How much does an AI automation consultant cost?

Discovery is $2,500, fixed, and runs one to two weeks. It is refunded in full if the arithmetic at the end says do not build — which is the only thing that makes a paid decision stage worth its own line item.

What if the answer is that we should not automate it?

That is a real outcome and it happens. The refund exists precisely so that saying so costs us rather than you; a consultant who only ever recommends building is not measuring anything.

What do we actually get at the end?

The process drawn as it really runs, a measured weekly cost for it, and every automation candidate scored against that cost. Not a deck — the arithmetic somebody else can check.

Do you build it as well, or only advise?

Both, and they are separate decisions. Implementation starts from $5,000 and puts one to three processes in production in two to four weeks, and it is a choice you make after the numbers rather than before.

All service tracks

AI automation consulting

Most automation advice comes from people paid to build. We measure the process first, name the steps that should be removed rather than automated, and return the fee when the numbers say do not build.

Fee: $2,500
Length: 1–2wks

What the engagement covers

Scope of work4

01Measure the process before quoting a buildTwo weeks on the real thing: where the hours go, how many handoffs a case makes, how often it goes wrong. A quote written before that is a guess with a number on it.
02Name the steps that should not existAutomating a step nobody should be doing makes it permanent. The audit separates work that is genuinely required from work that survives because it always has.
03Say when an off-the-shelf tool already does itAn app is cheaper than a project. Where a standard tool covers the flow we say so, and we say exactly where it stops being enough.
04Put the candidates in orderWhich process goes first, what it depends on, and what the second one costs once the first exists. Sequence is most of the saving.

What you get out of it

A measured weekly cost for the process, not an estimate
The fee back in full when the honest answer is do not build
Off-the-shelf options named rather than quietly skipped
A plan specific enough for your own team to execute without us

Where this sits7

Visibility AnalyzerfreeFree
Free AI DiagnosticfreeFree
Managed Automation$750/mo
Visibility Upkeep$900/mo
Visibility Sprint$1,500
Discovery Sprintyou are here$2,500
Implementation Sprintfrom $5,000

How it can end1–2

1–2

weeks of measuring

the arithmetic

it adds up · we build
it does not · money back

Both exits are real. Only one of them is paid for.

Two ways to get that number

You can measure the process yourself; the tools cost nothing. The difference is who pays for the inconvenient answer.

Do it yourself

Carry a stopwatch through one process for a week: how many steps, how many hand-offs, how often a request comes back
Turn that into a weekly cost and set it against the price of automating the same step
List separately the steps that should not exist at all — automating one of those fixes it in place forever
Then answer honestly whether to build, knowing a vendor does not answer that question for free

Discovery Sprint

Hand it to us

Two weeks on the same process, measured by somebody who is not paid for the answer being yes
The whole fee back if the arithmetic says do not build: that answer costs us, not you
The ranked list of candidates is yours either way, including if you go on without us

Not for you if

Where this is the wrong buy3

The decision is already made and you need a document that agrees with it. We measure to answer, not to endorse.
You need the whole company mapped. Two weeks buys one process on purpose: a map of everything at once is accurate nowhere.
It has to be working by the end of the month. What comes out of this is numbers and an order, not a running process. Building is the next rung.

Asked before anyone buys this

How much does an AI automation consultant cost?: Discovery is $2,500, fixed, and runs one to two weeks. It is refunded in full if the arithmetic at the end says do not build — which is the only thing that makes a paid decision stage worth its own line item.
What if the answer is that we should not automate it?: That is a real outcome and it happens. The refund exists precisely so that saying so costs us rather than you; a consultant who only ever recommends building is not measuring anything.
What do we actually get at the end?: The process drawn as it really runs, a measured weekly cost for it, and every automation candidate scored against that cost. Not a deck — the arithmetic somebody else can check.
Do you build it as well, or only advise?: Both, and they are separate decisions. Implementation starts from $5,000 and puts one to three processes in production in two to four weeks, and it is a choice you make after the numbers rather than before.

Next step

Put a number on the process first

Two weeks on one process: the map, a measured weekly cost, and every candidate scored before anything is built. If the arithmetic says do not build, we say so and the fee comes back.

Put a time in the calendar

Thirty minutes, free. The sprint is what the call is about.

Fee: $2,500
Length: 1-2 weeks

Refunded in full if we conclude you should not build.

Or start with the free 15-minute diagnostic