For humans · Act
Prove the action moved the number.
Every action is tracked against the metric it was meant to move, and impact is verified by the same pipeline that calculates your actuals. Nobody marks their own homework.
Prove the action moved the metric. Then the model learns.
Nobody marks their own homework. Verified against the metric
Actions are tracked against the metric they were meant to move and the impact is measured by the same pipeline as your actuals.
Learns from behaviour change. Not query counts
What worked strengthens the causal model; what did not gets discounted. Recommendations sharpen over time.
Your day starts with what needs you. The daily briefing
What moved, why, and whether yesterday's action worked, personalised to what you own.
Nobody marks their own homework
Every vendor promises insights to action; almost none can tell you whether the action worked, because the only record is what people said in the retro. Here, actions link to metrics, and when the metric moves, the action's impact is measured against it by the same pipeline that calculates your actuals. The verification is observed, not self-reported, which is what makes it worth anything.
Fix checkout flow
Verified · +£32kLinked to Revenue · measured by the actuals pipeline
Pause underperforming ad sets
Measuring…Most tools learn from queries. This one learns from behaviour change.
KPI Tree tracks the actions people actually took, not what they said they did, and which of them moved their metric. That verified outcome history tunes the causal model, re-weighting the drivers that prove out when someone owns them, and sharpening every future recommendation. It is also the asset no ad-hoc AI analysis can build: a stateless agent forgets each answer as the chat closes, while the tree accumulates proof of which levers move which numbers. The longer you run the loop, the smarter it gets.
Driver
Conversion rate
Outcome · 58% contribution
Revenue
Targets that track themselves
Metric goals auto-complete when the metric hits the target, and progress bars show how close every goal is in real time. Nobody updates a status field, because the data is the status, and a status nobody can fudge is the quiet foundation of the whole verification story.
Your day starts with what needs you
Turn on the daily briefing and every morning each person gets a personalised read: what moved, what needs attention, what is overdue, generated from the same causal model, ownership and outcome history that powers everything else. The attention strip puts the items needing action first. This is what a personalised action plan looks like in practice: not a dashboard to interpret, a list with your name on it.
KPI Tree app · 08:00
Good morning, Sarah. Tuesday briefing: revenue is tracking £31.5k/day behind target, down 13.1% MTD. Three things need you.
Engagement measured in actions taken, not queries run
The engagement heatmap shows who is viewing which metrics, who is taking action, and who needs a nudge, feeding straight back into the accountability loop. Query counts flatter tools; actions taken and impact verified measure whether behaviour is actually changing. Data culture stops being a slogan and becomes something you can see and manage.
The full loop
Metric moves. The named Accountable owner is notified with the driver context attached. The action is tracked against the metric it was meant to move. The impact is verified. Then the model learns and the loop starts again. Every vendor claims this sentence; this is the mechanism.
Understanding changes behaviour. Dashboards don't.
People change when they see the system, see their place in it, and see what moves when they do, and when the number has their name on it. That is the discipline this product exists for: connecting data intelligence to human behaviour change.