Home/ Journal/ Memo

How to measure ROI on a mid-market AI engagement.

You do not measure AI ROI across a company; that number is unmeasurable and the question is a trap. You measure it on the one differentiated workflow you commissioned, against a baseline you name before the build. State what the workflow costs you today in staff hours, error and rework cost, or lost cycle-time that maps to revenue. Then measure the same number after. Count hard dollars (headcount avoided, cycle time turned into revenue, errors removed) and discount vanity metrics like generic time saved and seat counts. If you cannot name the baseline, you are not ready to commission, and that inability is itself the finding.

"What is the ROI on AI" is the question every mid-market operator is told to ask, and asked at the level of the whole company it has no honest answer. This memo gives the version that does have one. ROI is measured per workflow, against a baseline you write down before the build, in hard dollars rather than vanity metrics. The trap of the company-wide number, the baseline test that tells you whether you are even ready, the payback-period framing, and why we would rather you walk away than commission a workflow you cannot measure.

MemoJune 2026
Read time7 minutes
AudienceOwner-CEOs, COOs, Managing Partners

The honest answer, up front.

You cannot measure the ROI of AI across your company, and you should stop trying. There is no single number that captures what AI did for an $8M-$50M business, because a company is not a workflow. The question only becomes answerable when you shrink it to the thing you actually commissioned: one workflow, with one baseline cost, measured before and after. Ask "what is the ROI on AI" and you get a story. Ask "what did this workflow cost us before the build, and what does it cost us now" and you get a measurement. That shift, from the company to the workflow, is the whole memo.

The reason the company-wide question is a trap is that it sounds rigorous while being impossible to answer. Spread AI across a dozen workflows and the gains and costs blur into a figure nobody can attribute or defend. Any number you produce is assembled from estimates, and estimates are where vendors live. The disciplined operator refuses the company-wide frame on purpose and measures the one place a clean before-and-after actually exists.

This matters more for a mid-market company than for an enterprise, because you do not have a finance team standing by to model attribution across a portfolio of initiatives. You have a workflow, a cost, and a question about whether the spend was worth it. The good news is that the smaller frame is the more honest one anyway. A single workflow with a named baseline is something an owner-CEO or a COO can verify personally, in an afternoon, without commissioning a study to measure the study. The company-wide ROI deck is the thing that looks sophisticated and proves nothing; the one-workflow before-and-after is the thing that looks modest and settles the argument.

The baseline you can name, or cannot.

Here is the readiness test, and it is harsher than it looks. If you cannot name the number the target workflow costs you today, you are not ready to commission a build. Not "we think it is a lot." A number: this workflow consumes roughly this many staff hours a month, or it produces roughly this much error and rework cost, or it loses us roughly this much cycle-time and speed that maps to revenue we can see. If you cannot produce that figure, the inability is itself the finding, and it is a valuable one. It means the workflow is not understood well enough to build against yet.

Naming the baseline is step one of measuring ROI, and it is the step almost every operator skips. People reach for the projected upside ("this could save us a fortune") before they have anchored the thing it is saved against. The baseline is the anchor. Without it, every later number floats, and a floating number is indistinguishable from a guess. The work of naming it (counting the hours, pricing the rework, tracing the lost speed to dollars) is not overhead on the engagement. It is the engagement starting.

This is also where the cost side of the decision gets concrete, and it pairs with the spend you are weighing; we walk through the real numbers in what a mid-market AI engagement costs. A baseline you can name on one side and a commission price you can name on the other is the only honest version of an ROI calculation.

Hard dollars versus vanity metrics.

Once you have a baseline, the measurement is a matter of counting the right things and refusing to count the wrong ones. Three categories are hard dollars, and they are the only categories that prove an engagement paid for itself.

Headcount avoided or redeployed. The workflow used to require people, or a fraction of people's time, and now it requires less. That is a real number you can put against payroll, whether it shows up as a role you did not have to hire or hours given back to people who now do higher-value work.

