Second Brain

The last 3 months has radically changed the potential of how 80% of small businesses will operate.

The M&A valuation process for most has not caught up yet. Here’s what you need to know if you’re going to be on either side of an M&A deal between today and 2030.

How SMB Valuation Actually Works

The process for valuing an SMB is radically different than the valuation process for most other assets. There are 5 levers that drive the price of a small business:

Distributable cash flow to the owner (CF)
Market rate for owner operator compensation (OC)
Durability of cash flows over time (D)
Transferability of cash flows (T)
Expected Growth Rate of Cash Flows (G)

I call this the SMB asset value equation, broken down as such:

(( CF - OC) / (D + T)) * G

The History of how SMBs have been valued

Consider this typical example for a small business (which has largely unchanged for the last 20 years):

distributable cash flow of $1.25M
Operator replacement comp of 250K yields $1M in EBITDA
Durability is considered healthy (recession resistant, long standing business model)
Transferability is deal dependent, but largely in tact if an operator can be installed, customer/vendor concentration is low, and there is no severe key man risk
The “real” growth rate of the company is 0% after factoring inflation. Most SMBs once matured (as most are at exit), have found a comfortable level and are priced as if they will be steadily maintained, not grown

This profile has typically yielded anywhere from 3-5x EBITDA depending on company specific factors.

How will this change in the AI era?

These five levers of valuation are also what’s known as the “quality” of earnings.

The magnitude, owner dependence, durability, transferability, and growth rate of earnings ultimately dictate how valuable your earnings are to a buyer.

Let’s examine how each lever stands to change with the advent of agentic AI.

Magnitude of cash flows

A company’s cash flows are largely driven by two factors:

It’s revenues
It’s margin structure

I’ve examined a common pattern present in hundreds of owner operated service businesses. Revenue tends to hit a ceiling around $2-3 million yielding about 400-750K of SDE.

Why? Because most service businesses operate on manual labor. A typical SMB looks like this:

Owner operator who is the brain for everything
Office manager / assistant who brute force works the entire back office
A Hodge podge of low to medium cost labor via W2 and contractors that the owner operator must mechanically command on a daily basis

As the business hits around 2M revenue, the business is typically serving several hundred customers. The Business faces two choices.

Cross the SMB chasm: Invest disproportionately in new systems that will help them scale past their operational bottleneck. Lose money for 1-2 years until they hit escape velocity. Non zero percent chance of bankruptcy. This is because the margin structure temporarily becomes hostile.
Uncomfortably profit: Stay put and operate that old thing for 10-20 years, cashing in those unscalable profits each year and avoid the risk that comes with scale. The owner optimizes margins at their local maximum.

How does this change with AI?

The Chasm Disappears: The ability to build new workflows, tools, and agents in hours instead of months means the trade off of today’s profits for tomorrows growth is no longer be required. We may very well see a significantly higher % of companies escape the $2M revenue bottleneck and scale to $5-$10M revenue.
Margin structures change: An overwhelming % of SMB costs are labor related. Even the indirect costs are typically connected to headcount. In an increasingly agentic world, a $2M company that used to be $500K SDE may now very well be a $1M SDE company.
Commodity services evaporate: A significant % of SMBs have historically existed solely because of the natural fragmentation that takes place from having labor heavy, manual operations. As AI reduces and removes the labor bottleneck from many industries, they will go from being fragmented to hyper-consolidated.

Recap: With the messy middle drying up, prepare to see a lot of companies die, and a lot double or triple in size.

Operator replacement comp

The impact of operator replacement comp will be fascinating as the AI landscape transforms SMB.

Historically, the following has always been true:

A significant % of cash flows were a reflection of the owners effort, and therefore cut deep into true EBITDA
Owner dependence made SMBs a shitty investment
Hiring an operator has been notoriously difficult

As SMBs become more agentic, the following will happen:

Agents will be easier to manage than unpredictable employees
Operators will go from default managers to default builders / sales leaders
The old school operator role might evaporate completely

The radical transformation of what it means to manage a business in the AI world should be a massive release valve for SMBs.

