Is this a tool for saying no to training?
No. It is a tool for saying yes on purpose. When the diagnosis confirms a real knowledge or skill gap, it helps you design the right training, and surround it with the supports that make it stick. It simply stops training being the default answer to every request.
Doesn’t diagnosing away training make L&D smaller?
The opposite. It gives L&D a value number instead of completion rates, proven by what you prevent. The function grows into a performance partner, not a course factory.
How long does a diagnosis take?
Three to five minutes. It records the intake conversation, adds whatever data you already have, and returns the gap, the causes, the next steps, the open questions, and a confidence score.
Will it replace our L&D team or our consultants?
The opposite. It turns your advisors into performance consultants and carries the function from Order Taker to Performance Specialist. It is a capability jump, not a headcount cut, and it means you no longer buy in an external diagnosis for every request.
How reliable are the answers?
Every cause is tagged with how far the evidence reaches, hypothesis, substantiated, or triangulated, and the diagnosis carries a single confidence score with exactly what would raise it. It marks its own uncertainty rather than hiding it.
What does it cost, and are there usage bills?
One predictable investment. No meter, no per-diagnosis charge, no surprise bill. Every license includes unlimited diagnoses within fair use, a self-study program, and in-product guidance. The small-department tier is €19,000 / $19,000 / £16,900 a year.
What is free?
The Performance Dividend Calculator, no login, always free. Run it to see the cost of a single request and a standing dividend number for your function.
Is our data safe?
Performance Detective AI is built for reliable knowledge handling, source provenance, isolated per-customer storage, and clear data governance, with enterprise controls (SSO, retention policy) on higher tiers.
We already run intake conversations. What does this add?
Intake captures the request. This tests it. The same conversation is carried through a structured diagnosis, so what comes back is a cause, a recommendation and the evidence behind them, rather than the request restated. The board report is written for you.
Can AI really diagnose our context?
It shows its reasoning as it works. Every cause is tagged with how far the evidence reaches, and the diagnosis carries a confidence score with exactly what would raise it. You see the argument, not just an answer, so you can challenge it wherever your context differs.
Where do the numbers on this page come from?
The worked examples are illustrative simulations built from real situations, not named clients. Your own numbers come from the calculator, using your participants, hours and costs. Nothing here is a claim about your organization until you run a diagnosis on a real request.
Who is behind it?
Performance Detective AI is a Tulser product, built on fifty years of human-performance practice by the Tulser team and engineered into an everyday diagnostic tool.