Approach
Facts first. Then a plan small enough to finish.
Most troubled programs are not short of effort or intelligence. They are short of an agreed version of reality, and of increments small enough that finishing one proves something.
Five things we hold to
These are not values on a wall. Each one changes what we do in a specific, sometimes uncomfortable way.
Evidence before opinion
Every finding we present is traceable to an artefact, a delivery record, or an interview we can name. Opinions are labelled as opinions. It slows the first three weeks and saves the next six months.
One plan, one owner
Not a program plan plus a vendor plan plus a workstream tracker that disagree quietly. One plan, one baseline, and a single named owner per increment who can actually make the calls their name implies.
Small enough to finish
If an increment cannot be finished and demonstrated inside a few weeks, it is not a plan item — it is a wish. We decompose until completion is observable, because that is the only reliable signal of progress.
Governance that decides
A forum that reviews status and adjourns is overhead. Every recurring meeting we run has decision rights, a quorum, and a standing question: what is blocked, and who is unblocking it by when?
Same story in every room
The version the sponsor hears is the version the team hears. Programs fail slowly when the reporting layer smooths the news on its way up, and we would rather lose the engagement than participate in that.
Designed to end
The measure of the work is what holds after we leave. Successor named early, cadence run by your people while we are still there to catch it, artefacts owned by you. A consultancy you cannot get rid of is not a partner.
The four phases
Timings below are typical for a program of eighty to two hundred people. Smaller programs compress; regulated ones stretch.
Ground truth
Weeks 1–3 · Establish what is actually true
We interview across levels — not just the leads — and read the artefacts against the delivery record. The central question is simple and rarely asked: over the last six months, what did the plan say would be done, and what was demonstrably done?
That gap, expressed as a ratio, is usually the single most predictive number on the program. It also tends to end the debate about whether the current date is achievable.
Ends with
- A written findings pack with evidence attached
- The real critical path, and the constraint driving it
- A ranked list of what will break the date
One plan, one owner
Weeks 3–6 · Rebuild the plan around the constraint
Re-baselining is done with the delivery leads in the room, not to them in a back office. Scope is decomposed into increments that can be finished and shown. Each gets a named owner, an outcome you could demonstrate to a sceptic, and an explicit set of dependencies.
Where capacity genuinely cannot meet the date, we surface the trade-off with options and costs rather than absorbing it into optimism. This is the phase where the difficult conversation happens — deliberately, and early.
Ends with
- A single re-baselined plan owned by your leads
- A defensible date, with the assumptions it rests on written down
- Agreed trade-offs, signed off by the sponsor
Run it
Ongoing · Hold the cadence
A weekly delivery review that decides. A 48-hour escalation path with a named resolver. A burn-up the sponsor can read unaided, and a risk register where every item has an owner and a next action dated this month.
We are in the detail here — chasing dependencies, testing "done", and having the conversations that the org chart makes awkward. This is the part that consultancies usually delegate downward and we do not.
Running artefacts
- Weekly delivery review with decision log
- Increment burn-up against baseline
- Live risk and dependency register with dated next actions
- Monthly sponsor re-forecast, unsmoothed
Hand back
Final 4–6 weeks · Make ourselves unnecessary
The successor is named in the first month, not the last. In the final weeks they chair the cadence while we sit in the room, and then we stop attending. Documentation is the artefacts you have been using all along, not a farewell deck.
We offer a check-in at thirty and ninety days after exit, unpaid. If the cadence has decayed by then, that is useful information about how we built it.
Ends with
- Named successor running the cadence unaided
- All artefacts owned and maintained in your systems
- A short close-out note on what remains fragile
What we don't do
Being clear about this early saves everybody a procurement cycle.
- We don't staff up. No bench, no pyramid, no junior team learning your business on your budget.
- We don't build the thing. We are not an implementation partner and hold no reseller or referral agreements — so our assessment of your vendor has nothing riding on it.
- We don't run methodology projects. Nobody needs a framework rollout. They need this program to land.
- We don't write the report you ordered. If the evidence contradicts the brief, you get the evidence.
- We don't take work we can't staff properly. Two active engagements at a time. If we are full, we will say so.
Next step
Start with three weeks and a straight answer.
The delivery assessment is fixed fee and self-contained. You can commission it without a procurement cycle and owe us nothing afterwards.