A new venture becomes most dangerous at precisely the moment it becomes plausible.

There is a credible customer problem. A respected executive is prepared to sponsor it. Early conversations are encouraging. A forecast shows meaningful revenue in year three. The leadership team can picture the strategic story.

That is often enough to secure a pilot. It should not be enough to secure a scale budget.

The problem is not a shortage of intelligence or ambition. It is that the evidence collected to support a growth idea is often much weaker than the decision being requested. Customer interest is treated as demand. Gross margin is presented before the cost to acquire, serve and retain the customer is understood. Access to an existing client base is mistaken for a right to win. A successful pilot is assumed to be repeatable.

The result is a venture that remains "promising" for several quarters while consuming cash, scarce specialists and leadership attention. Nobody wants to close it because the strategic logic still sounds attractive. Nobody can scale it with confidence because the commercial proof remains incomplete.

A better approach is to make the evidence standard explicit. The Five-Gate Test helps a leadership team decide whether a growth bet should advance, be reshaped, pause for a specific dependency or stop.

Growth bets fail between enthusiasm and evidence

Most businesses have some form of idea generation. Many also have a budget process. The weak point is the space between them.

An idea enters that space as a set of assumptions. It should leave as one of three things: an investable growth case, a bounded experiment designed to resolve a material uncertainty, or a stopped proposal with its lessons recorded.

Instead, assumptions often harden into facts through repetition. A market estimate becomes the sales plan. Positive interview feedback becomes the conversion rate. A founder's relationship becomes the distribution strategy. A capable project team becomes the operating model.

The Five-Gate Test interrupts that progression. It asks five different questions:

  1. Do we have a strategic right to win?
  2. Is there evidence of paying demand?
  3. Can the economic engine become attractive?
  4. Can we deliver the proposition repeatedly?
  5. Is there a credible route to scale?

These gates are deliberately separate. A venture can have strong customer demand and poor economics. It can have attractive unit economics and no realistic distribution route. It can fit the strategy but depend on capabilities the business is unlikely to build.

The objective is not to remove uncertainty. That is impossible. It is to identify which uncertainty is being funded, what evidence would reduce it and what decision will follow.

Before the gates, define the growth bet

Leadership teams should not score a vague ambition. Start with a one-page definition containing six items:

  • Target customer: A specific buyer or user, not a broad market.
  • Problem: The costly, frequent or important problem the venture will solve.
  • Proposition: The product, service or outcome being offered.
  • Revenue logic: Who pays, how much, how often and for what unit of value.
  • Strategic logic: Why this opportunity belongs with this business.
  • Decision requested: The capital, people, time and executive support required for the next stage.

If the team cannot state those items clearly, the venture is not ready for an investment discussion. It may be ready for discovery work, but that is a different request with a smaller commitment and a learning objective.

Gate 1: Strategic right to win

The first gate is not whether the market is attractive. Attractive markets draw capable competitors. The question is whether this business has, or can build, an advantage that matters to the target customer.

A right to win might come from trusted customer relationships, proprietary data, specialised knowledge, a difficult-to-replicate process, distribution access, regulatory permission, brand credibility or an installed asset base. It must be relevant to the buying decision.

"We already serve these customers" is not sufficient. Existing access helps only if customers see the new proposition as a credible extension of the relationship. A payroll provider may have access to thousands of employers, for example, but that does not automatically give it permission to sell every form of workforce advice.

Evidence to examine

  • The venture uses at least one distinctive asset that competitors cannot easily match.
  • Target customers recognise that asset as relevant to the proposition.
  • The opportunity is consistent with the organisation's strategic choices and risk appetite.
  • The new venture strengthens the core, or any tension with the core is understood and intentional.
  • The leadership team can explain why ownership is better than partnering, licensing or acquiring.

Warning signs

  • The case rests mainly on market growth.
  • The proposed advantage is a feature that can be copied quickly.
  • The venture relies on the brand while creating material brand risk.
  • It competes for the same scarce capability needed to protect the core.
  • Nobody has made an explicit build, buy or partner decision.

