This is a sample report, for Fieldgate Consulting (fictional).
A 60-person technology consultancy: three delivery leads, a dozen concurrent engagements, and a delivery method that mostly lives in people's heads. Every number below was calculated by the same arithmetic your report would use. Your own report takes about 15 minutes and is free.
Get your own reportDelivery maturity assessment · TDMM v0.2
Your delivery maturity is Level 3 — Established
46/100
There is a defined way of working, and most projects follow it. Roughly 95% of organisations are expected to land at or below this level, on the model's calibrated distribution.
You are an Established organisation with a visible weak seam: the work is run competently, but how it is checked, evidenced and landed still depends on individuals.
An overall score of 46 out of 100 places you at Level 3 — Established. There is a recognisable way of working here, and most of your delivery stages sit at a consistent mid-level: engagements start properly, requirements are documented, and the build is tracked against a plan.
The shape of the result matters more than the headline. Two things stand out. First, Test & Readiness at 25 is your weakest stage by some distance — testing is squeezed into whatever time remains, and the deployment is never rehearsed end to end. Second, your Data & Visibility enabler at 33 means delivery data still lives in personal spreadsheets, which quietly limits almost everything else: forecasting, margin visibility, and any AI ambition you may have.
Governance at 42 completes the picture: risks are managed and decisions get made, but nobody independently checks project health until something has already gone wrong. Taken together, this is a capable organisation whose results depend more on its people than its system — a very common position, and a workable one, provided you strengthen the seam before growth widens it.
Where to start
Written by TIM
Start with assurance: a lightweight, independent health check on your larger engagements, because until you have that, you cannot trust what your projects are telling you. Rehearse your next significant deployment end to end — it directly de-risks your weakest stage. Then begin consolidating delivery data into one shared source; it is unglamorous, but it is the rung everything else on your ladder stands on.
Your three biggest gaps
Ranked by how much each one matters — weight × severity, not simply your lowest scores.
- 1
How is the health of a project independently checked?
You answered at level 2 of 5
Written by TIM
Without independent health checks, you find out about trouble when it is already expensive. This is the single biggest driver of margin surprises in organisations your size.
- 2
Do you rehearse the deployment before doing it for real?
You answered at level 1 of 5
Written by TIM
An unrehearsed deployment means your cutover plan is a hypothesis. Rehearsal is the cheapest insurance delivery offers — it converts the runbook into a proven procedure.
- 3
Where does delivery data — time, cost, progress — actually live?
You answered at level 2 of 5
Written by TIM
Personal spreadsheets cap everything downstream: estimating calibration, live margin, trustworthy forecasts, and any meaningful AI adoption all need one shared source of truth.
Worth a second look
One pair of your answers is difficult to hold at the same time.
You describe a detailed cutover runbook with rollback triggers, but also say the deployment is never rehearsed end to end. A runbook that has not been executed under realistic conditions is a plan, not a proven procedure — and cutover is where untested assumptions become very expensive very quickly.
Written by TIM
Your strongest stage score rests on a runbook that has never been executed under realistic conditions. In my experience that combination holds until the first genuinely complex cutover, at which point the runbook's untested assumptions surface at the worst possible moment. A dress rehearsal would convert this apparent strength into a real one.
The fixes, matched to your gaps
These address exactly what the assessment found — save your report below to open them free.
Test Strategy & Dress Rehearsal Guide
LockedOnline guide
What 'tested' actually means: a test strategy covering scope, environments, data and exit criteria, plus how to run a dress rehearsal that genuinely de-risks go-live.
Project Health Check Playbook
LockedOnline guide
A step-by-step method for running independent project health and risk reviews, with the evidence to look at and how to track findings to closure.
Delivery Data Consolidation Guide
LockedOnline guide
A practical path from scattered personal spreadsheets to a single shared source of delivery data — the foundation almost everything else depends on.
Where AI would actually help you
Based on your weakest areas — not a general case for AI.
Written by TIM
With delivery data at 33, the honest advice is to fix the foundation before automating on top of it — AI assembling reports from five personal spreadsheets automates the inconsistency. Two moves are worth making now: use AI to generate test cases from your requirements, which attacks your weakest stage directly and costs almost nothing to try; and let it draft health-check evidence packs so the assurance reviews you are about to introduce take half a day, not two. Anything touching client data waits until your AI guidance is formalised into a policy.
Your delivery, stage by stageSeven stages in project order, calculated from your answers.
Written by TIM
At 50, engagements start on a reasonable footing — scope is broadly clear and kick-off happens — but quality still depends on who runs it.
Written by TIM
At 50, requirements and estimates are produced with care but without the calibration that would make them defensible under pressure.
Written by TIM
At 50, the build is tracked against a plan and quality is checked, inconsistently. Solid, and reliant on good people.
Written by TIM
At 25, this is your weakest stage. Testing has no agreed definition of done, and the first full run of any deployment is the real one. This is where your delivery risk concentrates.
Written by TIM
At 58, go-lives are your strongest stage — a detailed cutover runbook exists. The consistency finding below explains why that strength is more fragile than it looks.
Written by TIM
At 50, support after go-live is agreed but loosely scoped, which is how free support quietly extends.
Written by TIM
At 50, handover happens but whether it sticks depends on the people involved, and outcomes are checked informally at best.
Enablers, across every stageThe capabilities that decide whether each stage goes well, whoever runs it.
Written by TIM
At 42, risks are managed on most projects, but project health is only checked when something has visibly gone wrong — so bad news arrives late.
Written by TIM
At 50, a defined method exists and is broadly followed, with ad hoc tailoring. Consistency is aspiration more than fact.
Written by TIM
At 33, delivery data lives in personal spreadsheets. This is the enabler holding the others back — forecasting and margin visibility cannot outgrow their data source.
Written by TIM
At 50, margin is reviewed during delivery and most scope changes are logged, though sign-off before work starts is inconsistent.
Written by TIM
At 50, onboarding and capability development exist but are not yet systematic — knowledge still leaves when people do.
Written by TIM
At 50, AI is used in pockets with partial guidance. Sensibly, adoption has not outrun governance — keep it that way as you scale it.
Where this leads: the solution ladderManual → systemised → AI-assisted, with your current rung read from your answers.
- 1.Manual & templatesYou are here
Spreadsheets, documents and discipline.
Where almost every organisation starts, and where this toolkit's templates live. Nothing above works without consistent manual practice first — a tool cannot systemise a process that changes with every project manager.
- 2.SystemisedNext rung
One delivery system instead of personal spreadsheets.
A central source of truth for time, cost and progress: reporting assembles itself, margin is live, and forecasts become checkable. Worth the investment once manual practice is consistent — systemising chaos just makes the chaos faster.
- 3.AI-assisted & agenticNext but one
AI drafts, assembles and monitors; agents run defined workflows under human direction.
We believe project managers will increasingly direct teams of AI agents. It only works on top of the rungs below — governed data, consistent method, and an AI policy — because an agent executing an inconsistent process makes the inconsistency faster and less visible. An option to grow into deliberately, not a leap to take first.
This could be your organisation
Thirty-nine questions following the life of a project, scored deterministically, with the assets matched to your own gaps unlocked free.