AI is cheap. Bad decisions are not.

Everyone audits your tools.
We audit your roles.

The Role Audit maps every role in your business against what AI can genuinely take on, before you spend a dollar on tools. Fixed fees, published up front. Start with a $300 Workbench Session, or go straight to the Audit.

80%+

of AI projects fail, not on the tech

Source: RAND Corporation, 2024

~50%

of paid software licenses sit unused

Source: Zylo SaaS Management Index, 2025

Why now

Three shifts arrived while you were busy running the business.

Your buyers stopped clicking.

When an AI summary appears in search results, people click through to a website 8 percent of the time. Without one, 15 percent. Gartner projects traditional search volume down 25 percent by 2026 as buyers ask AI assistants instead. Being found now means being the answer, not the link.

Source: Pew Research Center, 2025. Gartner, 2024.

Your software bill grew in the dark.

Roughly half of paid software licenses sit unused. Stacks pile up quietly: a pilot that fizzled, two tools doing one job, seats for people who left. Nobody owns the bill, so nobody reads it. The money leaks every month either way.

Source: Zylo SaaS Management Index, 2025.

AI spending is outrunning AI results.

More than 80 percent of AI projects fail, twice the rate of ordinary IT projects. The failures are rarely technical. They come from buying tools before mapping the work. The fix is a map, not another subscription.

Source: RAND Corporation, 2024.

You do not need to become an expert in any of this. You need someone accountable for it. That is the job. It starts small.

The real problem

Every AI audit examines your tools. Almost none examine what your people actually do.

That's the whole ballgame. AI doesn't replace tools, it changes tasks. And tasks live inside roles. If you don't know which tasks make up which roles in your business, every AI purchase is a guess wearing a scorecard. Tools change every quarter. Roles change on the timescale of your business. We audit the durable thing.

AI is cheap. Bad decisions are not.

What lands on your desk

Findings you can hold, not slides you can nod at.

The Role Map

Every role, broken into tasks. Every task, scored.

The Role Map takes each role in your business and breaks it into the work it actually contains. Every task gets a verdict: AI takes it, AI assists, or it stays human. High-judgment work staying human is a finding, not a failure.

Role: Client Coordinator 6 tasks JRC Reconciling vendor invoices 252 AI fit Drafting renewal proposals 443 Assist Approving client refunds 325 Keep human Escalating at-risk accounts 535 Keep human J judgment · R repetition · C consequence of error, scored 1–5. Sample format.
The Tool Bill Audit

Your software bill, with a verdict on every line.

Every subscription mapped against the roles that actually use it. Keep, cut, or downgrade, each with a dollar figure attached. We don't sell software and we don't take referral fees, so a smaller bill is a report we can actually give you. It is the most common first finding.

Tool Bill Audit 14 subscriptions CRM · used by 9 of 12 seats Keep Social scheduler · last login 4 months ago Cut Project tracker · 12 seats, 5 active Downgrade Monthly change if adopted −$740/mo Sample format. Figures come from your stack.
Build or keep

Some of what you rent, you should own.

Where a subscription is doing simple work at a rented price, we can build you a replacement you own outright. We document it, hand you the keys, and new features are a quote away. Where the subscription earns its fee, we tell you to keep it. The audit tells you which is which.

Where the math usually wins:

  • Pipeline tracking rebuilt inside your email and calendar, replacing per-seat CRM licenses.
  • Brand-trained drafting inside the document suite you already pay for, replacing standalone AI writing tools.
  • Content scheduling moved to the platforms' own free native tools, replacing third-party schedulers.
  • Thin internal tools you own outright, replacing rented interfaces over your own data.
Rented $189/mo Renews every year 5% of features used Price set by vendor Owned Built once Yours outright Documented and handed off New features quoted, not forced Only where the math wins. A system holding years of client records usually stays rented, and we will say so.
How the Role Audit works

Every task in every role, scored on three axes.

Judgment required

Not "could AI do this in a demo." How much context, taste, and consequence-awareness does the task carry in your business? High-judgment work stays human. That's a finding, not a failure.

How often it repeats

AI compounds on repetition. A task done once a quarter rarely justifies any tooling. A task done forty times a week is where the return lives. We count, so you're not guessing.

Consequence of error

A misformatted internal report costs a shrug. A wrong number on a client invoice costs a relationship. Error consequence, not technical capability, decides where AI runs unsupervised.

Score every task on those three axes and the AI opportunities stop being a matter of opinion. The map draws itself.

How it works

Four moves. Repeatable every time.