Gate question: What advantage gives us a credible chance of earning an attractive return in this market?

Gate 2: Paying demand

Customer interest is easy to collect and easy to overvalue. People are generous with positive feedback when there is no price, switching cost or implementation effort attached.

The strongest evidence is behaviour. A signed contract, paid trial, deposit, pre-order, completed procurement step or allocation of customer resources says more than a favourable interview. For an early proposition, a carefully designed test can still be useful, but the test should expose the customer to a real choice.

Suppose a professional-services firm is considering a subscription compliance dashboard. Ten existing clients may say they like the concept. The more useful questions are whether a defined buyer will pay a specified annual fee, provide the data needed for onboarding, assign an internal owner and use the output frequently enough to renew.

Evidence to examine

  • A clearly defined buyer has authority, budget and a reason to act.
  • The problem is frequent, costly or strategically important.
  • Customers accept a realistic price or make another meaningful commitment.
  • The venture has tested acquisition beyond the sponsor's personal network.
  • Reasons for rejection, delay and non-use are recorded, not filtered out.
  • There is early evidence of use, outcome and willingness to renew where the model requires it.

Warning signs

  • The team reports conversations rather than conversion.
  • Most prospects are friendly existing contacts.
  • The proposition requires extensive explanation before its value is understood.
  • The forecast assumes a shorter sales cycle than the pilot revealed.
  • Custom requests are being mistaken for a repeatable market need.

Gate question: What have target customers done that they would not do if the problem were unimportant?

Gate 3: Economic engine

Revenue can disguise a weak business. The economic gate asks whether each additional customer can contribute enough value to cover acquisition, delivery, support, product development and the overhead needed to scale.

Early economics will contain ranges rather than precise figures. That is acceptable. False precision is not. The team should show the variables that matter most and the evidence behind each range.

For a recurring-revenue service, those variables might include contract value, implementation effort, ongoing service hours, gross margin, acquisition cost, sales-cycle length, retention and working-capital timing. For a product business, the critical variables may include landed cost, returns, channel margin, stockholding, warranty exposure and reorder rate.

The purpose is not to demand mature margins from a pilot. It is to prove that a plausible route to attractive economics exists and to identify the milestone at which that route must become visible.

Evidence to examine

  • Price has been tested against a defined proposition and buyer.
  • Direct delivery and support costs are measured from actual activity.
  • Customer acquisition assumptions reflect the intended channel.
  • Working-capital and cash timing are visible.
  • The model includes downside and slower-growth cases.
  • The break-even point is linked to operational requirements, not only revenue.
  • There is a clear view of which costs decline, remain fixed or increase with scale.

Warning signs

  • Gross margin excludes substantial implementation or founder time.
  • The forecast assumes price rises that have not been tested.
  • Customer acquisition is treated as free because the first clients came from existing relationships.
  • The plan needs high volume before unit economics become positive.
  • The venture appears attractive only in the management case, not the downside case.

Gate question: What must be true for this venture to generate cash at an acceptable return, and which of those conditions have been demonstrated?

Gate 4: Delivery repeatability

A pilot can succeed because exceptional people compensate for an incomplete system. The founder sells it. The strongest operator manages every handoff. Product gaps are bridged manually. Senior specialists solve issues before the customer sees them.

That may be appropriate during discovery. It is dangerous when pilot performance is used to justify scale.

The delivery gate tests whether the proposition can be sold, onboarded, delivered, supported and governed without permanent heroics. The aim is not full industrialisation. It is evidence that the work can become repeatable at the required quality, speed and risk level.

Evidence to examine

  • The end-to-end customer journey and delivery workflow are defined.
  • Roles, decision rights and escalation paths are clear.
  • Critical capability gaps have owners and realistic acquisition plans.
  • Data, technology, suppliers and partners have been tested in the intended environment.
  • Service quality and risk controls can survive more customers and less executive intervention.
  • The team understands which elements should remain bespoke and which must be standardised.
  • Front-line learning changes the proposition and the operating process.

