Examples are illustrative unless otherwise attributed.
Make the goal falsifiable and useful
Specify the metric, window, population, and business purpose. A million cumulative registrations can coexist with low current use. A million monthly sessions can produce little contribution. Define the customer outcome that must survive as the volume grows, then choose an appropriate scale target.
Use stage gates as investment decisions. The question is not whether a team has crossed an impressive counter; it is whether the evidence supports the next increase in spend, scope, or operational complexity.
Identify what each stage must demonstrate
| Stage | Primary work | Evidence before the next investment |
|---|---|---|
| 0 → 1 | Reach one relevant person and deliver one useful experience | Observed understanding and a working next step |
| 1 → 10 | Repeat the experience manually with similar people | Recurring need and concrete feedback |
| 10 → 100 | Improve onboarding and document one distribution route | First-value completion and early return behaviour |
| 100 → 1,000 | Develop a repeatable offer and acquisition routine | Cohort quality, delivery capacity, realistic costs |
| 1,000 → 10,000 | Strengthen a proven channel and useful content system | Retention, contribution, and stable measurement |
| 10,000 → 100,000 | Add distribution partnerships and operational capacity | Incremental audience, quality control, channel resilience |
| 100,000 → 1 million | Expand formats, markets, or channels selectively | Sustainable economics and value at the larger scale |
The stages are a planning model, not empirical growth laws. A change in market, distribution, or product can invalidate earlier evidence. Treat an expansion into a new audience as a new hypothesis rather than assuming the original funnel will transfer unchanged.
Better evidence makes the next decision more useful.
Model stocks and flows correctly
Registered accounts are a stock. New signups are a flow. Active users are defined by behaviour within a window. For mutually exclusive categories, next-period active users come from retained actives, newly activated users, and reactivated users. Build those definitions before predicting the total.
In an illustrative cohort of 1,000 active users, 800 return, 250 new users activate, and 50 previously inactive users reactivate: next-period actives are 1,100. That is 10% growth, not the 30% suggested by counting additions alone. The return and activation assumptions need evidence.
Find the binding constraint
Early constraints may be unclear demand or poor first value. Later constraints may be acquisition cost, sales capacity, delivery margin, retention, or market saturation. Diagnose the constraint before adding another channel. More top-of-funnel volume can increase losses when the downstream system is not viable.
Use contribution and capacity scenarios. Include downside assumptions, variable support work, and the cash timing of acquisition versus revenue. Avoid funding an expensive growth plan with optimistic lifetime-value estimates from an immature cohort.
Build a repeatable learning and distribution system
Connect customer questions to content, content to a relevant action, and actual use to the next product or editorial decision. Preserve attribution limits. A final-click report is not a complete model of why a customer chose you.
Expand through a channel or partner only after you can explain the mechanism and maintain quality. Document the parts that relied on founder attention. As the audience grows, editorial review, support, moderation, and data governance become capacity requirements rather than optional polish.
Use a scale decision memo
Before the next investment, record the current evidence, the mechanism you expect to repeat, the assumptions most likely to fail, and the guardrail that would stop expansion. Name who owns the review and allow enough time for the relevant customer outcome to appear.
A goal of one million can help align ambition. The useful management practice is a series of bounded decisions that keep acquisition, customer value, and economics connected. Choose a smaller sustainable outcome if that better serves the business and its customers.