1

Fit call

30 minutes. We find out if the Audit is the right next move for your business.

2

Audit

Two to three weeks. We map every role, the actual work rather than the job titles, interview your team, and score every task for AI fit.

3

Findings

A written report and live walkthrough, plus one process already automated and running. You get a prioritized roadmap you can act on with or without us.

4

Build (optional)

If you want us to implement, we start with the highest-return item and work down the list. Some of what you rent, you should own. Where a subscription is doing simple work at a rented price, we can build you a replacement you own outright. The audit tells you which is which.

The product

The Role Audit

A fixed-scope engagement that answers one question: where does AI genuinely belong in your business, role by role, task by task, and in what order?

Not a discovery call. Not a proposal. A real deliverable with real findings, from two people who stay with you the entire time. And because we don't sell software or take referral fees, "cancel three subscriptions and change nothing else" is a report we can actually give you.

Book a Fit Call
Scope Fixed. No scope creep, no surprise invoices.
Timeline Two to three weeks from kickoff to delivery.
Your time A fit call, a few short interviews, and access to how things run today. We handle the mapping. Your team keeps working.
Deliverable Four things: the Role Map, the Tool Bill Audit, Fit Verdicts ranked by payoff, and a 90-day sequence, delivered in a live walkthrough.
Quick win The clearest, safest opportunity we find gets built and running before the engagement ends.
Leave-behind Your team keeps a briefing trainer: a simple tool that turns a rough request into a well-structured AI prompt.
Credit The full audit fee applies against implementation if you continue with us. The findings are yours either way.
Team The same two people from start to finish. You are never handed off.
Next step Optional. Implement with us, take the roadmap elsewhere, or both.
Pricing

Fixed fee. Published, because you shouldn't need a sales call to learn a price.

Rung one: the Workbench Session, $300 flat. One person, one live video call, 60 to 90 minutes. We build your first AI setup with you, in your own tools, on a real task from your week. It is running before we hang up. See the Workbench Session

Compact

$4,500

Teams under 10 people

  • Up to 5 roles mapped
  • Owner + 2 operator interviews
  • All four deliverables
  • Two weeks, kickoff to walkthrough
Book a Fit Call

Extended

$12,500

31–75 people

  • Departmental role maps
  • Up to 10 operator interviews
  • Light compliance pass for regulated industries
  • Three weeks, kickoff to walkthrough
Book a Fit Call

The ladder has three rungs. A Workbench Session for one person. The Role Audit for a whole team. Then, if you want us to build what the Audit found, the build work. Each step is optional and each step earns the next.

At $7,500, the audit breaks even on $625/month of tool cuts, and the typical stack carries 20–40% overlap or underuse.

The full audit fee credits against implementation if you continue with us. The findings are yours either way.

Founding client rate: $5,500 for our first three Standard engagements, in exchange for a documented, publishable results number. 3 of 3 remaining.

Where the money is

Six leaks that quietly drain a busy operation.

Work that lives in one person's head

A process only one person knows how to run. When they are out or they leave, it stalls. Anything you can write down, you can make repeatable.

The same data, entered twice

Staff keying the same information into two or three systems because the tools do not talk. Hours a week, and every extra keystroke is a chance for an error.

Tools you pay for that nobody uses

Overlapping software, a license bought for a pilot that fizzled, three tools doing one job. The fix here is usually fewer tools, not another purchase.

Reports rebuilt by hand every week

Someone pulling the same numbers from the same places into the same spreadsheet, every week. A rules-based task quietly eating senior time.

Requests that pile up in an inbox

Leads, tickets, or internal requests sorted and routed by hand. Slow answers, dropped items, and no clear view of what is still waiting.

Drafts that start from zero every time

Proposals, briefs, and summaries written from scratch when most of the content repeats. The pattern is already there. It just has not been captured.

We start by finding the one that's costing you most, not by selling you all six.

Run the free Role Check

Two free tools, two doors. Curious about your own job? Try Staves, the 60 second job split. Curious about your team? The Role Check above.

Why Cooperage

A partner earns the next step. So we made the steps small.

Three hundred dollars shows you how we work. The Workbench Session helps one person. The Role Audit maps a whole team. The Audit fee credits toward the build work if you continue. The same two people from first email to final build, and the findings are yours at every rung whether you continue or not. Partnership is a structure, not a slogan.

Specialists, not a platform

We don't sell software. We don't have a preferred vendor. Every recommendation starts with your operation, not with a tool we're incentivized to place.

Outcomes first, not tech first

Our measure is the return on your investment, not the sophistication of the implementation. Boring automation that saves $40k is better than impressive AI that saves nothing.

