Most companies cannot answer three basic questions about AI inside their own walls: which tools are being used, on what, and with which data. The diagnostic answers them in one to two weeks, function by function, and the result is the map everything else is built on. It is step one, and the 30-minute call that scopes it is free.
Short interviews with a handful of people per function, plus an internal survey that everyone can answer in a few minutes. The survey is anonymous and we say so up front, because the single most useful finding is the usage people would not put their name to. We need no access to your systems, no integration and no IT project: the whole thing runs on conversations and a form.
A map by function showing current use, the risks ranked rather than listed, and a proposed scope: which functions to start with, how long their tracks should be, and which use cases the lessons should be written on. You also get the raw material for the acceptable use policy, because the data categories that need a rule are the ones the survey surfaced. If you stop there and do the rest yourself, the map is still yours.
This matters more than it sounds. A diagnostic that people read as a hunt for rule-breakers returns nothing useful, because the unofficial use is exactly what stays hidden. We report at the level of the function, never the individual, and we say that in the survey itself. The companies that get the most out of this are the ones whose management says out loud, before it starts, that nothing found here will be held against anyone.
Companies often start with the training, pick a generic catalogue, and discover three months later that half the content covered tools nobody uses while the real exposure was in a function nobody thought about. Starting with the measurement costs two weeks and changes what the rest of the programme is. It is also the cheapest step to run, which is why it is the one we give away the scoping call for.
Diagnostic, acceptable use policy, interactive tracks by function, adoption follow-up. The four steps work together, and the free 30-minute call scopes all of them.
See the four stepsFind out how your teams actually use AI
No. The call is where we go through how your teams use AI today, what is missing to frame it, and what the first quarter would look like. You leave with a scope and a quote. The diagnostic is the one-to-two-week piece of work that follows if you decide to run it.
Two or three per function in scope is usually enough for the interviews, because we are looking for the shape of the usage rather than a statistically representative sample. The survey goes to everyone, and a response rate above half is what makes the map trustworthy.
No. We ask people what they do; we do not read their accounts, their files or their prompt history. If you want a technical inventory of which licences are active, that is a job for your IT team and it complements this rather than replacing it.
That is the normal result and it is the useful one. The companies that discover nothing are usually the ones whose survey nobody answered honestly. Being behind on written rules while your teams are already using the tools daily is the ordinary starting position, and it is the position the four-step approach was designed for.
We go through how your teams use AI today, what is missing to frame it, and what the first quarter would look like. You leave with enough to decide, whether you work with us or not.