Delivery constraint diagnostic
Based on DORA 2025 research (Google Cloud)
Find where your delivery waits — before you point AI agents at it.
16 questions, about 5 minutes. You leave with your team’s probable constraint, where AI will actually speed up your delivery, and your five DORA metrics against the 4,867 respondents of the 2025 survey. Nothing is kept on the server.
- 16 questions
- About 5 minutes
- No sign-up
- Nothing kept on the server
- Independent tool
In figures
16
questions, one per screen, ‘I don’t know’ always available
5 min
or so; you can resume later in the same browser
4,867
DORA 2025 survey respondents your five metrics are compared with
How it works
Answer
Sixteen questions, one per screen. DORA questions come from the survey (CC BY 4.0); questions written by JDUCRAFT are labelled as such. You can resume later: your answers are saved in your browser.
Read
The probable constraint, the signals that confirm or contradict it, your metrics against the 2025 respondents, three first moves. A printable report, a link to share, no account.
Talk it through, if you like
A free 30-minute reading call: we go through your answers, name the constraint, you leave with two leads. Then, if the question deserves it, a three-consecutive-day audit at a fixed price.
What you get
Your probable constraint, and how confident we are
The model says what it knows and what it does not, with every answer that weighs for or against. A hypothesis to check, not a verdict.
Your five DORA metrics against 4,867 respondents from 2025
Lead time, deployment frequency, recovery, change failure rate, rework: the share doing better, the same and worse, with the report’s figure and page.
Where AI changes something, and where it changes nothing
From what your team already hands to AI and what DORA measures: three first moves, and what to re-measure in a month.
The author
Every figure carries its origin: DORA 2025 data with figure and page, JDUCRAFT reading, or estimate.
Method and sources