of their time is spent compiling reports, cleaning data, or reconciling accounts, tasks that are essential, but far from the forward-looking role CFOs enjoy and most businesses expect. Strategic planning and direct client interaction make up a much smaller share of their workload.

“I spend 8–10 hours per client per week just preparing reports.” — Fractional CFO, Services Sector
“I can’t manage more than 15 clients — impossible right now.” — Fractional CFO, SaaS

This misalignment explains much of the frustration we heard. Businesses hire fractional CFOs for strategic clarity, yet they themselves are constrained by manual work. As one interviewee put it, “Half of us are controllers, half are FP&A guys. We call them all CFOs, but they’re not.”

The Key Challenges Fractional CFOs Face

Despite working across different industries, fractional CFOs described remarkably similar challenges. The most consistent theme was the messiness of client data. Engagements often begin with months of unreconciled accounts, duplicate entries, or literal boxes of receipts. As one CFO put it, “Every client starts with a shoebox of receipts.” Another added, “It’s not access to data, it’s the quality that’s suspect.” Before any meaningful analysis can happen, time is consumed by cleanup.

That leads directly to the second burden, reporting. Nearly every fractional CFO said they spend considerable amount of time producing financial statements, board packs, and variance analyses and other types of reports. This workload caps their client capacity and keeps them tied to manual work, even as businesses expect more strategic guidance.

The third challenge is forecasting. While executives want forward-looking insight, most fractional CFOs rely on Excel models for forecasting and scenario planning. While this is great at producing accurate numbers, it is not as good at identifying root causes of variances or recommendations for corrective action. They know the limitations, but see no credible alternative.

Finally, all of this is underpinned by the trust gap around AI. CFOs expressed optimism about automation for categorization and reconciliation but were more cautious about AI’s application in forecasting and strategy. For most CFOs we spoke to, accuracy, and control are non-negotiable. “I don’t trust AI to forecast for me yet,” one CFO told us.

Where AI Fits (and Doesn’t Yet)

When asked about AI, fractional CFOs drew a clear line between the tasks they trust it for today and the ones they are not ready to delegate. On the trusted side, most pointed to routine work such as categorization, transaction coding, and variance analysis. These tasks are repetitive, rules-based, and low-stakes if corrected. As one CFO explained, “AI is great for classification - the stuff my bookkeepers spend hours on.”

CFOs were more cautious about delegating forecasting and strategic guidance to AI. CFOs repeatedly voiced hesitation to rely on AI for forward-looking insight, citing accuracy as a non-negotiable barrier. “I don’t trust AI to forecast for me yet,” one CFO admitted. Another added, “If it just tells me variance, that’s nice but I need to know it’s right.”

This divide underscores the CFO AI Trust Gap. Automation is welcomed when it saves time on bookkeeping and reporting, but adoption stalls when tools claim to replace strategic judgment. Until accuracy is proven, Excel will remain the CFO’s forecasting tool of choice.

Quick Wins vs Strategic Bets

When fractional CFOs talked about where AI could help them today, they consistently pointed to quick wins such as categorization, AR/AP monitoring, and variance analysis. These are tasks that consume hours of manual work but follow predictable rules. “AI is great for classification, the stuff my bookkeepers spend hours on,” one CFO told us. Automating this layer doesn’t just save time, it builds confidence by showing immediate, tangible value.

But the strategic bets are where fractional CFOs’ interest and skepticism both converge. Every CFO acknowledged the need for AI to help with strategic insights, forecasting and decision support. However, accuracy remains a concern that prevents CFOs from adopting these tools widely.

Vendors who succeed will be those that start with credibility at the base layer and earn trust with high stakes tasks.

Implications: The CFO AI Playbook

The conversations revealed both the promise and the limits of AI in finance today. Fractional CFOs are not resistant to technology; they are resistant to tools that promise more than they can deliver. The path forward is not about replacing CFOs but about augmenting them, starting where the pain is most acute and trust can be earned.

For fractional CFOs, three practical steps emerged:

1. Start with cleanup wins. Solve the messy data problem and automate transaction categorization, reconciliations, and variance analysis which frees hours each week and builds confidence in the underlying data.

2. Automate reporting. AI that can generate reports and board-ready narratives, cutting reporting prep time dramatically.

3. Pilot forecasting carefully. Use AI for cash flow forecast scenarios and sensitivity analysis, but always with human oversight. Treat it as co-pilot.

For vendors building AI tools, the implications are just as clear:

• Prove accuracy first. Trust is the barrier. CFOs are not anti-AI, they are pro-trust

• Reduce time-to-value. Provide quick tangible wins that save hours and capture adoption.

• Integrate deeply with QuickBooks. It is the backbone of SMB finance; seamless integration is non-negotiable.

• Earn your way to strategy. Build credibility in cleanup and reporting before claiming forecasting authority.

The playbook is simple; start small, prove value and build trust.

Conclusion

Fractional CFOs are being asked to do more than ever. They step into messy financial systems, spend countless hours on cleanup and reporting, and are still expected to provide forward-looking strategic advice. Their challenge is not a lack of tools, but a lack of time and trust.

AI offers a way forward, but only if it meets fractional CFOs where they are. Automating cleanup, categorization, and reporting can free capacity immediately.

Delivering accurate and explainable outputs builds confidence. Only then can AI credibly move into forecasting and strategic guidance.

The message from fractional CFOs is clear, they are not anti-AI, they are pro-trust and cautious about hype. Tools that deliver immediate, verifiable wins will be adopted. Tools that over promise will be ignored. AI can help CFOs move from being trapped in manual cleanup drudgery to delivering the strategic insight that businesses truly need.

We strongly believe fractional CFOs will continue to be a key partner in unlocking value in main street businesses and are confident AI will be a key tool they use to achieve this.

The worst job I ever had was working as a debt collector at IC Systems, a collection agency in Fargo, North Dakota. Picture a warehouse-sized space about three-quarters the size of a football field, filled with endless rows of cubicles stretching as far as you could see. The moment you entered the building, you were greeted by a cacophony of sounds - ringing phones, clicking keyboards and hundreds of overlapping conversations.

I hated going to work every day, but it paid $10/hr and $15/hr for overtime. That was great money for a college student living on his own plus it funded my sneaker collection. I didn't know anyone who actually loved that job, but it was one of the easiest to get since turnover was so high and they constantly needed bodies to fill seats and answer calls.

So in mid-2024, when I first experienced a demo of Retell AI, one of the first voice AI customer support agents in the market, I was completely blown away. I couldn't believe it wasn't a human on the other end. The voice was scaringly natural.

I immediately thought back to my days at IC Systems with a mix of trauma and nostalgia. I imagined a world where call centers like that could be staffed entirely by AI agents trained on debt collection workflows, so no human would ever have to endure that soul-crushing work again.

Since 2024 when I first came across Retell AI, there has been an explosion of Voice Agents in the market. Every business vertical seem to have a specialized voice agent tool. In fact, Andreesen Horowitz, one of the most prominent investors in AI startups had about 90 companies in their market map for voice agents alone.

So, can AI agents really replace human customer support reps?

AI Claim vs Reality

Customer support has largely remained unchanged for decades. In fact, you could argue it's gotten worse. You still get an operator routing your call to a human or giving you a menu of options that rarely address your actual problem. Around the 2010s, companies started building those cringy chatbots on their websites that most people hated. The industry has been desperate for better solutions.

The prospect of a tool that can autonomously handle customer queries, adjust in real time, and scale infinitely would be a game changer. Imagine a small HVAC service business in San Antonio having a receptionist that works 24/7 and can handle unlimited customer calls and book appointments so the business never loses revenue. That would be incredible.

Think of AI voice agents as having three main parts working together and processing information just as a human will. First, you have the ear that converts your spoken words into text. This is very similar to how a human customer support agent will process information by first listening. For an AI voice agent, this process also happen instantly as you speak.

Next, this information is then relayed to the brain, which for an AI voice agent is akin to the same type of AI that powers ChatGPT. But here's where it gets interesting. The same way a company trains its customer support reps on specific workflows and routing processes, AI voice agents also need to be trained on your specific company's customer support workflows, routing processes, and actual conversation histories. It learns from thousands of real interactions your team has handled and understands which problems get escalated, how billing issues are typically resolved, what steps work best for technical troubleshooting, etc. This is training that gives the AI context to be able to handle real world customer scenarios and adapt to varying edge cases in real time.

Finally, the information processed by the brain gets relayed back to the mouth. With AI voice agents, there is a text to speech technology that converts the AI's text response back into natural-sounding speech. Just as humans, AI systems can even match the tone and emotion appropriate for the situation, sounding concerned when you're frustrated or excited when helping you solve a problem.

All these three parts happen in real-time. These newest generations of voice agents can even hold context throughout the entire conversation, remember what you said five minutes ago, access your account information, and even book appointments in your company's scheduling system all while sounding completely human.

While this sounds very compelling, this week I went down the Reddit rabbit for customer sentiment so I can help you separate the hype from the reality. I reviewed Reddit comments for two popular B2B voice agents, Retell AI and Decagon.

Decagon

Retell AI

The Verdict

Do AI customer service agents actually deliver on their promises? It depends on your use case and the complexity of your current customer support workflows.

Decagon is built for big companies with complex needs. Users like its analytics and smart responses, but it comes with months of setup, requires technical teams to implement, and costs way more than most small businesses can afford.

Retell AI focuses on voice calls and does it really well. Users consistently say people can't tell it's AI because the voice sounds natural and handles interruptions like a human would. I personally tried Retell and i was impressed by it human like voice and ability to adapt even through interruptions. A small business owner I spoke with got 1 meeting for every 5-6 calls they made (same as their human staff) and said it could easily replace a couple part-time employees. The pricing is also straightforward without hidden platform fees.

For small businesses with simple needs, Retell AI is the clear winner. It's easy to set up, pricing is transparent, and it works right away.

