Fund the next question, not the whole innovation dream

Staged innovation funding means approving the next responsible learning commitment before approving a much larger delivery commitment. It gives a team enough money, time and access to investigate a consequential uncertainty, then brings the evidence back to a decision-maker. It is not simply a large project budget released in instalments regardless of what happens.

The aim is to avoid two extremes: demanding a precise business case before meaningful learning is possible, and financing an entire vision because an early presentation is persuasive. A useful funding decision connects the question, the test, the resources and the conditions for continuing.

Separate the opportunity from the next commitment

Consider a fictional packaging business exploring a reusable delivery box for small ceramics sellers. The long-term idea could involve box manufacture, collection, inspection and repeated use. The immediate uncertainty is much narrower: can a proposed insert protect the product through a relevant handling cycle, and can the box be checked and prepared for reuse?

The team should not need approval for a complete national return network to investigate those questions. Equally, the cost of a few prototypes should not be presented as the cost of launching the whole service. Keep the opportunity estimate, next test budget and possible later investment separate.

Strategyzer’s discussion of innovation metrics distinguishes uncertainty and potential from time and cost. That is a useful management distinction, not an objective measurement of success probability. A risk score still depends on the evidence and judgement behind it.

Write the decision in one sentence: “Approve a bounded packaging and handling investigation so we can decide whether to revise the design or prepare a controlled service trial.” Someone reading it should understand what is being authorised and what is not.

Fund an answerable question

A test brief should describe the assumption, why it matters, how it will be investigated and what result would change the plan. “Explore reusable packaging” is too broad. “Compare whether two insert designs remain usable after a defined handling sequence” is specific enough to plan, provided qualified people define appropriate test conditions.

Use assumption mapping to distinguish consequential unknowns from easy questions. A team can spend its entire budget refining the box’s appearance while avoiding the difficult return-and-inspection problem. Cheap tests are not useful if they address the wrong uncertainty.

GOV.UK’s alpha guidance encourages focused work on risky assumptions. Apply that principle without copying government phase names into a business that does not use them. The important connection is between uncertainty and evidence, not a particular process vocabulary.

Do not create a misleadingly small test by omitting necessary safeguards. Product protection, worker safety and customer commitments require appropriate expertise and conditions. A learning budget is not permission to use real customers as an uncontrolled experiment.

Count more than cash

List the people, equipment, access, materials and coordination the investigation needs. Include the work that will be displaced. A specialist who is nominally available but committed to current orders is not free capacity.

For the fictional packaging team, resources might include design time, sample materials, a qualified testing provider, delivery-operator input and someone responsible for documenting results. Later trials could require customer support, return handling and storage. These belong in the planning conversation even when they sit in different departments’ budgets.

Separate spent resources from commitments that cannot easily be reversed. A prototype material purchase and a long equipment lease create different obligations. Document cancellation conditions and ownership of outputs before treating both as equivalent “experiment costs.”

The Business Portfolio Map explanation describes increasing investment as evidence develops. Use that as a staged-commitment idea, not as a guaranteed funnel in which a predictable percentage of projects succeeds. The number and size of commitments should reflect your actual constraints.

Build a short funding brief

The following format is an original planning aid. It is not a financial model or a universal approval standard.

Funding-brief element	Fictional packaging example
Decision after the work	Revise the insert, stop the concept or plan a controlled trial
Main uncertainty	Protection and inspectability under defined handling conditions
Work authorised	Prepare samples, obtain qualified tests and document handling steps
Resources	Named staff capacity, materials, testing access and review time
Boundaries	No public durability claims or full service launch
Review evidence	Test conditions, observed failures, unresolved questions and next options
Decision owner	A named sponsor able to approve, narrow or stop further work

Specify the amount and timing using your own costs. Illustrative budget percentages are rarely helpful if they obscure access delays, specialist costs or the consequences of failure. Estimate a range where uncertainty is real and explain what could move the cost.

Also record dependencies. If the proposed test needs access to a customer’s packing operation and that access is not agreed, the test is not ready merely because funding exists. Resolve the dependency or fund a different first step.

Agree the review before the test begins

Decide what the team must bring back and who must attend. Include people who understand the method and people who would inherit the proposed operation. A review dominated by presentation skill can miss practical weaknesses.

Ask for the original question, the actual test conditions, observations, interpretation and limitations. Record any changes made during the investigation. If the team altered the handling sequence after early failures, the final result should not be described as though the original test passed unchanged.

GOV.UK’s discovery guidance recognises stopping as a legitimate outcome of investigation. Make that option credible in your own review. If every stopped project is treated as personal failure, teams have a strong reason to conceal unfavourable evidence.

A review can continue, revise, pause or stop the work. “Continue” should authorise a named next commitment. “Pause” should identify a condition for restarting, not leave an inactive project consuming attention indefinitely.

A worked review with mixed evidence

Suppose the fictional team finds that one insert survives the defined tests better than another, but checking it for damage takes longer than expected. The result is neither a complete success nor a reason to discard every aspect of the concept.

The team can distinguish what has improved from what remains unresolved. It has evidence about those samples under those conditions. It does not yet know whether inspection can be made dependable in a busy return operation, whether sellers will participate or whether repeated use is economically attractive.

A reasonable next commitment might investigate inspection and return handling before expanding manufacture. Alternatively, the team may revise the design so that damage is easier to see. The funding decision follows the limiting uncertainty rather than automatically moving to a larger prototype batch.

Notice what the review does not do. It does not count the number of meetings as progress, convert positive feedback into sales forecasts or assume that money already spent justifies spending more. The previous investment bought information; the next decision concerns the value and cost of additional work.

Manage the portfolio’s shared constraints

Several modest experiments can collectively overload one testing facility, one procurement team or one sponsor. Review the portfolio as a set of competing commitments, not merely a list of individually affordable requests.

Compare the learning value of the next step with other uses of the same resource. Sometimes a promising initiative should wait because another has a more urgent decision. Sometimes a small additional test is worthwhile because it could prevent a much larger avoidable commitment.

Do not prescribe a universal split between existing-business improvements and new opportunities. Connect the allocation to the innovation strategy, the state of the evidence and the organisation’s ability to deliver. An attractive ratio cannot resolve a shortage of qualified people.

Recognise the limits of staged funding

Some work cannot be divided into tiny independent experiments. Infrastructure, regulated development and tightly coupled systems may require substantial commitments before key evidence becomes available. Staging still helps clarify decisions, but it cannot make every investment cheap or reversible.

Evidence can also decay. A supplier change, new customer context or different operating scale may reopen an assumption that once looked settled. Review what the evidence covers instead of treating approval as permanent validation.

This is management guidance, not financial advice or a formula for investment returns. Its practical value is a clearer agreement: what we are trying to learn, what we will commit, what we have not approved and how we will decide next. A good learning budget buys a better decision, not a more convincing story.