Warning signs

  • Every customer receives a different version of the offer.
  • Senior people remain embedded in routine delivery.
  • The pilot team cannot explain the true cost of manual workarounds.
  • A critical partner or platform has not committed.
  • Risk, legal, security or compliance work is deferred until after the scale decision.
  • Quality depends on one individual whose time is already constrained.

Gate question: What breaks when customer volume triples, and do we know how to fix it?

Gate 5: Scale case

Scale is not a larger pilot. It introduces different problems: channel productivity, management capacity, cash absorption, recruitment, service consistency, technology reliability and competitive response.

The scale gate tests the mechanism by which the venture will move from a handful of supported customers to a material business. A list of target accounts is not a scale model. Nor is a top-down share of a large market.

The team should define a repeatable growth unit. This could be one salesperson reaching a target level of annual recurring revenue, one site achieving a contribution threshold, one partner producing a known volume of qualified demand, or one implementation pod onboarding a set number of customers at an agreed quality level.

Until one growth unit works, multiplying it in a spreadsheet adds volume but not confidence.

Evidence to examine

  • There is a defined beachhead segment with a clear reason to buy.
  • At least one acquisition channel shows repeatable performance.
  • The growth unit and its capacity constraints are understood.
  • Capital is released in stages against observable milestones.
  • The core business can provide shared assets without becoming a bottleneck.
  • Leadership capacity and venture governance are designed for the next stage.
  • The team has defined the conditions for further investment, a change of direction and exit.

Warning signs

  • The scale plan assumes several channels will work despite none being proven.
  • Growth relies on the founder remaining the principal salesperson.
  • The forecast ignores the time needed to hire and develop scarce capability.
  • Shared functions are expected to absorb work without a capacity plan.
  • The investment request funds a calendar period rather than a value milestone.
  • There is no stop rule because the total addressable market remains attractive.

Gate question: What repeatable unit will turn today's evidence into a material business?

Score evidence, not confidence

Each gate should receive a score from 1 to 5. The language matters:

Score each gate independently. Do not average away a fatal weakness. A venture with strong strategic fit, delivery capability and market excitement should not advance to scale if customers will not pay or the economic engine remains structurally unattractive.

Confidence should also be recorded separately. A score of 3 with high confidence is different from a score of 3 based on a small or biased sample. For each gate, note:

  • the score;
  • the evidence supporting it;
  • the material assumption still open;
  • confidence in the evidence;
  • the next test; and
  • the owner and decision date.

Use a different threshold at each stage

An early idea should not be rejected because it lacks scale evidence. A scale request should not pass because the idea is strategically interesting. The minimum standard must rise with the commitment.

Discovery to pilot

Require at least 10 out of 25, with no score of 1 for strategic right to win or paying demand.

At this stage, the business is funding learning. The proposal should identify the most important assumption, the cheapest credible test and the decision that will follow. The team should cap time and money.

Pilot to repeatable offer

Require at least 15 out of 25, with no score of 1 for the economic engine or delivery repeatability.

The business is now proving that the offer can work beyond a single favourable case. Customer selection should begin to reflect the intended market, and hidden founder or specialist effort should be made visible.

Repeatable offer to scale

Require at least 20 out of 25, with no gate below 3 and scores of at least 4 for paying demand and the economic engine.

At this stage, the venture is asking the organisation to commit meaningful capital and capacity. The scale mechanism, leadership model and stop conditions should be as explicit as the revenue ambition.

These thresholds are a starting point, not a substitute for judgement. A regulated, capital-intensive or brand-sensitive venture may require a higher standard. A small reversible experiment may justify a lower total commitment, but it should still have a precise learning objective.

Make one of four decisions

An investment review should end with a decision, not a request for a better presentation.

Advance

The venture meets the stage threshold, there is no critical gate failure, and the next release of capital is tied to a clear milestone.