Pro Tip

These tools are not perfect and you should expect them to make mistakes. My pro tip for small businesses looking to deploy voice agents in their customer support stack is to follow a safe and sequential roll out process. Start with high control and low agency support task such as appointment scheduling and provide some basic guardrails such as defined times and and days the agent can use. Assess how the agent performs on this task and as you gain more confidence slowly increase the scope to higher agency task. Always have a human to monitor the agent’s interactions and use them as fall back strategy in case things go wrong.

Six months ago, on February 2nd, 2025, I did something I'd been thinking about for months - I published my first newsletter for aiformainstreet. The voices in my head, and maybe my village people (IYKYK), were working overtime to convince me that I'm a bad writer, unqualified to talk about the future of AI, and that no one would read or care about what I have to say. For months, they were winning.

The courage to start came the night of January 31st, 2025, though it still took me two days to hit publish. It wasn't that I was no longer scared or nervous about sharing my opinions to the world, I just decided to move forward despite the fear. I told myself that worst case, this would be a great way to learn much more about AI as I write about it.

The last six months of pouring into this blog has been incredible. I've spoken to many small business owners, written 21 newsletters, and heard from so many of you along the way. I've been amazed by the number of people who have reached out unprompted to say, "Cedric, what you're doing is wonderful, and I love reading your blog." It’s a reminder that fear really is just false illusions appearing real. So let me echo words you’ve likely heard before: if there’s something you’ve been thinking about doing - Go do it.

To all my Main Street champions reading this, thank you for all the support. Thanks to my wife and business partner for reviewing my writing every week. Thanks to Claude and Grok for generating all the images in my blog because God knows I don't have one graphic design bone in my body.

I'm excited to keep writing and sharing because I'm more convinced than ever that the divide between AI-powered and traditional businesses will soon become irreversible. AI-powered businesses are establishing efficiency advantages that will be impossible to catch up to. Main Street businesses are disproportionately at risk of being left behind. The hyperscalers (Amazon, Google, Facebook, Apple, Tesla, etc) have spent over $320 billion in infrastructure to build out their AI capabilities. 64% of the $160+ billion in startup funding this year has gone to AI companies. Always follow the smart money. I don't think the world is fully ready for what's about to hit us with AI's rapid diffusion into every part of our lives, but we'll all have to grapple and eventually adjust to this reality.

At the turn of the millennium during the dot-com boom, many small Main Street businesses were left behind, especially in retail and wholesale. Internet giants capitalized on new technology to move commerce online while small businesses watched from the sidelines.

Not this time. We must empower every small business with the tools and knowledge to not only survive but thrive in the AI age. The best way to build and empower our underserved and overlooked communities is by empowering the small businesses that serve them. When small businesses create wealth, it stays local and drives prosperity within communities.

I look forward to continue contributing towards this ideal both through my aiformainstreet and other ventures I am involved in.

With love,

Cedric

click here to read my very first newsletter

My first few weeks on the job, I basically followed the president around to board meetings, community engagement events, and operational meetings. I thought to myself, "I can't believe I'm getting paid $25 an hour just to learn."

Eventually, I got my first real project to conduct a study on why patient experience and quality of care in the medical-surgical units were lagging behind peer healthcare systems. It took me about three months of patient interviews, shadowing doctors, poring through survey data and care practices, researching what other systems were doing, and analyzing historical data on care outcomes to finalize my report and recommendations.

I ended up getting a full-time offer based on that work. I was lucky my experience was unlike the typical intern experience where you basically serve as the errand person and do little of anything value-added. Reflecting on this experience got me thinking about how AI could potentially upend the traditional apprenticeship pipeline for young interns while also simultaneously giving small businesses access to intelligent interns at 1/100^th of the cost of hiring a full-time intern.

How much more could a business do if the cost of knowledge ultimately approaches $0 and can current tools already approximate the work of some interns do such as basic market research.

AI Claim vs Reality

The core challenge that most small business owners are trying to solve is how to get more market share and wallet share from their customers. Many small business owners conduct competitive analysis on pricing to understand how competitors are pricing their products. Often they'll pay a consultant to handle this type of work. So the promise of a tool that can conduct deep market research at a fraction of the cost of hiring a consultant would be a game changer.

So how does AI work in the background to complete reasoning tasks like this where it needs to think for a while, just like a human would? Think of AI research tools as incredibly fast readers that have already read billions of web pages, reports, studies, and articles. When you ask "What are coffee shops in downtown Portland charging for lattes?" the AI doesn't just search like Google. Instead, it breaks down your question and starts making connections between different pieces of information it has seen before.

The AI looks for patterns across thousands of coffee shop websites, menu listings, review sites, and local business directories. It identifies which businesses are actually competitors, extracts their pricing information, and starts noticing trends—like whether shops near universities charge less or if organic coffee costs more across the board.

What's different from a simple search is that the AI can reason through the information. It might notice that three shops increased prices in the last six months and connect that to a news article about rising coffee bean costs. Or it might identify that shops with similar atmospheres tend to price similarly, giving you insight into positioning strategy.

The AI then packages all this analysis into a comprehensive report with the specific insights you need - competitor prices, market trends, and recommendations - completed in minutes instead of the weeks or months it would take a paid consultant or intern.

This week I put this claim to the test by using three AI tools (ChatGPT, Claude and Manus) to create a competitive pricing report for Playa Bowls – a healthy superfruit acai bowl, smoothie and fresh juice fast casual restaurant in Philadelphia.

In order to create this report, i constructed a prompt similar to what a business owner might do when providing requirements to an intern or paid consultant to run this analysis. I used the same prompt across all three AI tools

Prompt

Research competitive pricing for Playa Bowls (açaí bowls, smoothies, fresh fruit) in Philadelphia market to assess pricing strategy and identify optimization opportunities. Competitors to Research: Vitality Bowls, Jamba Juice, SoBol, Rush Bowls, local açaí shops: Sweetgreen, Freshii, CoreLife Eatery, local juice bars Adjacent: Tropical Smoothie Cafe, Smoothie King, Starbucks. Collect pricing for: Açaí bowls (all sizes, signature vs build-your-own, add-ons) Smoothies (all sizes, protein vs fruit-based, supplements) Other items (oatmeal bowls, toast, fresh fruit cups). For geography focus on Center City, University City, Northern Liberties, King of Prussia, Main Line suburbs. Your key Deliverables are: Comprehensive report with: Pricing comparison tables by product category Market positioning analysis (where Playa Bowls ranks) Geographic pricing variations Value proposition comparison (portion sizes, ingredients, quality) Strategic recommendations for pricing optimization.

Manus

Competitive pricing report from Manus AI

Playa Bowls Pricing Report

320KB ∙ PDF file

Download

Download

ChatGPT

Competitive pricing report from ChatGPT

Competitive Pricing Analysis For Playa Bowls In Philadelphia

1.36MB ∙ PDF file

Download

Download

Anthropic Claude

Competitive pricing report from Claude

Philadelphia Açaí Bowl Market Pricing Analysis

99.7KB ∙ PDF file

Download

Download

The Verdict

So, is the claim that AI can help you complete market research using a simple English prompt accurate? Yes! I was impressed across the board by Manus, Claude, and ChatGPT's ability to reason and create market research reports.

Stylistically, there were few differences between the three tools. Manus' output looked the most like a structured report, even though the formatting (tables, line breaks, etc.) wasn't the best. But it was the only tool that clearly outlined its methodology and provided succinct summaries that made the information very digestible. It intuitively knew to not just look at different sources for pricing but also create an average across multiple channels, which showed that Playa Bowls was priced more competitively.

While ChatGPT's report was beautifully designed, the information wasn't as digestible for me. ChatGPT also didn't disclose its methodology, which made it hard to understand the basis for ranking Playa Bowls as a more premium offering. I did love that ChatGPT provided verifiable sources, though many of their pricing sources were from delivery apps like Uber. Claude's output was very similar to ChatGPT.

I personally preferred Manus' output because it was digestible and the methodology was transparent. One area for improvement is their report design template. Manus also transparently shows you in real time the pages they're visiting and provides a full transcript of every step they take. I didn't find ChatGPT and Claude to be as transparent.

Thirteen years ago, I was getting paid $25 an hour as an intern to conduct deep research for a hospital. Today, I paid the equivalent of $2 to build three competitive pricing research reports in a matter of minutes.

Pro Tip

When using AI tools to conduct market research, always ask the AI to disclose its methodology and sources when it is unclear and always run a sanity check on those sources to ensure the output and recommendations you are getting are grounded in truth.

When I moved to the US, I was astounded to find multiple computer labs at school. I'd never seen that many computers in one place. For the first time, I could have a computer all to myself. It was fascinating, and I wanted to take any class where I could use one. At first, it was awkward, I was self-conscious about my typing speed, which must have been 10-15 words per minute at the time (I know that’s atrocious please don’t judge me 🤭). I was fascinated by classmates who could type full papers in 30 minutes without looking at the keyboard or while talking to someone else.

Then I took a web development class. Suddenly, my keystroke speed didn't matter because we were building cool stuff. That was when I first learned CSS and HTML using a software called Dreamweaver. I could write a few lines of code and watch it render into an actual webpage in real time. It felt like magic, and I didn't want to stop.

Things have evolved rapidly since 2005 when I was writing HTML in high school using Adobe Dreamweaver. From drag-and-drop editors like Wix to new AI-native tools like Lovable, web development is now accessible to anyone who can speak and write English.

But with the new advancements in AI can any small business owner really create a website for their business just using natural language?

AI Claim vs Reality

One of the first things almost every small business thinks about when they first start out is creating a website. Without a digital footprint, customers can’t fine you and a website is still the most important first step to take to establish your digital presence. Many main street business owners still hire the services of a contractor on platforms like Fiverr and others to build a website. So, the promise of a tool that can allow business owners to create high quality professional websites in minutes would be a game changer.

Think of an AI website builder as a virtual website developer that's been trained on millions of real websites and the underlying code for each. When you type "I need a one page website for my flower shop business with a neutral theme and a contact form," the system breaks down your request into pieces and uses what it's learned about page layouts, form fields, and styling to predict the exact code you need, one line at a time.