The promise: a person stays in the loop

Every engagement is run by a strategist and a senior engineer. Not a junior team, not a subcontractor. The same two people who scope the work are the ones who deliver it.

Common questions

Things people ask before they book.

Do I need to know anything about AI before the Audit?

No. The Audit is designed for business owners, not engineers. You bring the context about how your operation runs. We bring the expertise about where AI can help.

What size businesses do you work with?

From a growing company to a single team or business unit inside a larger company. What matters is a defined operation with repeatable work and real money moving through it, not headcount.

What's the difference between the fit call and the Audit?

The fit call is free and takes 30 minutes. It's a conversation to make sure the Audit is the right next move for your business. The Audit is the paid engagement where the actual mapping and findings happen.

What does the Role Audit cost?

Fixed fee, published right here: $4,500 for teams under 10 people, $7,500 for 10–30, and $12,500 for 31–75. No hourly billing, no scope creep, no surprise invoices. The full fee credits against implementation if you continue with us, and the findings are yours either way.

How is a Role Audit different from an AI readiness audit?

A readiness audit examines your tools, data, and workflows. The Role Audit examines what your people actually do: every role broken into its real tasks, each scored for judgment required, repetition, and consequence of error. Tools change every quarter; roles change on the timescale of your business. Auditing roles produces answers that stay true after the next model release.

What is the quick win?

During the audit we build the single clearest, safest automation we find, so it is running before we deliver the findings. It is chosen for confidence and low risk, never client-facing or compliance-sensitive on a first pass. If nothing meets that bar, we say so in the findings and explain why.

What is the briefing trainer?

A small tool your team keeps after the audit. Paste in a rough request and it returns a well-structured prompt, built around how your business actually talks about its work. Most AI underperformance traces back to vague instructions. The trainer fixes that at the source, for everyone, every day.

Can you replace software we pay for with something we own?

Sometimes, and the audit tells you when. A subscription doing simple work at a rented price is a candidate: basic pipelines, schedulers, thin interfaces over your own data. A system holding years of client records usually is not, and we will say so. Every build recommendation comes with the ownership cost included: what it takes to run, maintain, and fix. If the math favors keeping the subscription, that is the finding.

Do you implement after the Audit?

We can, but it's not required. Some clients take the roadmap and run with it internally or with another team. Others hire us to build, and the full audit fee credits toward that work. Either way, you own the findings.

We already use AI or have tools in place. Can you still help?

Yes. A stack you already paid for is usually where the most waste hides and the fastest wins sit. We map what you have, keep what earns, cut or consolidate what does not, and add only where the return is clear. The answer is sometimes less, not more.

Will my team actually use what you recommend?

That is the point. Most AI fails for human reasons, not technical ones. A tool nobody adopts, a workflow that ignores how people actually work. We map the work the way your team really does it, across the people and departments involved, so the roadmap fits the operation you have. Adoption is built into the findings, not bolted on afterward.

How is this different from hiring a consultant?

Most consulting engagements are open-ended and measured in hours. This is fixed scope, fixed deliverable, fixed timeline. You know exactly what you're getting before you start.

What industries do you focus on?

The Audit fits any team with repeatable processes and people doing manual work. That covers professional services like law firms, agencies, financial advisors, and consultancies. It also covers a single business unit or leadership team inside a larger company. The common thread is a defined operation worth mapping, not a particular industry or company size.

What is the Workbench Session?

One person, one live video call, 60 to 90 minutes, $300. You bring one real task from your week. We build the AI setup for it with you, in your own tools, and it is running before we hang up. You get a one page recap after, and a ten minute check-in two weeks later. It is the smallest way to start with us.

Should I book a Workbench Session or a Role Audit?

Simple rule. Just you? Book a Workbench Session. Your whole team? That is the Role Audit. A session helps one person with one task. The Audit maps every role on a team and puts dollar figures on the map.

Do you do SEO or AI search optimization?

No. We are not a marketing agency and we do not sell visibility services. Buyers moving to AI answers is real, and when an Audit shows it matters for your company, the finding says so and we point you to a specialist. We stay vendor neutral so every verdict you get from us stays honest.

Find out where your operation is leaking.

30 minutes. No pitch. Just a clear-eyed look at whether the Audit makes sense for your business right now.

Want to start smaller? Book a $300 Workbench Session. One person, one call, working help by the end.

Or email us directly: hi@cooperage.works

Or pick a time right now. The calendar below books directly with us.

Open the booking page in a new tab