Reshape

The underlying opportunity remains credible, but the proposition, segment, channel, price or delivery model needs to change. Reshaping should produce a new thesis and test, not an indefinite extension.

Pause

A specific external or internal dependency prevents a fair test. Examples include regulatory permission, access to essential data, a partner commitment or a capability appointment. A pause must have an owner, condition and review date.

Stop

Evidence contradicts a critical assumption, the venture has exceeded its agreed time or capital envelope, or the expected return no longer justifies the risk and attention required.

Stopping is not evidence that the original decision was poor. A well-designed growth process expects some ideas to stop. The failure is continuing to fund a contradicted thesis because stopping has become politically difficult.

Separate the sponsor from the evidence

Every growth bet needs a senior sponsor. It should not depend on that sponsor acting as prosecutor, witness and judge.

A robust review has four roles:

  • Venture lead: Owns the proposition, evidence and delivery plan.
  • Executive sponsor: Removes enterprise barriers and protects legitimate learning.
  • Independent challenger: Tests assumptions, sample quality and economic logic.
  • Investment decision-maker: Allocates capital and records the decision conditions.

In a smaller business, one person may hold more than one role, but the perspectives should still be distinct. The founder who originated the idea can ask another director or external adviser to challenge the case before the investment meeting.

The meeting itself should spend little time on background. Review the five scores, the evidence that changed since the last gate, the largest remaining uncertainty and the decision requested. Record what would cause the next release of capital, a reshape or a stop.

Run the first review in ten working days

A business does not need a venture office to use the method. One growth bet can be reviewed in a focused ten-day cycle.

Days 1 and 2: Define the bet

Write the one-page venture definition. Agree the current stage and the precise decision being requested.

Days 3 to 5: Assemble evidence

Collect customer behaviour, pipeline conversion, price tests, delivery activity, cost data, cash requirements, capability dependencies and competitive evidence. Separate observed facts from estimates.

Days 6 and 7: Score independently

Ask the venture lead, sponsor, finance lead and an independent challenger to score the five gates before discussing them. A wide scoring gap is useful evidence. It shows where the organisation is relying on different facts or standards.

Day 8: Resolve the critical uncertainty

Identify the one assumption most capable of changing the decision. Do not produce a long risk register. Define the quickest credible test, its cost and what result would count as support or contradiction.

Day 9: Hold the gate review

Choose advance, reshape, pause or stop. If advancing, approve only the resources required to reach the next evidence milestone.

Day 10: Communicate and record

Document the scores, evidence, decision, conditions and next review date. Tell affected functions what has and has not been approved.

Five questions for the next leadership meeting

If a growth proposal is already seeking funding, start here:

  1. Which part of the case is observed evidence, and which part is still an assumption?
  2. What has a target customer done that demonstrates willingness to pay or change behaviour?
  3. Which two variables have the greatest effect on cash and contribution?
  4. What did exceptional people do manually in the pilot that cannot continue at scale?
  5. What result would make us stop, and have we agreed it before releasing more capital?

If the team cannot answer those questions, it does not necessarily mean the venture should stop. It means the next investment should buy evidence, not growth.

Fund the next proof, not the full story

The quality of a growth portfolio is determined as much by what a leadership team refuses to scale as by what it backs.

A strong growth bet does not arrive fully proven. It earns confidence in stages. Strategic fit creates permission to investigate. Customer behaviour creates permission to pilot. Attractive economics and repeatable delivery create permission to scale. The scale mechanism creates permission to commit serious capital.

The Five-Gate Test gives leadership teams a shared standard for those transitions. It protects promising ideas from being judged too early, and it protects the business from funding a persuasive story long after the evidence has weakened.

Allington Advisors helps founders, CEOs and leadership teams test strategic growth choices, strengthen the commercial case and turn selected opportunities into executable plans. A focused Growth Bet Review can assess one proposed venture or a small portfolio and define the evidence needed for the next decision.