It's similar to how the AI video and image tools we've discussed work, but instead of predicting pixels or frames, it's predicting code. The AI stitches together headers, image galleries, text blocks, and buttons for your website based on patterns it's seen across millions of websites. It even handles the behind-the-scenes stuff like ensuring form submissions actually get emailed to you.

Once everything is drafted, the tool packages it into a ready to go website which you can preview live and push to a hosting service like GoDaddy and others. You can also make changes using natural language and the AI will update your website instantly. What you end up with is a fully functional website built in minutes, without ever writing a line of code.

This week I put this claim to the test by creating a one page website for a nail salon in Conshohocken, Pennsylvania using three of the most popular AI website builders; Lovable, Bolt and v0.

I used the same prompt shown below for each tool and have provided a full review of the results in the videos that follow (click to open video).

Prompt: build a one page website for my nail salon shop called *Destination Nails* located in Conshohocken, PA. i want to have a contact form and also an appointment scheduler where customers can set appointments with one of my three mani and pedi technicians named Haley, Hazel and Harriet respectively.

Lovable

Bolt

v0

The Verdict

So, is the claim that AI can generate a professional website using a simple english prompt accurate? Yes! I was very impressed by these tools' ability to create a functional one-page website. There were some minor design issues with font colors and images, but these could largely be resolved by asking the AI to edit the output.

I personally preferred Lovable for creating my website. While Bolt's output was equally impressive, I loved the user experience and overall feel of Lovable. It seemed more flexible. I could move components around and edit things directly if I didn't want to use the AI. The interface felt more intuitive to navigate.

Twenty years ago, when I started learning HTML with Dreamweaver, I could never have imagined a time when anyone could build a website by simply typing an English prompt. Using Lovable brought back that same childlike excitement I felt when I first walked into the computer lab at James Hubert Blake High School in Silver Spring, Maryland, and wrote my first line of code.

I'm thrilled that now, with AI, a small nail salon in Conshohocken, PA can create a website and get their business online in less than 15 minutes. This creates a new level of agency that unlocks boundless opportunities for entrepreneurs across the globe.

Pro Tip

When editing your website in Lovable, select the specific section you want to change, then describe what you want in natural language. This gives the AI better context and produces more accurate results.

Though I'm not a betting guy, I'd take the over on AI-generated films winning Oscars in the coming years. But as someone who spends considerable time thinking about how small businesses can harness AI effectively, this Kalshi moment sparked a different question for me: if AI is good enough to create content for national television, can it help small business owners produce professional quality video ads for their social media campaigns?

AI Claim vs Reality

As I discussed last week, many small businesses rely on social media as their primary distribution channel, spending countless hours and sometimes thousands of dollars creating advertising content. A tool that could help business owners produce more content faster and cheaper would be a complete game changer.

AI video generation works similarly to image models we discussed last week, but instead of creating a single picture, it's making smart guesses about how things should move and change over time. When you type "create a video of a dog running through a park," the AI doesn't actually understand what dogs, running, or parks are. Instead, it's been trained on millions of video clips paired with text descriptions, learning what these things typically look like in motion.

Think of it this way: imagine you're a video editor and I showed you hundreds of thousands of cake decorating videos while describing each technique out loud. Eventually, you'd start recognizing patterns such as how frosting flows, how decorations are placed and how finished cakes should look. When a client asks you to create content for their custom gourmet treat company, you'd have context to create the best approximation of what it should look like in motion. That's exactly what AI does, but at a massive scale across millions of video clips.

My goal this week was to explore whether AI is already good enough to help small business owners create video content for social media. Take Kabsy, the owner of a Houston based custom gourmet company called KabsyKreations. Kabsy spends considerable amount of time creating content for her social media pages. The clip below was produced by a real video editor based on real footage provided by Kabsy of of her cake preparation and assembly.

My goal was to try to recreate this video using Sora or Veo 2, two very popular AI video generation tools created by OpenAI and Google respectively.

Original clip created by human video editor

1st Shot

For my first shot, I created a prompt to replicate the scene in the original clip above. I started by describing my best interpretation of each scene in the video. Based on what I saw, I identified a three-stage process: preparation and batter mixing, cake assembly, and decoration. I described each phase in my own words, then passed my description to ChatGPT and asked it to act as a film director and video editor to convert my description into a script for high-fidelity video generation.

After iterating on the script a few times, I broke it down into three prompts representing each stage of cake preparation (see prompt below). I then passed each prompt to Sora and Veo 2 respectively to generate video clips and used CapCut to stitch the clips together.

Prompt Generated by ChatGPT

Scene 1: Begin with a cinematic close-up of real butter stacked in a yellow bowl on a marble kitchen countertop. The lighting is warm, with soft shadows and natural tones. At 3 seconds, cut to a side view of a Black baker operating a stand mixer. The mixer blends butter and flour while the paddle spins and flour dust gently rises. Emphasize motion, texture, and a cozy kitchen atmosphere with shallow depth of field.

Scene 2: At 7 seconds, cut to a top-down close-up of silky white batter swirling inside a metal bowl. The batter is glossy and smooth. At 11 seconds, transition to a Black baker’s hand folding vibrant red velvet batter using a red silicone spatula in a stainless steel bowl. Focus on the spatula’s slow swirl and the deep, rich red texture. Highlight contrast between batter and bowl using natural daylight from the side.

Scene 3: Start with a wide-angle shot of multiple cake layers wrapped in plastic and stacked on a metal tray. Slowly pan across. At 16s, show the Black baker’s hands placing pastel-colored cake tiers on a spinning turntable. One hand adjusts the top tier while the other spreads pink frosting with an offset spatula. At 18s, cut to a slow push-in on the finished cake: pastel frosting, candy sprinkles, a waffle cone topper, and colorful lollipops. Emphasize handcrafted detail and visual delight.

Sora clip output

Veo 2 clip output

Though both clips looked different from the original KabsyKreations video, I was particularly impressed by the Veo 2 output. Its ability to understand context, camera angles, and produce hyper-realistic video that follows the laws of physics was promising. Veo also naturally included relevant sound even though I didn't prompt it to do so. The Sora output was decent, but some movements looked awkward and unnatural, making it far easier to tell the clip was AI-generated.

2nd Shot

For my second shot, I tried something different. I kept the same prompts but extracted still frames from the original KabsyKreations clip (see below) and attached them as additional context to my Sora and Veo prompts. I was very specific about which frames to reference for each scene. The output from both tools improved significantly. Sora's result was much closer to the original video, though some object movements still looked perceptibly unnatural. Veo 2 consistently created superior quality that was nearly indistinguishable from human-generated clips.

Sstill frames from original KabsyKreation clip

Sora clip output

Veo 2 clip output

The Verdict

So is the claim that AI can generate professional video clips that can be used by small business owners like Kabsy in social media ads accurate? Yes! However, you'll need patience and willingness to experiment extensively with prompting to find great clips. You'll also need post-production work to stitch clips together for a coherent video.

Veo 2 from Google Gemini was clearly superior, though Sora can create respectable quality with enough prompt iteration. This process took me about 3-5 hours using three paid tools: Sora, Veo 2, and CapCut.

I learned three key lessons for optimizing your output. First, break it down, don't one-shot your video production. Long prompts often produce worse results, and some models like Sora truncate them. Break your video into distinct 5-10 second scenes and generate each separately. Second, be precise. For a 10-second clip, clearly specify what happens first, then next, breaking it down by timeframe to guide the AI (see sample below). Third, provide context. I included still frames from Kabsy's actual kitchen preparation, and the output improved dramatically. The additional context makes a material difference in quality.

Sample prompt
Scene 1: Begin with a cinematic close-up of real butter stacked in a yellow bowl on a marble kitchen countertop. The lighting is warm, with soft shadows and natural tones. At 3 seconds, cut to a side view of a Black baker operating a stand mixer. The mixer blends butter and flour while the paddle spins and flour dust gently rises. Emphasize motion, texture, and a cozy kitchen atmosphere with shallow depth of field.

So, for small business owners, my recommendation will be to use Sora and CapCut to generate videos and ChatGPT to iterate on your prompts. Though you will have to invest time upfront once you get good at this and even create some prompt libraries you can reuse, this can save you significant time and money and allow you to churn content with much higher velocity.

Pro Tip

When generating videos with AI, think about your storyboard first and describe each part in plain English. Then pass this storyboard to ChatGPT and ask it to create prompts for video generation. A sample prompt you could use is: "Act as a film director and video producer and convert this storyboard into cinematic prompts to generate a video using popular models like Sora and Veo 2. Please keep the length of description for each prompt to clips of 5-10 seconds each."

Once you get your first iteration of prompts, run them through Veo 2, assess the output, and go back to ChatGPT to iterate on the parts you didn't like or want to improve. Repeat this process until you get your desired output.

My older brother had a more artistic way of immortalizing Goku. He was a talented pencil artist who could create realistic drawings of a Saiyan. Like many younger brothers, I looked up to my elder brother and wanted to be like him but had zero artistic ability - That gene skipped right over me. I couldn't even draw a perfect circle to save my life.

So when ChatGPT recently released its advanced image generation model, while everyone else was busy creating Studio Ghibli art, - a unique visual style developed by the renowned namesake Japanese animation studio - my first instinct was to finally create that Saiyan Gohan drawing I'd always wish I could. I was completely blown away. For a moment, the young, inartistic kid in me believed I was Da Vinci.

Vote and then read to the end to find out which of these drawings was created by ChatGPT

I wasn’t alone, the ChatGPT image generation model release created such a viral moment that OpenAI CEO Sam Altman implored folks to “chill on generating images because his servers were melting”

Beyond Studio Ghibli art and my nostalgic attempt to relive my childhood, the real question is: can AI help small business owners create professional ad graphics and logos that can be used commercially? The short answer is not quite yet except you know exactly what you want and have some technical background of design elements.

AI Claim vs Reality

Most small businesses today use social media as their primary distribution channel and spend hundreds, sometimes thousands of dollars monthly with agencies for logos and graphic designs. So the promise of a tool that can create professionally designed logos and ad assets for a fraction of the cost is certainly exciting.

