First test
Most arguments about a delayed programme are opinion against opinion, because nobody has counted either side. This page counts both. Give it eight things you already hold and it returns what one month of slipping is actually worth, and how much change is already landing on the same people, set against twenty five programmes read from the public record.
It runs entirely in your own browser. Nothing you enter is sent anywhere, stored anywhere, or seen by us. Your result downloads straight from this page.
Two things. What one month of delay is actually worth in money, and how much change the affected people are already carrying before this one lands.
Before the next irreversible commitment. A contract award, a gate, a board paper. Once the money has moved, a count changes nothing.
Turning two arguments into two numbers. Most delay conversations are opinion against opinion because nobody has counted either side.
Every large change has a date, and somebody chose it. Usually a contract ending, a regulator, a political promise or a supplier's roadmap. Almost never a measurement of what the people can absorb.
Then the plan is built backwards from that date, and the business case is written to fit the plan. The question of whether anyone can carry it stops being something you could test and quietly becomes something the date has already answered.
In all twenty five programmes read from the public record, the date came from somewhere other than evidence about absorption. Not one exception. That is why it is the first thing this page asks, and the single strongest signal it collects.
A completed example, run on the Emergency Services Network from figures the Home Office published on 4 August 2026. Three pages. It is the actual output of this page, not a mock up.
Sense Shift
Developers of MOSAIC · The first test
What this tells you
The twelve month figure, the load on every group you listed, where each one sits against the twenty five, and which of the five board questions you can answer.
Your result appears on this page and downloads from it. We are not going to email it to you. We hold these two lines only so we know who to follow up with, and only if you ask us to.
Three counts, no score. Nothing above is a judgement about your programme, because a judgement needs evidence this page has not seen.
Where the 60 per cent comes from. It is a stated assumption, not a measured one, and it is deliberately conservative. It says that when a slip happens, roughly six tenths of the scope is still in flight and therefore still burning cost, while the rest has either landed or has not started. Programmes that slip late in delivery are lower. Programmes that slip early, or that hold a large standing team, are higher.
If you know your own figure, put it in the last field and the page uses yours instead. Nothing here is a forecast, and every step is shown so you can disagree with a specific number rather than the whole thing.
Load is the number of other changes landing on that group, plus this one.
What counts as a change. Anything that requires those people to work differently and has its own plan, budget or go live date. A new system, a restructure, a policy shift, a site move, a regulatory deadline they have to meet. Not business as usual, not a project that touches somebody else, and not a task.
One consolidated plan. Ford in 2006 sat here, and it is the only one of the twenty five that met all five conditions.
Carryable if somebody has counted it and sequenced it. In twenty of the twenty five, nobody had counted it at all.
Above what most organisations have ever measured. At this level the order changes are landing in matters as much as the number.
Greater Manchester Fire and Rescue in 2026 carried nine further changes on the same 2,100 staff. HMRC moved 170 offices into 13 while fourteen other major programmes ran on the same workforce.
These boundaries are a judgement anchored to named cases, not a statistical threshold. We say so rather than dress them up.
Every figure above is from the published record. Sources are named on the evidence page. This is what the count looked like in a real organisation, not a prediction about yours.
Why these five. They are the five things that were missing, in the same order, across twenty five programmes in nine countries and more than two decades, at budgets differing by a factor of a thousand. One programme of the twenty five had all five. The rest are missing at least one, and the ones that went worst are missing all five.
Unanswered does not mean the answer is bad. It means nobody has looked, so the assumption sits in the business case untested. That assumption is usually the largest single line in the case and the only one with no owner.
You can answer one of five. The read answers the other four.
Nothing, unless you want it to. This count is yours to use however you like, including against us.
Three questions it should trigger, if the numbers are uncomfortable. Who owns the assumption that these people can absorb this at this pace, and where is it written down. What would have to be true for the date to move, and who is allowed to move it. And if the load is above two, has anyone put the changes in an order, or are they all simply landing.
