AI adoption framework
Most AI adoption frameworks are a diagram you read once and never open again. This one is a working document: five branches, checkboxes that mean something, and a progress gauge that moves as you tick them. Fill it in below — in your browser, with no account, and without anything leaving your machine.
Opens the real app with the template already loaded. No sign-up, no upload, no install.
The framework, in full
This is the actual template, not a summary of it — the five branches below are what opens when you click above. It is deliberately short. An adoption plan that takes a week to write gets written instead of executed.
1. Inventory — candidate tasks
List the tasks, not the tools. The filter is one question: is it repetitive and text-heavy? That pair is what today's models are reliably good at, and it rules out most of what gets proposed in an enthusiastic meeting. Three candidates is enough to start; a list of thirty is a way of avoiding the next step.
2. Score each task
- Value — hours saved per week. A number, not "significant".
- Risk if the AI is wrong — and specifically who checks. A task with no named checker is not low-risk, it is unmeasured.
- Data — is it allowed to leave the building? This is the question that quietly kills half the candidate list, and finding out at the scoring stage costs nothing. Finding out after the pilot costs the pilot.
3. Pilots — pick two, not ten
Each pilot carries three checkboxes, and they are the point of the whole document:
- Owner named
- Success metric agreed before starting
- Two-week review booked
Agreeing the metric afterwards is how every pilot succeeds and nothing changes. Booking the review in advance is how a pilot ends instead of quietly continuing forever.
4. Rules of the game
- What may never be pasted into an AI
- Who validates AI output before it ships
Two lines, written down, beat a policy document nobody reads. If you want the longer version later, the app also ships an AI usage policy template.
5. Training
- Everyone has written 10 real prompts
- A shared prompt library exists
Adoption fails at the individual level far more often than at the strategy level. Ten real prompts each — on their own work, not exercises — is the smallest thing that turns a rollout into a habit.
Fill it in here
Below is the real application, running in this page with the template already open. Click a node and type. Tab adds a child, Enter adds a sibling, and ticking a checkbox moves the progress gauge in the toolbar.
The AI adoption framework, editable — nothing is uploaded, no account
.tabtree file to your disk, and automatic backup to a folder. Those unlock with
the licence.
Using it in five minutes
- Fill the inventory first, alone. Three tasks, ten minutes. Bringing a blank framework to a meeting produces a discussion about the framework.
- Score them with the person who owns the data. The third line — is this data allowed to leave the building? — is usually not yours to answer.
- Pick two pilots and name owners in the room. If nobody will take one, that is the finding; write it down rather than assigning it to the group.
- Switch to the kanban view to run them week to week, or the Gantt view if the two-week reviews need dates. Both are views of this same document — nothing is converted, nothing is lost.
One more thing worth knowing: press F5 and the map becomes a presentation, one slide per branch. It is the same document again — useful when the pilot review needs a five-minute readout rather than a deck.
Why this is not a PDF
A framework you can't edit gets copied into a slide and stops being maintained the same week. This one is a document you keep: the pilots move across a kanban board, the checkboxes drive a progress gauge, and the whole thing lives in a single file on your own disk.
That last part is the actual design decision. TabTree is one HTML file — you double-click it and it runs, offline, with no account and no server. Your AI adoption plan names your processes, your data rules and probably your people; it is exactly the kind of document that should not be sitting in someone else's database. Nothing here is uploaded, and you can prove it by turning off your Wi-Fi before you start.
Free forever: all six views, all 87 templates, ten maps synced across your devices, and image, video and PDF export. Pro is $9/month for unlimited sync, AI credits and the Claude connector — 14 days free, no card.
Common questions
What is an AI adoption framework?
A structure for deciding where AI actually helps, instead of adopting it everywhere at once. A usable one answers four questions in order: which tasks are candidates, how each scores on value and risk, which two you pilot first, and what rules everyone follows. If it does not end in a named owner and a success metric, it is a reading exercise.
How many pilots should we run at once?
Two. Ten pilots means nobody owns any of them and none get reviewed. Two — each with an owner, a metric agreed beforehand and a review already in the calendar — is the smallest thing that produces a decision.
Is the template free?
Yes, and it opens in the demo above with nothing uploaded and no account. The demo holds back only the exports and the saving. The same is true of every other framework the app ships — the Business Model Canvas, Lean Canvas, SWOT, AARRR and the rest are all on the frameworks page. If you'd rather see the rest of the app first, the set-up guide covers the backup folder and sharing, and the XMind import page lets you test with a file of your own.