Two weeks ago I wrote about smart guesses and how LLMs generate text. AI image generation works differently than text models, but the core concept is the same; AI is still making smart guesses. When you type "create an image of a kid playing in the mud and eating ice cream," the AI doesn't actually understand what kids, ice cream, or mud are. Instead, it's been trained on billions of images paired with text descriptions, learning what these things typically look like.

Another way to think of this is; imagine you're a designer and I showed you hundreds of thousands of event planning agency logos while describing each one out loud. Eventually, you would start recognizing patterns. So, when a client asks you to design a logo for their new event planning business, you'd have context to create the best approximation of what it should look like. That's exactly what AI does, but at a massive scale.

But can it create a visual image that actually works for your specific brand? That's what I set out to test this week.

My goal was to recreate the logo of Rima, which was designed by a human using three AI tools - ChatGPT, Canva AI and Leonardo AI.

Original Rima Logo

1st shot

For my first test, I gave the three AI tools above a prompt that represented my best description of the logo, as I would explain it to a human designer (see the prompt below). The output was poor overall, with Leonardo AI performing the worst. This wasn't surprising given how image models work. The AI tries to map the content of my prompt to the closest approximation it can find based on patterns from all the images and descriptions it was trained on, then creates what it thinks is the best fit based on its training data.

2nd shot

To get closer to what I wanted, I decided to give more explicit context by sharing the exact image of my logo and asking ChatGPT to modify its previous output to match it exactly (see the follow-up prompt below). The other AI tools - Canva and Leonardo AI - don't support image to image generation, so I could only test this approach with ChatGPT. This improved performance significantly and created a close approximation of my original logo, though the color shade and gradient density were still slightly off from the original.

3rd shot

For my third shot, I tried something different. I uploaded my original logo to Claude and asked it to create the exact prompt that could replicate the image on other platforms like ChatGPT, Canva AI, and Leonardo. Then I used that exact prompt across all three tools. It was interesting to see how specific and technical Claude's prompt instructions were. I was also surprised by the huge variation in output across the three tools. ChatGPT's output was very impressive, though I noticed that using the same prompt three different times gave me three different results, similar to each other but with slight deviations in color and shadow. Leonardo and Canva AI performed poorly with one-shot image generation prompts across board.

Claude prompt: Create a minimalist SVG-style image of a large, centered circular shape on a light background. The circle should have a soft radial gradient with these properties: Center color: #9B59B6 (a medium lavender-purple) Outer color: fades smoothly to white or fully transparent at the edges Gradient type: radial, smooth and subtle Background: pure white or very light gray (#F5F5F5) Design style: modern, clean, and ethereal — no text, borders, icons, or embellishments Effect: the circle should appear like a softly glowing, diffused light source Format the image as a square or circle-centered composition suitable for logo use.

The Verdict

So is the claim that AI can design a professional logo for you that can be used commercially accurate? Not quite yet. It depends heavily on the tool you use and your ability to be very specific about what you want.

Today, I believe ChatGPT's image generation works best as a brainstorming partner for experienced designers who can write detailed, technical prompts and iterate quickly on ideas. For small business owners with limited design experience, AI can spark some ideas, but you'll still need to work with an experienced designer to create something truly professional and commercial-ready.

ChatGPT was clearly the best performer among the tools I tested, producing significantly better results than Canva AI or Leonardo. However, even with the best prompts, the outputs from chatGPT still had inconsistencies in color, shadow, and overall polish that would matter for professional use.

So, for small business owners, my recommendation will be to use AI for initial concept exploration, but partner with a designer for final logo. The technology is advancing rapidly, so this gap will likely close soon but we're not quite there yet.

Pro Tip

Use AI to enhance your prompting skills. If you find a picture, design, or color scheme online that you love, upload it to your favorite LLM chatbot and ask it to create a prompt that other AI tools can use to recreate the design. Then use that prompt in ChatGPT or other state of the art image generation AI tools to recreate it, and use that initial design as a canvas for different iterations with follow-up prompts.

The answer to the poll question is A. Did you get it right? 😊

I’ve known my wife for 14 years, and we’ve been married for 8. She has this frightening ability to predict exactly what I want to talk about based on my mood, complete my thoughts mid word, and perhaps most discomforting call me out on a lie with striking accuracy. Sometimes it's actually scary how accurate she is. I feel so exposed at times and I’m still working on beating her psychic abilities - so far, no luck. If you’re reading this, baby I love you.

Here's the thing though, my wife’s psychic abilities is a product of being trained on 14 years of data about me. She knows my speech patterns, what gets me excited, what I ramble about when I'm stressed, and exactly which topics I'll inevitably circle back to in any conversation. So when she makes these predictions about what I'll say or do next, her accuracy is remarkably high. Her training set is hyper-focused on one very specific subject - me.

This is exactly the promise many AI companies are now making; that AI can understand you and your business just as deeply and write marketing copy, emails, and posts in your unique voice, tone, and style, automatically. It's a compelling promise, especially for small business owners who spend hours crafting the perfect message to their audience.

So can AI guess what you want to say and say it in your voice, style and tone without coming off cringe to your target audience? The short answer is, it depends.

AI Claim vs Reality

Under the hood, Large Language Models (LLM) - the technology behind tools like ChatGPT don't think the way humans do - yet. They are prediction engines. When you interact with tools like ChatGPT, Claude, or Google Gemini, it takes your input and makes an educated guess one word at a time about what’s most likely to come next. Next time you hear someone throw around jargon like token-prediction to sound smart, this is exactly what they're talking about.

These AI models have been trained on billions of words from the best ads ever written, thousands of blog posts, and examples of how top brands communicate with customers. So when you ask for marketing copy, the AI makes a smart guess from this massive dataset to generate text that sounds good for most people in most situations.

But can it guess exactly what I want to write to my audience in my voice and tone? I decided to test this claim with three of the most popular AI tools used by billions of people globally.

I took a LinkedIn post (shared below) I'd handwritten to announce my new project Rima which I also shared in my newsletter last week—and tested whether ChatGPT, Claude, or Google Gemini could reproduce something close to it. Here's my process and what actually happened.

Original hand written LinkedIn post:

This week, I’m pausing my scheduled blog post to share a project close to my heart. Since boarding a plane to Ghana in 2017 to build my first startup, I’ve been driven by a singular mission: to create wealth in local communities. Over time, I’ve come to believe the most powerful way to do that is by helping the small businesses that power them.

I’ve been tested, humbled, and shaped by my journey so far - and now, I’m channeling those lessons into a bold new goal.

My partner in crime Gwanygha'a Gana and I just co-authored a post on why we’re building Rima. It captures the heart of our mission and the scale of the opportunity in front of us. I encourage you to read it.

Success is far from guaranteed. In fact the odds of failure are high. But I’ve never been one to follow conventional wisdom. Some calls you can and should ignore. Others feel too important not to answer. This is one of those.

1st shot:

For my first test, I gave the AI a one-shot prompt to help me write a LinkedIn post. I provided some basic guidance, and while none of the results fully captured my voice, tone, or style, I was still impressed. The output was much better than what I’ve seen from earlier versions of these models. Claude’s response stood out the most, it felt the warmest and most human.

2nd shot:

To get the AI’s output closer to my actual voice, tone, and writing style, I decided to give the models a sample of my work. I shared a link to a blog post I had written about why I’m launching the project and asked the AI to edit the LinkedIn post to match that voice. The result from ChatGPT came much closer to how I actually write and speak while the output from Gemini and Claude got a little worse.

3rd shot:

To improve the chances of getting an output closer to what I wanted, I refined my prompt by giving the AI more detailed guidance, including specific examples of what I wanted the post to cover. I also asked it to take on a specific persona to match my writing style. While Google Gemini delivered the weakest result, Claude’s response came the closest to my original post. It felt like something I’d actually write and share on LinkedIn

The Verdict

So is the claim that AI can write expert marketing copy for your business correct? Yes, but with an asterisks. The quality and usefulness of the output depends entirely on how well you guide it. In my testing, Claude performed best, delivering content that actually felt like something I'd write and share.

Remember that beyond the hype, LLMs are just prediction engines making smart guesses. So to help them make better guesses personalized to your business, you need to provide better guidance.

Pro Tip

The key to getting better outputs is prompting. There are four key building blocks to creating a great prompt that improves the accuracy of the guesses AI makes to match your desired output.

• Specify the exact task

• Provide context about your business/audience

• Give sample examples of your writing style

• Specify the persona you want the AI to adopt

In my first attempt above, I just gave the AI a task (create a LinkedIn post) and got generic results. My third attempt provided context with a sample blog post to mirror my voice and style, and specific examples of what I wanted. The output was dramatically better.

So yes, AI can definitely write marketing copy for you. But you'll need to do the work upfront to guide it properly. The smart guess is only as good as the information you provide.

I’ve been tested, humbled, and shaped by my journey so far - and now, I’m channeling those lessons into a bold new goal: helping small businesses around the world unlock $10 billion in profits over the next decade with the help of AI.

My friend and co-founder, Gana, just shared a post on why we’re building Rima. It captures the heart of our mission and the scale of the opportunity in front of us. I encourage you to read it.

Success is far from guaranteed. The odds of failure are high. But I’ve never been one to follow conventional wisdom. Some calls you can and should ignore. Others feel too important not to answer. This is one of those.

I met Ced through my childhood friend Kevin at a conference in Durham. Kevin had met him in Minneapolis and said, “There’s this dude into startups - I think you two would click.” He was right. From our first conversation, it was clear we needed to build something together. Cedric was as driven and obsessed with building as I was, and couldn’t care less about traditional career paths.

The challenge was finding an idea we were both excited to pursue. After countless back-and-forths, we landed on a pain point I faced regularly, shipping spare parts from the U.S. to Africa for my dad, who was constantly getting ripped off by local mechanics. With this idea, we launched Garage - a B2B marketplace for auto parts traders - got into Y Combinator, raised a few million dollars, and built the largest network of neighborhood auto parts retailers across Africa.