Cycle time that converts to revenue or capacity. The workflow used to take a week and now takes a day, and that speed maps to something countable: deals closed faster, more volume through the same team, a response time that wins business you were losing. Speed only counts when you can name what it buys.

The cost of errors removed. The workflow used to produce mistakes that cost money to fix, or cost money when they reached a client. Remove them and you remove a real expense that was hiding in rework and goodwill.

Everything else gets discounted, hard. Generic "time saved" estimates with no dollar attached, usage and seat counts, "productivity" as a feeling: these are how vendors inflate ROI decks, and they are seductive because they are always large and always favorable. A metric that cannot be traced to money that shows up in the business is a vanity metric. It may be fine for tracking adoption. It does not prove the engagement was worth its cost, and you should never let it stand in for the proof.

The payback period, in nameable months.

A properly scoped single-workflow commission should pay back in a number of months you can state out loud. This is not optimism; it is arithmetic. If the workflow costs a concrete amount every month in staff hours and error cost, and the commission costs a concrete amount once, you divide one into the other and you have a payback period. The number is only available because the baseline is concrete, which is exactly why naming the baseline came first.

The diagnostic is the inverse. If nobody on the team can state the payback period, the problem is almost never that the build is too risky to forecast. The problem is that the workflow was not scoped tightly enough to measure. A scope that sprawls across "improving operations" has no baseline and therefore no payback period, and that is the tell that the engagement is shaped wrong. The fix is to tighten the scope until a single workflow with a single baseline emerges; that workflow is the one to build. This is the same discipline that separates a buy from a build in the first place, which we lay out in build vs buy AI for a mid-market company.

A payback period also keeps everyone honest about magnitude. A workflow that costs you a few hundred dollars a month in hours is not worth a six-figure commission no matter how satisfying the build would be, and the arithmetic says so immediately. A workflow that quietly costs you tens of thousands a month in rework or lost speed justifies a serious build and justifies it fast. The point of the number is not to produce a flattering slide; it is to tell you, before you spend, whether the math closes at all. Most of the engagements that go wrong were ones where nobody did this division, because the moment you do it, the bad ones disqualify themselves.

Set the measurement before the build.

The single most common way operators fool themselves on AI ROI is measuring against a story told after the fact. The build ships, something feels better, and a number gets reverse-engineered to justify what was already spent. That number is not a measurement; it is a press release. The only honest comparison is against the baseline you wrote down before the engagement started, because that baseline was recorded when you had no incentive to flatter it.

So the measurement plan is part of the scope, not an afterthought. Before the build begins, you decide which hard-dollar numbers you will track, you record their baseline values, and you agree on when you will re-measure. This sounds obvious and is almost universally skipped, because the moment of greatest enthusiasm (kickoff) is the moment nobody wants to slow down and write down what failure would look like. Writing it down anyway is what makes the eventual ROI claim defensible. A build that drifts past this point is also how engagements stall without anyone noticing, a pattern we trace in why mid-market AI rollouts stall in month four.

How to run the measurement without us.

You can do most of this before any vendor is involved. Pick the one workflow that is your actual bottleneck. Name its baseline cost in the three hard-dollar categories: hours, error and rework, lost speed that maps to revenue. If you cannot, stop; that is your answer for now, and it is a finding worth having. If you can, you have both the case for the build and the yardstick to judge it by, and you can hold any vendor to it.

If you want the sorting done for you, our AI Maturity Index surfaces the workflow worth measuring and the baseline questions to ask of it, and our diagnosis process exists to set that measurement before any money is spent. The framing to keep is simple. Measure per workflow, not per company. Name the baseline first. Count hard dollars and discount the rest. And set the yardstick before the build, not after, because a number you wrote down in advance is the only one anybody should believe.

Field-note context

Where the ROI question actually lives.

ROI is measured per workflow, not per company.

The mistake operators make is asking for one ROI number for "AI" the way they would ask for one number for a marketing channel. AI is not a line item; it is whatever workflow you pointed it at. Almost every honest ROI figure in this space comes from a single isolated workflow with a clean before-and-after, and almost every dishonest one comes from summing across workflows where nothing can be attributed. Shrink the question until it has an answer, and it will be one workflow at a time.

