Eight quick questions on the things that actually predict whether an AI build will land: a clear process, reachable data, an owner, and buy-in. You'll get a readiness score and a concrete next step. No email required.
1. How clearly can you name the first process you'd want AI to handle? No specific process in mind yet A vague area, but nothing concrete One process, roughly defined A specific, named process with clear inputs and outputs
2. How often does that task actually happen? Rarely, or it's a one-off A few times a month Most days Many times a day, across the team
3. Is the process written down anywhere? It lives in one person's head Loosely, in scattered notes Mostly documented Fully documented and repeatable, step by step
4. How accessible is the data this would need? Locked in PDFs, inboxes, or someone's memory In a few systems, hard to pull together In systems we can export from In connected tools with APIs we can reach
5. Are your core tools connected to each other? Everything is manual copy-paste A couple of integrations exist Most tools talk to each other A connected stack with automation already running
6. Who would own this internally? Nobody has the time Someone, in theory, with no bandwidth A clear owner with some bandwidth An owner who is ready and accountable
7. Where does leadership stand on AI investment? Skeptical or unaware Curious but uncommitted Supportive, exploring budget Committed, with budget allocated
8. Could you review AI output before it acts on something important? No review process at all We'd check it ad hoc We have a clear review step in mind We already run human-in-the-loop checks elsewhere
Ninsei Labs