This journey transformed us. We developed a deep appreciation for how difficult it is for small businesses to start, run, and scale and we learned that the only way to create enduring value is to build sustainable profits.

At scale, Garage operated across three countries - the U.S., Ghana, and Nigeria. Just four months into starting the company, Ghana was hit with its worst macroeconomic crisis in history with skyrocketing inflation and the Cedi tanking into the worst-performing currency in the world. A year later, Nigeria followed. Its currency depreciated over 300%, with inflation gutting buying power. We found ourselves operating in an environment where fuel prices quadrupled, distributors raised prices 40%+ each quarter and customer demand dried up. Our unit economics crumbled, and we were facing a life or death scenario.

Faced with this existential crisis, we had to innovate and reinvent ourselves. So, we leaned into our strengths in data analytics, finance and automation. We set three goals to achieve within six months: cut direct costs by 60%, automate 90% of sales, and grow gross margins to drive profitability.

Our team of data scientists, engineers, and finance managers worked around the clock. We built a machine learning model to optimize fuel allocation and delivery routes, reducing fuel costs - 60% of our direct costs - by more than 50%. We built another model to guide what to stock, when to buy, how much to hold, and how to price. That narrowed our inventory to high-churn, high-margin SKUs and boosted gross margins by five percentage points in six months. Our engineering team also built an AI-powered WhatsApp agent that took orders from retailers, allowing us to automate 90% of sales and retain only our top performers. By the end of this transformation, our largest city in Ghana was profitable and Nigeria was near breakeven.

Through this experience, we saw a bigger opportunity: What if every small business could do what we did without a full team of analysts, engineers and finance operators? What if a business owner could work with an AI agent to define a goal like growing margins by 5% and receive a fully executable plan, powered by both humans and AI, tailored to their business? In late 2024, this idea sounded like science fiction. Today, we are confident it’s within reach. That’s why we decided to pivot from Garage and build Rima.

What we are building at Rima

Jeff Bezos famously said you should build businesses around desires that won’t change not those that change. Business owners will always want higher profits, and investors will always demand better returns. It’s hard to imagine a founder saying, “I wish we earned less this year,” or an investor hoping for lower returns. That core desire for growth, margin, and efficiency is timeless.

AI is destined to help business owners optimize profits and unlock a new era of abundance by narrowing the gap between idea and execution. Tools and capabilities once reserved for large enterprises will soon be accessible to small businesses at a fraction of the cost.

Our purpose is to create this future.

We’re starting by building a full-service, AI-powered CFO advisory firm that combines the best AI tools with the top 1% of finance talent to help VC/PE-backed small businesses, like Garage, uncover insights that drive profits.

Next, we’ll scale this service to thousands of businesses, delivering strategic finance and operational clarity to thousands of businesses and their leaders.

Then, we’ll build an autonomous, self-improving AI co-worker that can deliver enterprise-grade financial insights and execution at a radically lower cost, unlocking profitability for even more small businesses.

We’ll do this until Rima helps generate $10 billion in profits for Main Street businesses around the world over the next decade.

When we started Garage, our goal wasn’t just to fix the auto parts supply chain, it was to help neighborhood traders in Ghana and Nigeria grow their profits and build enduring businesses. That mission hasn’t changed. With Rima, we’re scaling that vision globally, unlocking $10 billion in profits for Main Street businesses over the next decade.

If you are as obsessed as we are about creating wealth in communities by empowering the small businesses that serve them, come join us!

Over the last few weeks, I've been talking to fractional CFOs across the country. I run Rima, an applied AI company focused on helping small businesses increase their equity value, which ultimately builds wealth in the communities they serve. My goal was to learn from the professionals who handle financial stewardship for these businesses, to understand why so many struggle with cash flow and where the real opportunities lie.

One question I've been asking everyone is; what AI tools do you use today in your work, and how valuable do you think AI is in that regard? The responses shocked me across every age group. Almost everyone confirmed they've heard about AI, but they still don't understand what AI actually is or what it can do for them. Many of these fractional CFOs told me they trust Excel more than AI and wonder whether AI can really be trusted with forward looking financial analysis and decisions.

This is fascinating because while many technologists believe AI is already better than any single accountant at certain tasks, the people on the front lines of Main Street are saying "not so fast." We forget it's only been two years since this technology crashed onto the scene, yet we expect every boardroom, Main Street business owner, school district, and hospital to fully understand it and start deploying it immediately.

That disconnect I'm seeing isn't just about one conversation; it reveals a messy tension between AI's grand promises and operational reality. Things are moving incredibly fast, with everyone scrambling to build the next shiny AI tool that will supposedly transform business forever. The only problem is that many businesses aren't buying the hype - yet.

It's not that folks are opposed to AI, they just don't know what it can actually do. To most people, AI still looks like nothing more than a sophisticated chat bot with a search engine that can spit out information on demand. They haven't seen evidence that it's anything beyond that.

The people I talk to are wrestling with three basic questions: I don't understand what AI is, what it can actually do for me, and if I can trust it with important work? Part of this is simply a gap in awareness. One fractional CFO I spoke with was amazed when he recently found out that ChatGPT could take an Excel file, understand what was in it, run analysis, and answer basic questions about the data. Once he discovered that one use case, he never looked back.

This series isn't for AI optimists or skeptics - it's for pragmatists. Those on the front lines building the future on mainstreet who are curious about AI but genuinely don't understand it or know what it can really do. People who want to separate fact from fiction and make sense of the gap between AI claims and what actually works in the real world.

Every week, we'll take one specific AI claim or capability and have a grounded, educational discussion about its merits and practical value.

Bill Gates recently announced he plans to close his foundation within the next two decades. Not because it failed—quite the opposite. He's designing himself out because he believes the foundation's work should ultimately continue without him.

This moment struck me deeply. At its core, designing yourself out is the highest form of leadership. It's creating something so well-structured, so resilient, that it transcends your personal involvement. I've come to believe that capitalism at its best isn't about accumulation and extraction but about empowerment—building enterprises that generate abundance and improve communities long after we're gone.

For business owners running Mainstreet businesses, this carries a profound responsibility. They’re not just building livelihoods for themselves; they’re creating economic engines for entire communities. If you are such a leader, the true measure of your success isn't how essential you remain, but how effectively you can step away while your dream and life’s work continues to thrive.

In our previous articles, we explored gaining visibility across your business data and extracting actionable insights from that information. But information alone doesn't create freedom. The final, most challenging step is building systems to act on those insights even when you're not there. That's what it means to design yourself out.

The Greatest Act of Leadership is Letting Go

When small businesses succeed, wealth stays local, creating a virtuous cycle of prosperity. This is why empowering Main Street enterprises might be the most effective path to building thriving communities across America and the world.

For these businesses to create lasting impact, they must grow beyond their founders. The dream of every business owner is to have a company that can grow without them and eventually continue to thrive long after they're gone.

The harsh reality though is that 82% of businesses that fail cite cash flow problems. Cash flow isn't just one isolated issue—it's like a flesh-eating bacteria that affects every part of the organization while simultaneously being impacted by every decision you make.

What compounds this problem is how many Main Street businesses remain dependent on the founder's intuition and personal decision-making. Without robust systems and processes, these businesses struggle to scale beyond what one person can manage; regardless of how talented that person might be.

This is where modern AI tools change the game entirely. As we explored in our previous posts on gaining visibility across your data and extracting actionable insights, small businesses now have access to capabilities that were once exclusive to large corporations with massive budgets. AI enables small businesses to combine the best of human judgment with powerful automation, creating systems that can operate beyond the owner. The best way to create an enduring enterprise isn't by making yourself essential—it's by designing yourself out of it.

Give Away Your Legos

I first heard about "giving away your legos" from my co-founder, . The concept is simple: to grow, leaders must continually give away control of things they've built and mastered.

At Garage Mobility, I led finance and operations, with inventory management as a critical function. For our first year, I personally managed all procurement—deciding which items to buy, quantities, suppliers, and handling all approvals. This worked initially, but as we scaled, cracks appeared. My demand forecasting was inconsistent, leading to both stockouts and dead inventory that strangled our working capital and cash flow.

I realized I needed to give away this particular lego and build a process that didn't revolve around me. This is where the principles from our previous articles came into play. We built machine learning models to analyze demand patterns and cash cycles across our inventory, creating accurate predictions of what we needed in stock and when.

The transformation was remarkable. Every morning, we generated a forecast sheet showing exactly what to buy, how much, and from which suppliers. Sales team members could annotate it with real-time customer requests, and procurement associates executed purchases within model-stipulated limits. I only needed to step in for exceptions.

By designing myself out of this process, we not only eliminated a bottleneck but actually improved performance beyond what was possible with my direct involvement.

Take The First Step

Designing yourself out doesn't happen by accident—it requires a deliberate framework. Start with tasks that drain your energy but don't leverage your unique expertise. Next, focus on processes with clear rules that could be documented and systematized. Finally, target decisions that could be automated with the right data inputs. The best part? You don't need a machine learning team or technical expertise to achieve any of this.

For a comprehensive roadmap to implement this approach in your business, I've created a step-by-step guide: AI Implementation Guide for Small Businesses. This e-book walks you through the entire process from identifying strategic priorities to setting clear implementation goals using proven frameworks for prioritization and execution.

The journey from overwhelmed owner to superhuman business leader isn't about working harder; it's about building systems that amplify your impact while reducing your direct involvement. From identifying what's happening in your business, to understanding why it's happening, and now implementing systems that act without you, this series has mapped the complete journey from owner-dependent to truly scalable business. By gaining visibility across your data, extracting actionable stories from that information, and designing processes that operate independently, you create something truly remarkable: a business that generates value beyond your personal capacity. That's not just freedom for you; it's the foundation of lasting prosperity for the communities you serve.

