Every plan contains assumptions.
Leaders assume funding will arrive, suppliers will perform, people will change behaviour, systems will integrate, customers will adopt, benefits will materialise, approvals will land, capacity will hold and external conditions will stay manageable.
Assumptions are not the problem. Unowned assumptions are the problem.
A visible assumption can be tested. An owned assumption can be managed. A hidden assumption becomes a quiet commitment. It shapes plans, dates, costs and expectations without receiving the challenge a real decision would receive.
Delivery confidence often rests on these hidden commitments. When they fail, the organisation may describe the issue as late risk, unexpected delay or external change. In many cases, the weakness was present much earlier. It sat inside an assumption no one owned.
Assumptions are unavoidable
No leader has complete certainty. No team begins complex work with perfect information. Assumptions allow progress when evidence is incomplete.
This is legitimate. Good delivery requires judgement under uncertainty. The aim is not to eliminate assumptions. The aim is to know which assumptions matter, who owns them, how they will be tested and when leaders need to know if they no longer hold.
The danger comes when assumptions are treated as facts.
A forecast becomes accepted as affordable. A dependency becomes treated as secure. A stakeholder position becomes described as agreed. A benefit becomes expected before adoption has been tested. A supplier commitment becomes relied on before commercial or operational evidence supports it.
The language sounds confident. The evidence may not be.
Some assumptions carry more weight than others
Not every assumption deserves the same attention.
Some assumptions are minor. If they prove wrong, the team can adjust without major consequence. Others are load-bearing. If they fail, the case for cost, schedule, value, scope, readiness or benefits weakens.
Leaders should pay most attention to assumptions linked to major commitments.
Common load-bearing assumptions include funding availability, affordability, supplier capability, stakeholder agreement, user adoption, operating model readiness, data quality, legislative or regulatory timing, workforce capacity, technology integration, benefits ownership and market conditions.
These assumptions do not sit at the edge of delivery. They sit at the centre. They influence whether a decision is responsible.
Assumption logs are not enough
Many organisations already record assumptions. The register exists. The table has columns. The review pack includes a page.
But a list does not manage an assumption.
An assumption log can become a storage place for uncertainty. It records uncertainty without changing behaviour. It captures what people are relying on but may not force anyone to test it, update it or escalate it before damage occurs.
A useful assumption record should answer four questions.
· What decision depends on this assumption?
· Who owns the assumption?
· What evidence supports it?
· When will it be tested?
If those questions cannot be answered, the assumption is not being managed. It is being carried.
Ownership means more than recording
Every material assumption needs an owner.
The owner does not need to control every factor. Many assumptions depend on external events, supplier behaviour, market change or stakeholder action. But the owner must monitor the assumption, test the evidence, judge whether confidence has changed and escalate if the assumption no longer supports the decision route.
Ownership should be clear enough for leaders to know who will act when the assumption weakens.
Without ownership, assumptions drift. They stay in the plan because removing them would be difficult. They remain in papers because no one has been asked to prove them. They survive review because they are familiar. Familiarity then becomes false confidence.
Assumptions and decisions are linked
Assumptions matter because decisions depend on them.
A decision to approve investment may depend on an assumption about benefits. A decision to proceed to delivery may depend on an assumption about supplier readiness. A decision to keep a launch date may depend on an assumption about operational capacity. A decision to accept risk may depend on an assumption about impact or likelihood.
If a decision depends on an assumption, the decision-maker should see it. The assumption should not sit in a separate register disconnected from the approval.
This is where assurance adds value. Assurance should trace the assumptions behind the decision and test whether they remain safe to rely on. It should not ask only whether assumptions are recorded. It should ask whether leaders understand the assumptions they are accepting.
Testing should happen before commitment hardens
The best time to test an assumption is before the organisation has built too much commitment around it.
Once money has been spent, suppliers mobilised, public commitments made, teams reshaped or benefits announced, weak assumptions become harder to challenge. The organisation may prefer to defend the plan rather than revisit the belief beneath it.
This is why assumptions need test points. A test point is not a vague intention to review later. It is a clear moment when evidence will confirm, weaken or replace an assumption before the next decision.
Good test points should sit on the decision route. They should occur early enough to influence the next commitment.
Challenge the words which hide uncertainty
Assumptions often hide inside confident language.
Phrases such as “on track”, “broadly agreed”, “expected to”, “subject to”, “to be confirmed”, “manageable” and “low risk” can be useful. They can also disguise uncertainty.
Leaders and reviewers should slow down when these phrases appear. What is agreed? Who agreed it? What evidence supports the forecast? What remains subject to confirmation? What makes the risk manageable? What would change the position?
Plain language helps. A clear assumption statement should be specific enough to test. “Users will adopt the service” is too broad. “At least 80 percent of target users will be trained and able to complete the core transaction before launch” is more useful. It can be evidenced, owned and challenged.
Assurance should focus on the assumptions which carry the most risk
Assurance does not need to test every assumption in equal detail. It should focus on the assumptions most likely to affect delivery confidence.
The strongest questions are direct.
· Which assumptions carry the most weight?
· Which decisions depend on them?
· Who owns them?
· What evidence supports them?
· What would prove them wrong?
· When will they be tested?
· What will leaders do if they fail?
These questions help leaders see whether confidence rests on evidence or hope.
The goal is not certainty. It is conscious commitment
Leaders will always make decisions under uncertainty. The issue is whether they know which uncertainty they are accepting.
A strong assurance conversation does not pretend assumptions can be removed. It makes them visible, tests the ones which matter, assigns ownership and links them to the decisions they support.
This changes the quality of commitment. Leaders can still proceed. They may choose to accept risk, set conditions, request further evidence, change scope, delay approval or stop. The difference is they decide with clearer sight of what the plan depends on.
An assumption without an owner is not a small gap. It is a risk waiting for permission