How to Pick Your First AI Workflow

The first AI workflow you automate sets the tone for every one after it. Pick well and the business sees a concrete win, trust builds, and the second project is easy to fund. Pick badly and "AI" becomes the thing that didn't work.

Most businesses pick badly — not because the option was bad, but because they chose on the wrong axis. They pick the workflow that's most interesting, or the one a competitor mentioned, or the one with the flashiest demo. None of those predict whether the project pays off.

Here's the framework we use to score candidates. It's three factors, and you can run it on a whiteboard in twenty minutes.

Score every candidate on three factors

For each workflow you might automate, score it 1–5 on each of these.

Frequency — how often does this happen? A workflow that runs 50 times a day has 50 times the surface area for savings as one that runs once a week. Automating something rare almost never pays back the build cost. Score by volume: a few times a month is a 1, many times a day is a 5.

Pain — how much does it hurt right now? Pain shows up two ways: time and errors. A workflow can be painful because it eats hours, or because when it goes wrong it's expensive — a missed lead, a botched invoice, a compliance slip. Score the current pain honestly. "Mildly annoying" is a 2. "We've lost real money to this" is a 5.

Clarity — how well-defined are the rules? This is the factor businesses skip, and it's the one that decides whether the build is smooth or a swamp. A workflow with clear, statable rules — if the email contains an invoice, extract these fields, match to a PO — is buildable. A workflow that depends on unwritten judgment your best employee makes differently every time is not, at least not first. Score how cleanly you could write the rules down. Vague is a 1. Fully specifiable is a 5.

Add the three scores. A candidate at 13–15 is a strong first project. Below 9, leave it for later.

Why clarity is the tiebreaker

Frequency and pain tell you whether a workflow is worth automating. Clarity tells you whether you can automate it well, now.

A high-frequency, high-pain workflow with low clarity is a trap. It looks like the obvious win, so you commit to it — and then the build stalls because every step surfaces an exception nobody had written down. The project runs long, costs more, and ships something fragile.

The same workflow is often a great second project. Once you've built one clean automation, you understand your own process better, you've documented the exceptions, and the clarity score has quietly climbed. First projects should be high on all three. Save the messy-but-valuable ones for when you have a win behind you.

Workflows that usually score well first

In practice, a few categories tend to land in the 13–15 range for small businesses:

• Lead intake and routing — new inquiry comes in, gets enriched, gets a drafted response, gets logged. Frequent, painful when slow, and the rules are clear.
• Appointment and reminder flows — scheduling, confirmations, no-show follow-up. High frequency, clear rules.
• Document processing — invoices, intake forms, contracts. The extraction rules are specifiable, and errors are expensive.
• Routine reporting — the weekly numbers someone assembles by hand. Frequent, low-clarity-risk, immediately visible savings.

These aren't the only good candidates. They just tend to score well because they're frequent, genuinely painful, and rule-bound — which is the whole point of the framework.

The discipline part

The hard part isn't the scoring. It's accepting the result.

The framework will sometimes tell you the unglamorous workflow — the reminders, the reporting — is the right first project, while the exciting one isn't. That's the framework working. The goal of the first project is not to be impressive. It's to produce a concrete, defensible win that makes the second project easy to say yes to.

Resist two temptations. Don't automate the rare-but-painful thing first — the math isn't there. And don't automate the high-pain, low-clarity thing first — you'll spend the build discovering rules you should have written down on a whiteboard.

Pick the boring, frequent, well-defined workflow. Ship it. Let the result argue for the next one.

If you want a second opinion on which of your candidates scores highest, that's exactly what a discovery call is for — bring two or three, and we'll run them through the framework together.