THIS WEEK ON MAINSTREET

Something is cookin’! We’re redesigning this section to better serve you

My wife's favorite crime drama is "Bones," and she particularly loves the lead character, Dr. Temperance Brennan. Like many fans, she admires Dr. Brennan not just for her characteristic blunt comic honesty, but for her brilliant ability to piece together evidence to uncover the story behind a crime. I've watched enough episodes to see how Brennan examines a fractured skull fragment, identifies unique tool marks, and connects it to a specific gardening tool — suddenly transforming scattered bones into a coherent narrative about the victim's final moments.

What business owners can learn from Dr. Brennan are the core principles of translating data into a great story that uncovers the right insights to drive action. Those principles are asking the right questions, having the full context (or close to it), and using the right techniques to analyze what's in front of you.

However, sometimes running a small business or walking into a newly acquired operation feels exactly like stepping onto a crime scene where evidence has been scattered everywhere, with no one around to help you make sense of it. You know something isn't right with the business, but you're overwhelmed just trying to figure out where to start looking for answers.

Stay Curious My Friends

"The only thing more dangerous than ignorance is arrogance" – Albert Einstein

This couldn't be truer when it comes to understanding your business's story through data. In my experience, business owners tend to approach their data in three distinct ways. The most successful are the genuinely curious—those who constantly probe beneath surface numbers, challenging assumptions and seeking deeper understanding. Then there are those who operate with incomplete information—not through any fault of character, but because they don't know which questions to ask or assume everything is fine until it's too late. Finally, there are those who rely too heavily on gut feeling—confidently dismissing warning signals and remaining resistant to what the numbers are telling them. While confidence is valuable, the second and third approaches can leave critical business insights undiscovered.

I learned this lesson firsthand during a period of hyperinflation in Africa running Garage Mobility – a B2B autopart marketplace. When facing steep quarterly price hikes from suppliers and performance pressure across all our cities, we had to make the difficult decision to drastically narrow our product catalog from dozens of brands to just half a dozen. To make these decisions intelligently, I couldn't just go with my gut or what had worked before. Instead, I stayed ruthlessly curious, asking pointed questions about our data: Which brands delivered consistent margins across regions? What products showed the least price elasticity, giving us room to increase prices? Which SKUs had the most vendors carrying them, reducing supply risk? Were locally made products outperforming imported ones during supply chain disruptions? How did seasonal patterns affect each product's profitability over time?

Today, AI can be a game-changer even if you're not naturally inclined toward this kind of curiosity. Instead of staring blankly at your P&L and heaps of operational data in excel to understand where to start, you could prompt your favorite AI chat bot with: "Based on my current cash balance of $125K, monthly revenue of $250K, and expenses of $175K, what five critical questions should I ask about my cash flow situation?" The AI might suggest questions about accounts receivable aging, collection timelines for outstanding invoices, which vendor payments could be strategically delayed, upcoming payroll obligations, or whether there's adequate cash runway for seasonal fluctuations—practical questions that reveal the gap between your healthy revenue numbers and the actual cash position of your business.

Translating data into story

But asking the right questions is only part of the equation. Even with complete data visibility and well-formulated questions, the final critical skill is interpreting what you find. This is where technique separates amateurs from the Dr. Brennans of the business world.

What made Dr. Brennan's team at the Jeffersonian so exceptional wasn't just access to evidence or knowing which questions to ask—it was their mastery of technique. Two forensic teams could examine the same bones, ask similar questions, yet reach entirely different conclusions. The differentiator? How they connected seemingly unrelated evidence points into a coherent, actionable story.

This mirrors my journey at Garage Mobility. I've always been naturally curious, constantly probing our operations with questions. But two critical gaps held us back initially. First, our business data resembled a crime scene with evidence scattered across departments, spreadsheets, and systems. We addressed this by creating a unified data layer that gave us the visibility we covered in our last article. But the game-changing decision was hiring a brilliant young ML scientist—someone who brought the right techniques to make sense of our mountain of data.

So, when hyperinflation hit and quarterly price hikes from suppliers became the norm, this foundation of visibility plus technique allowed us to construct a story that drove immediate action. We built a comprehensive understanding of our cash conversion cycles across different product lines, which revealed unexpected opportunities. The data told us which products we could stock longer in anticipation of price hikes—effectively creating margin through arbitrage while managing our working capital constraints. We identified which SKUs had price elasticity headroom, allowing us to pass some inflation costs to customers without losing volume. Most importantly, we could see which products were steadily becoming unprofitable due to supply chain dynamics, prompting our strategic decision to slash our catalog from dozens of brands to just half a dozen high-performers.

This creates an enormous opportunity for innovators to build full-stack AI solutions specifically for small business operations. Tools that can understand the complete context of a business—connecting financial results with operational realities—and help owners translate their scattered data points into coherent stories that drive specific actions. The Main Street businesses that can access these capabilities won't just survive; they'll thrive by making decisions with the same data sophistication that was once exclusive to enterprises with dedicated analytics teams. In the following we will revisit how you can use the actionable insights from your data story to drive results.

THIS WEEK ON MAINSTREET

📰Mainstreet Minute:

Visa unveiled the new era of commerce this week with AI agents that could use your credit card to perform tasks such as shopping on your behalf. While it is still unclear how long this will take to gain ground, it is clear Visa clearly believes the future of shopping is agentic and personalized. This presents a new paradigm and challenge for small businesses who will need to think about ensuring their offerings are also discoverable and compatible with AI-driven shopping.

📌Quick Win Tip:

Clipdrop is the fastest way to create stunning visuals—without mastering complex design software. Powered by AI, it offers tools like background removal, object cleanup, image upscaling, and generative fill to transform your photos effortlessly. Whether you're enhancing product images or crafting social media content, Clipdrop simplifies the process. Let Clipdrop handle the heavy lifting for your content creation so you can focus on your creative vision. Get a quick win this week by trying out Clipdrop's AI-powered editing tools.

I first came to understand the real impact of working capital management about three years ago, while running Garage—a B2B auto parts marketplace moving hundreds of thousands of dollars in inventory every month. We operated lean, with 80% of our stock on consignment and 1–2 week payment terms from suppliers. That meant every day was a balancing act—tracking cash flow, managing inventory, and ensuring we could stock up when needed and pay suppliers on time. I had five different spreadsheets opened daily just to keep up. But I didn’t truly grasp the stakes until I made a costly mistake.

One afternoon, I was staring at our inventory report and realized we were sitting on $150K worth of slow-moving SKUs. In an effort to drive growth, we had stocked aggressively with minimal controls. Now our cash was locked up. When our payables came due, we found ourselves in a bind and had to heavily discount goods to free up cash. We took a loss.

In hindsight, I wasn’t short on data. I was just missing the full picture. I looked at inventory and payables, but failed to spot patterns in sales, customer routes, and reorder cycles. I was managing the trees but missing the forest.

Many small business owners today fall into the same trap I did, spending hours digging through spreadsheets, email threads, and disconnected reports just to answer basic questions about their business. That’s the trees: all the scattered data points pulling your attention in different directions. But what really drives smart decisions is the forest, the ability to see trends, spot issues early, and connect the dots across your entire operation

When You Can't See The Forest For The Trees

This visibility crisis affects virtually every small business owner. We get so caught up in day-to-day operations that our overall business health remains obscured. It's not that we're not looking; it's that our data is so fragmented we can't see the complete picture.

At Garage all of my systems didn’t talk to each other. I still had siloed data across functions. I spent hours with inventory data and payables, but missed critical connections to sales patterns and delivery routes. Some routes moved specific SKUs rapidly, but we didn't service those routes frequently enough—yet we kept overstocking those parts. The result was tied-up capital and avoidable cash flow strain.

Why Business Owners Struggle With Visibility

When you’re wearing multiple hats and running from one fire to the next, three major obstacles tend to get in your way:

1. Data Fragmentation
   Critical data lives in separate places—accounting software, CRMs, inventory management tools, emails, google sheets and a graveyard of documents in the file room or sitting on some on premises server from the 80s. Without a unified system, your visibility is constantly compromised.

2. No Single Source of Truth
   When data lives in multiple places, which version do you trust? Teams end up second-guessing numbers, reconciling reports, and wasting time trying to confirm what’s actually accurate.

3. Time Constraints
   Even if you manage to bring all the data together, do you have the time to analyze it meaningfully? For most business owners, the answer is no. Between customers, vendors, and day-to-day firefighting, there’s little room left for deep analysis.

This leads to many business owners stuck in the trees without clear insights to make critical decisions. And when you can't see the forest, you’re more likely to miss the small leaks that eventually sink the company.

Knowledge Base Management To Help You See The Forest

Modern knowledge base management tools such as Glean and Rima act like a unified layer across all of your data and workspace tools. It eliminates the data fragmentation issue and provide business owners with a single source of truth to search and find the right answers quickly, eliminating hours spent fiddling across data and disparate tools just to find an answer to make a decision. These tools also allow you to see the forest – patterns and trends across your data – so you can spot issues and prevent cash leaks that ultimately sink the whole ship.

In the age of AI, these tools give every business owner super powers and businesses who do not use these tools will be at a severe disadvantage to those who do. These tools help turn your scattered operational documents into an intelligent unified knowledge base. This enables you to spend less time searching and more time growing.

In other words, for the first time, AI can help small business owners see both the forest and the trees.

How Do You Get Started

If your business is operating with disconnected systems (and let’s be honest, most are), don’t boil the ocean. Your first step isn’t to install a dozen new tools. It’s to find one AI solution that plugs into your existing data systems and tools and brings everything together.

Today’s knowledge base AI tools offer a range of options. Glean and Notion AI are great for search and information access across silos. Guru offers a streamlined experience for centralizing team knowledge and surfacing insights. And Rima AI, built specifically for small businesses, not only connects siloed tools but also helps spot cash leaks, suggest actions, and even automate fixes—all without needing a full ops team.

In the following week, we will discuss what to do once you have visibility and can make better sense of your operations instead of spending hours digging through paper trail.No major developments to report this week. We'll be back with fresh insights in our next update.

THIS WEEK ON MAINSTREET

📰Mainstreet Minute:

No major developments to report this week. We'll be back with fresh insights in our next update.