The baseline is a readiness test in disguise.

Naming what the target workflow costs you today is not just step one of measuring ROI; it is the cheapest diagnostic you can run on whether to commission at all. An operator who can state the baseline understands the workflow well enough to build against it. An operator who cannot is not ready, and the gap between those two is usually a few hours of honest counting, not a consulting engagement. If you do nothing else from this memo, go name the number. Where it comes out and whether you can name it tells you more than any vendor pitch will. The AI Maturity Index is built to walk you through exactly that.

Why we would rather you walk away.

We commission custom builds for a living, so telling an operator not to commission until they can name a baseline is against the obvious short-term interest. We do it anyway, because an unmeasurable engagement makes neither side look good. A build with no baseline cannot be proven to have worked, which means it cannot be defended internally, which means it sours the operator on the whole category and sours us on the reference. We would genuinely rather an operator walk away and come back when the baseline is nameable than take on a workflow whose ROI can only ever be a story. The honesty selects for engagements that succeed, which is the entire point.

Extended questions

The ROI questions buyers ask next.

How do you actually measure ROI on an AI engagement?

Not across the whole company; on the one workflow you commissioned, against a baseline you wrote down before the build. Name what that workflow costs you today in staff hours, error and rework cost, or lost cycle-time that maps to revenue. After the build, measure the same number and compare. Company-wide AI ROI is unmeasurable, which is what makes the company-wide question a trap. Per-workflow ROI is measurable precisely because the baseline is a concrete number, and that same concreteness is what lets you state a payback period in months.

What payback period should a single-workflow build hit?

A nameable number of months, because the baseline cost is a concrete figure you divide the commission price into. If the workflow burns a real amount every month in hours and error and rework, the payback period is arithmetic, not optimism. And if nobody can state the payback period, read that as a scoping problem rather than a risk problem: the workflow was not isolated tightly enough to measure. Tighten the scope to a single workflow with a single baseline before any money is spent, and the number appears on its own.

Why can a whole-company AI ROI number never be honest?

Because a company is not a workflow. Spread AI across a dozen workflows and the gains and costs blur into a figure nobody can attribute or defend, so whatever you report is a story stitched from estimates rather than a measurement. ROI is only measurable where you can isolate a baseline and compare it after, which is one workflow at a time. The differentiated workflow you commissioned is the single place a clean before-and-after exists, so that is the only place an honest ROI number can come from.

Which metrics count as real ROI and which are vanity?

Hard dollars count: headcount avoided or redeployed, cycle time that converts to revenue or capacity, and the cost of errors and rework removed. Discount the rest: generic time-saved estimates, usage and seat counts, and "productivity" with no dollar attached. Those soft metrics are how vendors inflate ROI decks, because they are always large and always favorable. If a metric cannot be traced to money that shows up in the business, treat it as a vanity metric, useful for tracking adoption but not for proving the engagement paid for itself.

What if we cannot name the baseline cost yet?

Then you are not ready to commission, and that is a finding worth having rather than a failure. The inability to state what the workflow costs you today means it is not understood well enough to build against, and naming it is usually a few hours of honest counting, not a consulting project. Go count the hours, price the rework, and trace the lost speed to revenue. If the number is real and large, you have your case and your yardstick. If it stays vague, the engagement was never measurable, and you have saved yourself the spend.

When should we lock in the measurement?

Before the build, never after. The most common way operators fool themselves is measuring against a story reverse-engineered once the work has shipped, and a number assembled to justify a spend is a press release, not a measurement. Decide which hard-dollar figures you will track, record their baseline values, and agree on when you will re-measure, all as part of the scope. A baseline recorded at kickoff, when you had no reason to flatter it, is the only comparison anybody should believe later.

Can you name what the workflow costs you today?

Start with the AI Maturity Index. Ten minutes, no call, and it surfaces the one workflow worth measuring along with the baseline questions to ask of it, so you only commission what you can prove.