Five questions.Forty-five seconds.An honest read.
Answer for how things really run today, not how the process document says they do. At the end you get a plain read on where AI fits in your business — including the answer “not yet”, if that is the true one.
- 45 seconds
- five questions
- no obligation
- no email to see it
- nothing sent unless you ask
Nothing here is stored or scored against anyone else, and nothing reaches us unless you ask it to. Answer, read your result, close the tab and it is gone. If you would rather keep it, the result page can email it to you — and that is the only way any of it reaches us.
The five questions, printed in full.
No trick questions and no trap answers. Every one of them is about how your work runs today, and none of them asks anything about you — not your name, not your company, not your numbers.
- Where does your most important workflow live today?From paper and signatures, through spreadsheets and chat threads, to connected systems end to end. This is the question that decides whether AI is even the right next move.
- When you need a number, what happens?Sales this week, stock, workload. Whether someone is asked and you wait, whether a spreadsheet gets assembled, or whether you open something live. It tells us what your data is actually like.
- How much of the day goes to repetitive work?Sorting, copying, chasing, re-typing. This is the one that decides whether an agent has anything worth taking off your team's hands yet.
- Has your team used AI tools in their work?From not yet, through one or two people experimenting quietly, to everyone using them under written rules. There is no wrong answer here, but there is a revealing one.
- How clear is your starting point?No idea where to begin, too many ideas and no way to choose, or a plan already in motion. Too many ideas and too few are different problems, and they need different first steps.
A read, a first step, and permission to stop there.
The result appears on screen the moment you answer the last question. There is no gate, no “enter your email to see your score”, and no report that arrives later.
Where you stand
Your answers place you in one of three honest bands, named plainly rather than scored out of a hundred. A low band is not a failure — it is a shorter, smaller first project than a high one.
The sensible first step
A short written recommendation in the same plain language as the rest of this site: what to do first, and why that rather than the thing you were probably about to take on. It can say the first step is not AI at all — that it is getting one workflow or one number into shape, and that the rest can wait.
Where to read more
Links to the services that match your answers — and, deliberately, not to the other ones. Then the check ends. Booking a call is offered once, on the same screen, and ignoring it commits you to nothing.
Nothing. That is the whole answer.
It is easy to build one of these as a form wearing a quiz costume: five easy questions, then your email address before the result. This one has nowhere to send anything, because the check is a small piece of code running inside your own browser.
- no sendYour answers are never transmitted. They are held in the page's memory while you answer, and used once to pick the band.
- no storeNothing is written to your device and nothing is kept after you close the tab. Reload the page and the check starts from the first question again.
- no askNo email, no name, no company. Nothing has to be filled in to see your result. Once it is on screen the page offers to email you a copy — that is the only field anywhere in the check, it appears after the answer rather than before it, and ignoring it costs you nothing.
- no trailNothing about your answers reaches us — not what you picked, not which band you landed in, not in aggregate. The page request itself appears in the host's standard logs, exactly like every other page on the internet, and that is the whole of what we could ever see.
- check itYou do not have to take our word for any of that. Open your browser's network tab and run the check — you will watch it send nothing. Or switch your connection off entirely and take it offline.
The chat assistant in the corner is a different thing entirely: that one does send what you type to a model, which is why it has its own section on the privacy page. The rest of what this site does with data is written down the same way on the trust page — including the parts we have to name rather than promise.
Four honest limits on a 45-second check.
Forty-five seconds gives a direction, not a diagnosis. Here is what this thing cannot do, so the result gets weighed correctly when you show it to someone.
-
Not a lead form
There is nothing to submit and nobody is notified that you took it. If you want us to see your answers, you have to send them yourself.
-
Not a benchmark
The band does not compare you to an industry average, because we do not have one and we are not going to invent one. It reads your five answers and nothing else.
-
Not a scope
Five questions cannot size work. What it involves depends on your data, your systems, your people and your exposure — none of which fit in a multiple choice answer.
-
Not the map
The map is the first stage of work, where the real process is watched and drawn as it actually runs. This is the quick check that tells you whether it is time for that yet.
Disagree with your result? Come and argue it.
Bring the band the check gave you and the one workflow sitting behind it. Thirty minutes, and you leave with the first step named — or with the same “not yet”, and the reason it applies to your business rather than to a set of five answers.
Ask us to call you backwith FuseLoom · [email protected] · or send the result by email and get the read back in writing instead