The Mirage

The False Positive

18 Primary in The 198

20.2% Named factor across all 198

4.1 Avg overall failure score

The pattern

The business was real. The conditions were not. Growth, revenue, valuations — all genuine in the moment, all contingent on an external environment that was temporary, not structural. Leaders in this zone did not misread their business. They misread the world around it — and assumed permanence where there was only a window.

The ZIRP-era company scaled on free capital. The pandemic business scaled on forced behaviour. The regulatory darling scaled on policy conditions that reversed. The dot-com play scaled on a market that didn't yet exist at the required size. In each case the metrics were accurate. The error was treating them as proof of something durable.

DZ3 is dangerous precisely because it looks like success until the moment it doesn't. The numbers validate the model. The team believes the model. Investors believe the model. The model is real — it just depends on conditions that will not last. By the time those conditions shift, the business has typically been built, capitalised and committed to a scale that assumes they will.

The signal you are probably not looking at

At least one significant driver of your current growth depends on a condition that did not exist three years ago. You have not stress-tested the business against its removal. You have been calling that condition "the market." It is not the market. It is a window. The question is not whether it will close. The question is whether you will have built the floor before it does.

You are approaching this zone

Other businesses have stood where you are standing. Some recognised the dependency and acted before the window closed. Some didn't. Both paths are documented below.

Businesses that didn't act — and what it cost them

Pets.com E-commerce · US · Failed 2000 · Age 1 Did not act

Signal visible 12 months before shutdown: $400 customer acquisition cost against <$30 average order value — a model only viable if dot-com capital availability continued indefinitely

Pets.com raised $300M and spent nearly $12M on marketing in its final year, including a Super Bowl ad, while burning through capital at a rate its revenue could never justify. The business model was not broken — it was contingent. Every unit of growth depended on cheap capital continuing to subsidise customer acquisition that the underlying economics couldn't support. When the NASDAQ corrected in Q1 2000, it didn't kill Pets.com. It removed the condition that had been keeping it alive. The business went from launch to liquidation in 268 days.

What the dependency looked like from inside

The plan was to reach a scale at which unit economics would work — a plan that required the capital environment to remain unchanged long enough to get there. The signal that it wouldn't was in the numbers every month. When the window closed, the company had nine months of runway and nothing underneath it.

Convoy Logistics · US · Shutdown 2023 · Age 8 Did not act

Signal visible late 2021: freight volumes normalising from pandemic peak while $3.8B valuation assumed those conditions persisting — no documented unit economics independent of elevated freight demand

Convoy built a digital freight brokerage that grew sharply during a period of exceptional freight demand driven by pandemic supply chain disruption. By late 2021, volumes were returning to trend. The company's headcount, cost base and $3.8B valuation all reflected peak conditions. The CEO's final memo described "an unprecedented freight market collapse" — an accurate description of what happened, but not of the underlying architecture. The dependency had been visible in the unit economics for two years before the end. The collapse wasn't unprecedented. It was a reversion.

What the dependency looked like from inside

A $3.8B valuation made honest internal reckoning with the underlying economics politically difficult. Every conversation about cost structure had to confront what a reset would do to the fundraising narrative. The number that was supposed to represent the business's future had become the ceiling preventing the floor from being built.

A business that recognised it in time

Patch Plants E-commerce · UK · Founded 2016 · Still trading Acted

Signal visible 2021: pandemic houseplant boom driving exceptional growth on behavioural conditions — working from home, closed garden centres, nesting — that were structurally temporary

Patch launched in 2016 as the UK's first online plant delivery brand, built on a simple proposition: make plants accessible to people who found gardening intimidating. The pandemic created extraordinary conditions — locked-down consumers, closed garden centres, a surge in home improvement behaviour — and Patch grew rapidly into them. By 2021 those conditions were visibly reversing. Offices were reopening. Garden centres were trading again. The behavioural wave that had pulled consumers online for plants was normalising.

Rather than defending the pandemic growth rate or scaling into conditions that were already turning, Patch focused on what was structurally theirs: the core convenience proposition, the novice customer, the care relationship. The plant doctor service, care content, gifting mechanics — all built recurring engagement that worked independently of pandemic behaviour. The business continued to grow through normalisation and has been recognised as the UK's best plant delivery service three consecutive years.

What they did differently

They did not try to hold the pandemic growth rate. They identified what was structurally durable — the proposition that worked without the tailwind — and invested in that while conditions were still favourable enough to fund it. The business they have now doesn't depend on locked-down consumers. That is the only version of the business worth having.

What the turnarounds had in common

Businesses that navigated out of DZ3 didn't wait for the conditions to change. Across the cases, three things consistently separated those that survived from those that didn't.

They named the dependency before it was forced on them

They ran a deliberate exercise: strip out the condition. Remove the cheap capital, the regulatory tailwind, the pandemic behaviour, the market boom. What is left? If the answer was "not much," they treated that as a structural finding requiring action, not a temporary problem to be managed. The businesses that failed named the same dependency — but in their post-mortems, not their planning cycles. The naming happened either way. The timing was everything.

They built the structural floor while they still had capital to build it

The window while conditions are still favourable is also the window in which you have the resources to act. Every business that survived used the good conditions to construct something that worked without them — a retention mechanic, a different customer segment, a unit economics model that didn't require subsidy. Every business that failed used the same conditions to scale the dependency instead. Both groups had the same window. One built through it. One grew into it.

They made the decision when it was still uncomfortable, not when it was unavoidable

The hardest version of this decision is the one made when the business is still growing, the team still believes, and conditions haven't visibly deteriorated yet. That is also the only version that works. By the time deterioration is visible, capital has contracted, options have narrowed, and the cost of the reset has multiplied. Every business that waited for the signal to become undeniable found that it became undeniable and terminal at the same time.

The signal is usually visible 18 to 24 months before it becomes undeniable. The businesses that survived acted in that window. The ones in this database did not.

Business 199 is always optional

The interrogator identified this architecture. What it cannot do is tell you how much of your window remains.

That requires someone who has been inside businesses at this stage — who can look at your specific dependency, your current capital position, and your runway, and tell you honestly whether you are still in the window or already past it. That is the conversation this tool is designed to open.

Take the interrogator →