Time to live
How quickly a defined problem becomes something users can use.
Rubric
Time, money, effort, fit, lock-in.
We score whether this AI deserves your time, effort, and money to get a real problem live. If a tool cannot produce something users can touch, it does not get a polite maybe.
What we write towards
Five questions. The listing body should answer them even when a numeric score is omitted.
Money to useful
What you spend before the tool is useful. We write this in the body. It is not one of the four scores, so we do not invent a number.
Effort to own
How much work it takes to keep the product yours after the first ship.
Problem-fit
Whether this AI matches the problem you know, not a generic model demo.
Lock-in
How hard it is to leave, export, or keep control of the artefact.
Scores we can record
Optional, from 0 to 10, and only a direction. Leave a score off when the evidence is thin. The labels below are the ones used on listings.
Time to live
How quickly a defined problem becomes something users can touch, not a deck or a prompt log.
Effort to own
How much work it takes to keep the product yours after the first ship: brand, code, and judgement.
Problem-fit
Whether this AI matches the problem you know, rather than a generic demo of what the model can do.
Lock-in
How hard it is to leave, export, or keep control. High scores mean you stay free. Low scores mean the tool owns the artefact.
Verdicts
yes
Yes
This AI deserves time, effort, and money for the stated problem. It can help you get something live. Not a blanket endorsement of the vendor.
yes-with-caveats
Yes, with caveats
Useful only if your success criteria match a narrow lane. Read the not-for list before you spend a week inside it.
no
No
Wrong problem, too much lock-in, or it will not get you live. Listed so you can skip the demo.
What we refuse to score
- Star averages scraped from other directories.
- Vendor-claimed time-saved percentages with no artefact to inspect.
- Affiliate economics. Status is recorded. Rates are not invented.