Durability

Durability is the variable that drives the most dispersion in multiples, and AI cuts both ways here.

For AI-resilient businesses, durability actually improves. A plumbing company’s revenue is as recession-resistant and structurally durable as it was five years ago, with the added benefit that AI-driven operational efficiencies are now making the cost structure more resilient too. Predictive maintenance scheduling smooths seasonal revenue dips. AI-assisted estimating reduces the lag between lead and contract, which stabilizes cash conversion. The business model hasn’t changed but the operational resilience underneath it has improved.

For AI-exposed businesses, durability takes a direct hit. A bookkeeping firm that has operated the same model for 15 years looks durable on paper. Long-standing client relationships. Low churn. Consistent revenue. But the durability of that model is now a function of how quickly AI alternatives penetrate the client base, and that’s a rate-of-change question that historical financials can’t answer.

The diligence challenge is that durability has traditionally been assessed by looking backward. How did the business perform in the 2020 downturn? What’s the client retention rate over five years? What percentage of revenue is contractually recurring? Those are still relevant questions. But they’re no longer sufficient.

The new durability question is forward-looking: is the business model itself durable against AI substitution? A business can have 95% client retention and five-year contracts and still face a durability problem if the underlying service is being commoditized by technology faster than the contracts can reprice.

In practical terms, this means the durability premium (the portion of the multiple that rewards resilient, recession-resistant cash flows) is going to start splitting. Businesses with AI-resilient models get a durability premium that may actually increase. Businesses with AI-exposed models lose the durability premium even if their trailing financials look spotless

Transferability

Transferability has always been the biggest structural risk in SMB acquisitions. The business is worth what it earns, but only if those earnings survive the owner leaving. Key-person risk, owner-dependent relationships, tribal knowledge locked in the founder’s head. these are the things that kill deals or force earnout structures that neither side loves.

AI directly addresses several of these transferability risks.

Tribal knowledge extraction is now possible at a level that didn’t exist 18 months ago. An owner who has spent 20 years building relationships and institutional knowledge can now work with AI systems to document processes, decision trees, and client management protocols in ways that make the business meaningfully less dependent on their continued involvement.

Operational systems that used to require years of owner training to transfer can now be partially automated. A new owner stepping into a business with AI-assisted operations (automated quoting, AI-driven customer communication templates, intelligent scheduling) has a shorter ramp to full operational capacity than a new owner stepping into a business that runs on the previous owner’s instincts and relationships.

The transferability improvement directly affects multiples. A business that a buyer would have discounted 0.5-1.0x for key-person risk may now warrant a smaller discount (or no discount) if the owner has systematically embedded AI into the operational infrastructure in a way that reduces the transition risk.

For sellers preparing for exit over the next 2-3 years, this is probably the highest-ROI investment they can make. Using AI to systematize operations and reduce owner dependency doesn’t just make the business easier to run. It makes the cash flows more transferable, which directly increases what a buyer will pay.

For buyers, the diligence question flips: has the seller used AI to improve transferability, or is the business still running on the owner’s judgment? A seller who can demonstrate AI-embedded operations is presenting a lower-risk acquisition. A seller who hasn’t engaged with AI at all is presenting a business that will require more post-close investment to stabilize.

Growth Rate

Most mature SMBs are priced at 0% real growth. The implicit assumption is that the business has found its level and will be maintained, not grown. The buyer is paying for the current earnings stream, discounted for risk and adjusted for transferability.

AI breaks that assumption in both directions, and the mechanism is older than most people realize.

In 1865, economist William Stanley Jevons observed that as steam engines became more efficient, coal consumption didn’t decrease. It increased. Dramatically. The efficiency gains made coal-powered applications economically viable in contexts where they previou