Skip to content

AI Audit & Adoption

An AI audit that ends with tools in use.

Find where AI changes how your organization works, and start building the first tools straight away.

Solutions · Applied AI

Plenty of organizations have AI licences now. Far fewer have changed how work gets done.

We learn how your team operates, find where AI changes what’s possible, and build the first tools with you.

Where AI adoption often stands

The licences

Licences for ChatGPT, Claude or Copilot have been bought. There’s been a training session, maybe an acceptable-use policy.

The use

A few enthusiasts use it every day. Others use it to tidy emails and summarize documents. Leadership suspects there’s much more available, but nobody has had the time to find it, and the people who know the work best are too busy doing it.

The gap

The tools are rarely the problem; they’re good and getting better. What’s missing is a close look at the work itself, and where AI would change it.

Why a strategy deck isn’t enough

What should an AI audit end with?

  • A strategy deck

    A list of possibilities doesn’t change anyone’s Tuesday.

    A lot of AI strategy work ends in a document: a maturity assessment, a list of use cases, a roadmap. It’s useful to have, and it tends to sit unread.

  • Working tools

    Working tools show what’s real.

    The audit still maps the opportunities, and it ends with working tools in the hands of the people they’re for. They show what AI can do with your data and your processes, and they tell both of us far more about what to do next.

How we run it

  1. Discover

    We spend time with the people doing the work, across the teams that matter most, and watch processes as they run. Where does the work repeat, or stall waiting on a handoff? Where do people copy information from one place to another? Which questions go unanswered because answering them takes a day? What do the best people on the team know that others don’t? Our AI discovery interviewer, built for this stage, helps us hear from more of the team.

  2. Map

    Each opportunity is sized by value, effort and risk. Some of the most valuable ideas carry real risks around privacy, accuracy or reputation, and we name those plainly.
    • Automate
    • Upgrade
    • Not possible before

  3. Architect

    We look at the platforms you already have and recommend how AI should connect to them: which tools, which data, which permissions, and what should stay out of reach.

    Often the right answer uses what you’re already paying for.

  4. Build

    We pick one or two opportunities with a clear owner and a visible payoff and make them real, in use before the engagement ends. That might be a custom skill that teaches Claude your proposal process, an automation that assembles a weekly report, or a small tool that answers questions from your own documents.

We sort what we find into three kinds

  1. Work to automate

    The output stays the same. It arrives faster and more reliably, and nobody assembles it by hand.

  2. Work to upgrade

    Output that’s clearly better: a summary that explains instead of lists, a document clients can explore.

  3. Work that wasn’t possible before

    Jobs too big to do by hand until now, like a weekly read of every customer comment, or advice clients can get at any hour.

Adoption is the work

AI rollouts often stall after the launch email.

  • With each teamWe sit with teams while they use the tools on their own work, fix what gets in the way, and adjust the tools based on what we see.
  • On the platformWe set up the platform properly, with access, permissions and data handling agreed with IT and leadership, so people aren’t guessing what they’re allowed to do.
  • Carrying it forwardWe help identify the people inside the organization who will lead it day to day, and we stay involved for what comes next.
People try it once, get a mediocre result, and go back to how they worked before. Adoption takes time with real tasks.

What we won’t do

We won’t recommend AI where a simpler fix works.

Sometimes the answer is a better form, a cleaner process, or one integration between two systems.

We won’t oversell the opportunity.

If it’s smaller than you hoped, we’ll say so.

When the audit ends

  1. A first result people can point to

    Something the team uses every week, and a clear case for what to build next.

  2. Licences that earn their keep

    People know how to use the tools on their own work, and what they’re allowed to do with them.

  3. A different conversation about AI

    With a working tool on the table, the question could move from whether to where next.

  4. Your own AI champions

    People inside who know the tools and the plan.

What comes after

Later tools can build on the data and access already set up.

The opportunity map becomes the plan:

  1. 01picking the next opportunity
  2. 02building the next tool
  3. 03reusing the connections and permissions already in place

  • Some clients stop once the first tools are in use and carry on themselves.

  • Others keep building with us.

Related

The expertise behind it

All expertise