We are deliberately giving you questions rather than actions. An action from eight inputs would be advice, and advice needs evidence.
If you want the other four answered, there are two routes, depending on which side of the decision you are standing on.
Four weeks, working entirely from evidence you already hold. The plan, the approved case, the change calendar, the org structure. Nothing is commissioned that does not already exist.
It ends in one dated verdict, Full, Phased or Hold, with the named conditions that must close before the next commitment, and a register entry with its scoring date so the call can be checked against the outcome later.
The fee is fixed in writing before week one, banded by the whole life value of the programme, and it does not move with the answer. A Hold earns exactly what a Full pace earns.
Five days, fixed fee. Before you commit to a pace in a proposal, an independent read on whether the client can actually absorb it.
You get the bid call, and the conditions drafted as wording you can put in the proposal itself.
A do not bid costs exactly what a bid costs, so there is no incentive to tell you what you want to hear.
Either way, one line is enough to start. Tell us what you have committed to and what fixed the date. If nobody has formally tested whether your people can carry it, that is the conversation worth having, and it costs nothing to have it.
It counts. It does not judge, and it is important you know exactly where the edge is.
Tell you whether to proceed. That needs evidence this page has never seen and a reader who has looked at it.
Test whether your leaders agree. Ask them on a form and they all tick yes. At NS&I the word deliverable carried three incompatible meanings across three parties.
Locate the culture gap. That comes from what employees actually wrote, not from a number you typed.
Say anything about technical feasibility, supplier capability, commercial terms, or whether the strategy is right. None of that is what MOSAIC reads.
Predict your outcome. Two forward calls are registered and neither has reached its scoring date. Nobody in this market publishes their record, and ours becomes checkable from 2027.
Everything above is what a four week read is for. The value of this page is that it stops exactly where the evidence stops.
Twenty five large programmes, read from published audits, parliamentary committees, regulators, court and securities filings, and independent evaluations. No client data, no survey, nothing self reported.
Fifteen public sector, eight private, two nonprofit. Nine countries across the United Kingdom, the United States, Europe, Asia and the Middle East. More than two decades, from the NHS National Programme in 2002 to a national AI commitment in 2025. Budgets differing by a factor of a thousand.
Twenty five is not a large sample and is not presented as one. It is large enough to show that the same five things are missing, in the same order, across sectors and countries and decades that have nothing else in common. A retail stock valuation problem and a clinical time per consultation problem turn out to be the same failure: a working practice nobody had located, treated as a configuration setting.
MOSAIC stands for Methodology for Organisational Strategy, Adoption, Integration and Culture. It is the framework a Sense Shift read runs on, and it tests one thing: whether an organisation can absorb a given change at the pace it has committed to, alongside everything else already landing on the same people.
It runs on evidence that already exists rather than on surveys. Interview notes, verbatim engagement comments, the change calendar, the org structure, real closed cases, real recent decisions. Where the record does not answer a question, the square is left open rather than filled with judgement.
Its base is twenty five large programmes read from published audits, parliamentary committees, regulators, court filings and independent evaluations. Fifteen public sector, eight private, two nonprofit, nine countries, more than two decades. One of the twenty five met all five conditions.
The scoring itself, the prompts, thresholds and decision rules, stays with Sense Shift. That is the instrument, and it is not published.
Nothing left this page. No figure you typed was sent to us, stored by us, or seen by us. There is no server behind this form and no account was created. If you download your result, it is generated in your browser from what is on screen.
We hold your email and organisation name only so we know who to follow up with, and only if you ask us to. We will not add you to a list.
This is a count produced from the figures you entered. It is not a MOSAIC verdict. Only Sense Shift issues verdicts on pace, and only after a read. The benchmark is twenty five programmes read from the public record, not client data.
Sense Shift Advisory LLP, in formation · Developers of MOSAIC · senseshiftadvisory.com
No verdict is issued from a form. This is the count. The read is the judgement.