📌Quick Win Tip:

Docwelo is the fastest way to generate professional documents—without wrestling with templates or legal jargon. Designed for freelancers and small businesses, it lets you create contracts like NDAs, service agreements, and employment offers in minutes. Just answer a few smart prompts, and Docwelo’s AI delivers a perfectly formatted, legally sound document ready to go. With bank-grade security, real-time collaboration, and a pay-as-you-go option starting at $2 per document, it’s the no-brainer way to handle paperwork. Get a quick win this week by automating your next contract with Docwelo.

Growing up, I was convinced medicine was the highest calling anyone could have. Everything I did up until 2011 was aligned to becoming a doctor. Then in the summer of 2012, I suddenly realized I didn’t want to be a doctor. After taking the Medical College Admissions Test (MCAT) twice, painstakingly writing essays, and finally submitting my medical school application, I made one of the hardest decisions of my life - I withdrew my application. I’ve never felt more lost or uncertain than I did that summer. For three long years, I drifted without direction. Despite having a successful job at the time, I was deeply frustrated and confused about what I wanted to do with my life. Then one day, I stumbled upon a TED Talk by Fred Swaniker, who spoke about his entrepreneurial vision to nurture the next generation of African leaders. Something clicked instantly; I felt an undeniable pull, an itch I couldn't ignore. I answered the call and haven't regretted it once. Whether you've started your own business or acquired one, you probably felt something similar: a voice, a passion, an urge you could no longer ignore. At first, that feeling is exhilarating, but nothing truly prepares you for the transition from inspiring idea to operator.

Transition Hell: From Idea to Managing Reality

70% or more of small businesses fail within their first five years of operations. I have started two companies, so I understand firsthand the mental trauma and challenges of transitioning from the dopamine rush of an exciting idea to the reality of running business operations. In 2022, I started Garage Mobility, a B2B auto part marketplace that connected traders to auto part suppliers and provided inventory financing to ensure their shelves remained stocked. Within 5 months of operations, we were generating about 250K USD in monthly sales volume. It was incredibly exciting to see our idea take off but no one had quite prepared me for what came next. Most days I felt like I was losing my mind and struggling to keep up.

By day, I was wearing multiple hats - handling sales, procurement, fulfillment, inventory, HR, and finance all at once. By night, I was moonlighting as a business analyst, pulling data from disconnected systems, trying to understand our inventory position, whether we had enough stock to meet demand the next day, how much we had sold, and if there were any cash or inventory reconciliation issues. Fair or not, the world expects business owners to be superhuman. Other than raising my cortisol levels, this constant context switching also had some downsides for the business.

When you’re focused on growth and wearing multiple hats, it’s easy for things to fall through the cracks. What kills most businesses isn’t a single catastrophic failure, it’s the small inefficiencies, unchecked leaks, missed signals that slowly drain you until it’s too late. In our case, even though sales were growing, we were unknowingly overstocking slow-moving SKUs; placing repeated orders without realizing they weren’t turning. Over time, this quietly compounded and increased our inventory days. It wasn’t until we started facing pressure with payables that we realized we had a working capital issue. To free up cash, we had to heavily discount the excess inventory, which led to losses and strained our cash flow at a time when we needed flexibility the most.

Why Do Small Business Owners Struggle Through Transition Hell

Transition hell is real for many business owners and is an implicit reason why businesses fail. Most small business owners are experts on the product and service they sell. However, as the business starts growing, they are expected to be superhuman and wear multiple hats. This leads to things falling through the cracks such as poor inventory management, high discount dependency, high churn that’s remains unattended, etc and all these have significant impacts on a business’ cash flow which is one of the top reasons businesses fail.

As I reflect on my own experience and that of other business owners I’ve been fortunate to chat with, I believe there are three major reasons why small business owners struggle through transition hell.

First is poor visibility. During this phase, business owners are wearing multiple hats; yet they lack centralized, easy access to the data needed to make quick, informed decisions across functions. Information is often scattered across spreadsheets, disconnected tools, and handwritten notes, making it nearly impossible to get a full picture of what’s really going on in the business at any given time.

Second is the lack of operational forensics knowledge. Even when business owners start pulling data from different sources, many don’t yet have the mental models or experience to clearly identify what inefficiency looks like. For instance, stocking slow-moving SKUs while celebrating growing sales may seem harmless until you run into cash flow issues. Or you may think discounts are helping drive volume, without realizing you’re eroding already thin margins. Without understanding the levers that drive financial health, even well-intentioned decisions can quietly bleed the business. And again, siloed data compounds this issue because if you can’t bring all your information together, you can’t diagnose what’s going wrong.

Third is limited access or knowledge on how to use automation tools effectively. Even when inefficiencies are spotted, many small business owners aren’t sure where to start, what tools are available, or how to deploy them in a way that creates repeatable, sustainable impact. This leaves them stuck manually solving the same problems over and over again.

Together, these three challenges—poor visibility, limited operational forensics, and a lack of automation trap many business owners in a loop where growth hides inefficiency, and hard-earned momentum is slowly drained by the cracks beneath.

The key to unlocking abundance on Main Street is to significantly reduce the failure rate of small businesses. Over the next 10 years, millions of businesses will be passed down from boomers to first-time owners and many more will be created from scratch. These new operators will be faced with the same challenges of transitioning into operators.

In this series, Superhuman Business Owner, we’ll explore how to harness AI to help small business owners transition successfully into operators and scale their businesses without loosing their minds.

THIS WEEK ON MAINSTREET

📰Mainstreet Minute:

OpenAI recently launched free AI education through its OpenAI Academy. This is an ambitious project to drive the mainstreet adoption of AI beyond tech savvy users. small business owners, teachers, students and others can now access free courses through the ChatGPT creator. Not to be outmatched, google also partnered with America small business development center (SBDC) to launch AI U, an initiative to train over 100K small businesses on AI literacy and tools to help grow their businesses.

📌Quick Win Tip:

Brand24 is the fastest way to monitor your brand across the internet—without digging through dozens of platforms. It tracks mentions from social media, blogs, news sites, and forums in real time, and uses AI-powered sentiment analysis to surface what really matters. Let Brand24 handle the noise so you can focus on building stronger customer relationships, protecting your reputation, and spotting opportunities as they happen. Get a quick win this week by setting up real-time alerts for your brand with Brand24.

This week marks the 10th consecutive week of writing this newsletter.

AIM started almost on a whim. I usually shy away from sharing my opinions or inner thoughts publicly. But I’ve always loved learning, and I get a deep sense of energy from solving problems. I’ve never been more excited about a technology than I am about AI, and I find myself constantly chasing knowledge on the subject.

While I’m currently building an AI venture focused on helping small businesses do more with less, I realized there are millions of Main Street businesses I won’t be able to reach through that venture alone. So, I decided to write; using this newsletter as a platform to share what I’m learning with you all. My aim has always been simple: to break down the challenges and celebrate the successes of AI adoption on Main Street, and to offer practical ways for businesses to harness AI effectively.

At times, it’s been uncomfortable. I’ve often wondered if I could really write or whether my content was useful to anyone. But I leaned into what I know best—stay consistent and learn as I go.

Now, 10 weeks in, I want to thank every subscriber who has followed along and engaged with my writing. I truly hope it’s added value to your work and thinking.

To celebrate this milestone, I’ve created a free gift for Main Street businesses, which I deeply care about. Small business owners wear many hats. I know this firsthand from running multiple ventures. And as the tidal wave of AI sweeps across industries, many are asking the same question: Where do I start?

So, I put together a simple, step-by-step guide based on my research, writing, conversations, and practical experience to help small businesses implement AI in their operations. It includes actionable worksheets you can use to complete each step.

If you're looking for theory, this guide isn’t for you. But if you want a hands-on, practical roadmap to get started with AI, download the free e-book with the link below and share it with a business owner who might find it helpful.

I am excited about what the future holds and to continue to grow and learn along side you all. if you believe there is someone that can benefit from this newsletter please share with them so they can join the conversation.

Some of you over the past weeks have reach out to me with words of encouragement or constructive criticism. I love it all and i love hearing from you guys so please keep it coming.

E-book:

Download for FREE

Thank you to Silas Adomako for contributing to this blog post as a co-author

In the early 1990s, my dad made a bold decision—he wanted to move our family to the U.S. Every choice he made after that, from sending us to English-speaking schools in Cameroon to how he structured our household, aligned with that vision. For ten years, he applied for the U.S. green card lottery and was rejected each time, yet he never wavered. Though he was a man of deep faith and conviction, he still carried the quiet voice in his mind all those years: am I making the right decision? Ultimately, his metric for success was simple—give 100% of his children access to the best education in the world—and in the end, things turned out exactly as he’d hoped.

Everyone contemplating a big decision carries that same quiet voice in their head: am I doing the right thing? For many small business owners, especially as they consider implementing AI across their workflows, that question becomes more pointed: how do I know whether I’m making the right financial decision? Unlike big Wall Street firms that can afford to experiment, small businesses don’t have the same margin for error. That’s why it’s critical for small business owners to measure the impact of AI clearly and quickly—to ensure the investment is driving real value.

The core promise of AI for businesses is increased productivity—in other words, doing more with less. But how do businesses measure whether it's working? In this blog post, I argue that Annual Revenue per Full-Time Employee (AR/FTE) is the most important metric to track to gauge the impact of AI on your company.

This ratio reflects the effect of AI on the most fundamental value equation in business: how much value each person creates. And while it’s often used in SaaS and tech companies under the term AR/FTE (Annual Recurring Revenue per FTE), its relevance is just as strong for small businesses that rely on limited human resources to grow and thrive—even if their revenue isn’t recurring.

Maximizing Economic Value

A business exist to maximize economic value for its shareholders. At its core, the economic value of a company can be defined simply as;

For small businesses, even a modest drop in revenue or a slight increase in costs can be the difference between staying afloat and shutting down. In fact, according to the U.S. Bureau of Labor Statistics, 20% of small businesses fail within their first year, often crushed by rising costs and stagnant revenue. Therefore, as long as a business makes more than it spends, the overall economic value of the company will be maximized, and the business increases its chances of survival.

Since Labor is often the largest expense for small businesses, accounting for up to 70% of total costs in small businesses, any solution that increases revenue per employee directly boosts economic value. That’s why Revenue per Full-Time Employee (FTE) is such a critical metric for assessing the impact of AI implementation. and properly assess value creation.

Earlier in the Mainstreet Champions series, I introduced the 4S-3R framework as a way for small businesses to prioritize AI opportunities. I recommend starting with quick wins - initiatives that offer immediate productivity gains with minimal effort. But those gains only matter if they ultimately translate into higher revenue or lower costs. That’s where the Annual Revenue per Full-Time Employee (FTE) ratio comes in - it captures productivity improvements in clear financial terms.

Traditional financial metrics like gross revenue, profit margins, and customer lifetime value (LTV) are important, but they often provide an incomplete picture of AI’s true impact on small business operations. While these metrics reflect overall growth or profitability, they don’t always highlight productivity improvements or hidden inefficiencies. For example, a business might grow revenue by 50% year over year but double its headcount in the process thus becoming less efficient. Similarly, profit margins might temporarily improve due to one-off cost cuts or favorable market conditions, masking deeper operational inefficiencies. Even metrics like LTV or average order value (AOV), which focus on customer behavior, can overlook whether growth requires disproportionate labor inputs.

So, for small businesses adopting AI, Annual Revenue/FTE cuts through the noise of vanity metrics and forces business owners to ask the tough question; is AI helping my team do more with less. A rising ratio signals that AI is amplifying productivity and increasing your businesses economic value. A flat or declining ratio, on the other hand Is a warning sign and an indication that business owners should reevaluate their AI implementation strategy.

Accounting Firm Case Study

Consider a boutique accounting firm that provides services in financial reporting, tax compliance, and bookkeeping. Today, the firm serves 120 clients and brings in $1.5 million annually with a team of 13 employees. While the owner is interested in expanding the client base, the firm’s manual onboarding and service delivery processes are a bottleneck. Rather than increasing headcount, the owner is exploring AI as a growth lever specifically to automate the client delivery workflow. He projects that with AI, he could onboard 30 additional clients and generate an extra $500K in revenue, bringing total sales to $2M annually - without hiring anyone new. The AI system would cost a $1,500 one-time implementation fee plus $95–$150 per month, a fraction of the alternative: hiring a new manager and associate for roughly $10,000 per month.

The table below compares the two growth scenarios. While both result in a 33% increase in annual revenue, AI implementation drives a 33% lift in Revenue per FTE, compared to just 15% with additional hiring. In other words, AI delivers greater productivity gains by enabling the firm to grow without adding headcount. Over time, this approach also sets a higher ceiling for both revenue and operational efficiency – driving stronger economic value for the firm

Limitations of AR/FTE

Some critics might argue that AR/FTE has its limitations since not every employee contributes directly to revenue, and the ratio doesn’t account for variable costs such as raw materials, which may not be directly affected by AI. These are valid points. It’s true that not all employees are revenue-generating, especially in the early stages of a business when owners may over-hire to build out capacity ahead of growth. However, for most small businesses, labor remains the dominant cost, often accounting for two-thirds or more of total expenses. That’s why, for the sake of simplicity, our focus in this piece is on the key drivers of economic value for a company that AI can directly influence namely, revenue and labor cost.

Furthermore, while over-hiring may be a temporary and strategic decision, sustained growth depends on improving output per employee. Our focus for this piece was to recommend a measure for assessing the impact of AI on sustained growth. This makes AR/FTE a useful directional metric for evaluating whether AI investments are truly translating into economic value creation.

AI holds great promise as a tool to help small businesses punch above their weight and unlock abundance. I strongly believe over the next few years, AI will drive unprecedented levels of productivity for Mainstreet businesses and create new opportunities that were unimaginable before.

I hope this Mainstreet Champions series has been a helpful guide for many small business owners thinking through how to implement AI within their organization and measure the impact of their investment.

THIS WEEK ON MAINSTREET

📰Mainstreet Minute:

No major developments to report this week. We'll be back with fresh insights in our next update.

📌Quick Win Tip:

Dextra is the AI-powered sales assistant built for independent contractors who want to win more B2B clients—without spending hours on cold outreach. It automates prospecting, crafts personalized emails, and manages replies so you can focus on closing deals. With the ability to send up to 1,500 targeted emails per month, Dextra helps you expand your client base effortlessly. Get a quick win this week by letting Dextra handle your outreach while you focus on growing your business.

With 26 seconds left in the game, its second and goal at New England’s 1-yard line, New England is up 28-24 and the Seahawks have the chance clinch the game with a touchdown. However, to everyone’s surprise, Pete Carroll calls a pass play, putting the ball in Russel Wilson’s hands who throws a slant pass to Ricardo Lockette but the ball is intercepted by Malcolm Butler. In what many pundits consider to be the most controversial call in super bowl history, Seattle looses the game. Coach Pete Carroll’s call left everyone puzzled – why didn’t he just run the ball. In the 2014 season, teams ran the ball 74% of the time when at the one-yard line with a success rate of about 54%. Therefore, by all logic, the safer and more straightforward decision would have been to run the ball. However, in most people’s eyes Pete used a sledgehammer to kill a fly and ended up costing his team a championship.

So, what does Pete Carroll’s decision in Super Bowl 49 have to do with AI and small businesses? As small businesses explore AI today, it’s easy to get fancy—choosing complex agentic tools that promise full autonomy and decision-making on your behalf. But while fully autonomous agents sound exciting, most business tasks today don’t need that level of complexity. In fact, the majority of tasks can be handled through simpler, AI-powered workflow tools that are still smart, efficient, and reliable. Hence, one key lesson small business owners can learn from Pete Carroll is that most times the simplest answer is often the correct answer. In most cases, business owners should opt for a straightforward, low risk and low complexity tool and only increase complexity gradually as these tools get better and the technology matures. Below I outline a simple and practical mental model small business owners can use when selecting the right tool to implement AI in their organization.

Selecting a Tool Type

In previous posts, we introduced a practical framework for prioritizing AI opportunities and building a clear implementation roadmap. We also outlined how to define the automation scope for each opportunity—categorizing them into task, process, or full operational automation.

This step is critical to selecting the right tool. Business owners should use a two-step approach:

1. Identify the appropriate tool type based on the automation category.

2. Evaluate tools within that category using a SIMPLE checklist framework to determine the best fit.

Before we dive into the checklist, here’s a quick guide to help match each automation category with the right tool type:

1. Task Automation

Task automation targets simple, repetitive actions that are time-consuming, error-prone, or frustrating such as scheduling meetings, generating invoices, or replying to basic customer queries.

Recommended tool type:

• SaaS tools with built-in automation

• Task automation platforms (e.g., Zapier)

• AI workflow platforms (e.g., Rima, Gumloop)

2. Process Automation

Process automation covers multi-step workflows across different functions such as customer onboarding, invoice processing, or inventory management. The recommended tool type for this category is a no-code/low-code workflow platforms that connect multiple systems

3. Full Operational Automation

This level integrates entire business systems such as CRM, ERP, or supply chain platforms into a cohesive, AI-powered operating layer. The recommended tool type for this category is an advanced no-code/low-code workflow platforms with support for complex logic and dynamic workflows.

Keep it SIMPLE: Practical AI Tool Evaluation Checklist

Now that you've matched your automation scope to the right tool type, the next step is choosing the right solution within that category. Businesses should use this SIMPLE checklist framework to evaluate AI tools that are practical, efficient, and built for real business impact.

S – Show Me the Steps
Transparency and usability are key. If it’s a workflow automation tool, it should offer pre-built templates and the ability to connect multiple systems without writing code. You should always have visibility into how the tool operates—what goes in, what comes out, and what happens in between.

I – Implementation Timeline
Speed matters for small businesses. Look for tools that can be implemented in days—not weeks, with structured onboarding, minimal friction, and a clear path from sign-up to value. Long setups and complex integrations are red flags.

M – Measurable Outcomes
If you can’t measure the impact, it’s not worth implementing. The tool should clearly demonstrate ROI through metrics like time saved, fewer errors, or improved revenue. Look for built-in dashboards or reporting to make tracking simple.

P – Proven Track Record
Choose tools with a track record of success in businesses like yours—by size, industry, or need. Prioritize vendors who offer case studies, testimonials, or live references. A mature product with real-world results is always better than a shiny new experiment.

L – Low Technical Debt
Avoid tools that require heavy maintenance or technical support. The right tool should be easy enough for non-technical staff to use, with clear documentation, strong customer support, and a well-explained data flow that shows how inputs turn into outcomes. The less hand-holding required, the better.

E – Economic Value
Pricing should be simple, transparent, and tied to value. Avoid tools with confusing tiers, hidden costs, or per-seat pricing models that charge before you see results. The best tools let you pay as you go and scale as you grow—so you’re always paying for actual value, not just access.

When it comes to tool selection, it is important to keep it simple. Starting simple not only gets the job done but helps businesses avoid tools that over promise and under deliver in real-world use. As these technologies mature, complexity can be layered in gradually when it truly adds value. As we close out our series on Mainstreet Champions, we will explore ways small businesses should be measuring impact in the age of AI.

THIS WEEK ON MAINSTREET

📰Mainstreet Minute:

Verizon has entered the AI arena with the launch of Verizon Business Assistant, a generative AI-powered text messaging solution tailored for small businesses. This innovative tool automates responses to common customer inquiries, providing 24/7 support and allowing business owners to focus on higher-value tasks. Over time, the assistant learns from interactions, enhancing its accuracy and efficiency. The service is available for $35 per month per line.

📌Quick Win Tip:

Framer is the fastest way to design and publish beautiful, responsive websites—without writing code. It seamlessly integrates with tools like Figma and offers built-in SEO, hosting, and animations to eliminate development bottlenecks. Let Framer handle the technical complexity so you can focus on creating, launching, and growing your brand. Get a quick win this week by building your next landing page with Framer.