Find the growth constraint before adding another growth tactic.

Growth systems: research and decision guide

Start with a growth equation, not a channel list

A growth plan should make explicit how customers enter, convert, remain and expand. Traffic growth can be irrelevant if conversion falls. Acquisition can accelerate while payback deteriorates. Strong new-logo growth can conceal weak retention. Expansion can lift revenue from the installed base while a narrow segment reaches saturation. The useful starting point is a simple system map: qualified demand, conversion, average revenue, retention, expansion and acquisition cost. Then identify which variable currently constrains the desired outcome.

Separate acquisition volume from acquisition quality

More leads or sign-ups are not necessarily better if they convert poorly, require expensive support or churn quickly. Segment acquisition by source, customer type and expected economics. A channel that appears expensive on first-touch CAC can be attractive if it produces larger contracts or stronger retention. A cheap channel can be destructive if it attracts low-fit customers. The objective is not minimum CAC in isolation; it is efficient acquisition of customers whose contribution and retention justify the cost.

Treat retention as part of the growth engine

Recurring businesses compound only when enough customer value survives from one period to the next. Stripe's NRR framework distinguishes churn, contraction and expansion within an existing customer cohort, which makes it useful for diagnosing why the revenue base changes. But the metric still needs decomposition. Product usage, service quality, customer selection, onboarding, pricing and contract structure can all affect retention. Growth teams should connect churn events to observable causes instead of treating retention as a customer-success metric owned by one function.

Design expansion around increased customer value

Expansion revenue is most defensible when customer value increases through more users, more usage, additional workflows, geographic expansion or higher-value capabilities. Upsell that is detached from value can increase short-term revenue but damage retention. A good expansion system identifies observable triggers that indicate broader need, gives the buyer a clear upgrade path and preserves price-to-value logic. This connects product instrumentation, packaging, sales and customer success rather than treating upsell as a separate campaign.

Align growth rate with unit economics and operational capacity

Fast growth can stress support, onboarding, infrastructure, cash and management capacity. Unit economics should therefore be evaluated at the growth rate the company is trying to sustain. If CAC rises as a channel scales, if implementation capacity becomes constrained or if infrastructure costs grow faster than pricing, historical economics may not extrapolate. The planning question is how marginal economics behave as the next cohort is acquired, not whether the average of all past customers looked attractive.

Run growth experiments against a causal diagnosis

Experiments are most valuable when they test a specific explanation for a bottleneck. If activation is weak, test onboarding or time-to-value. If conversion drops at enterprise procurement, test proof, security readiness or buying-process support. If churn clusters in one segment, investigate customer fit or unmet workflow needs. Define the expected mechanism before launching the experiment and choose a metric that reflects that mechanism. This avoids accumulating isolated wins that do not improve the end-to-end growth system.

Research sources

Related business and technology research ecosystem

Related model: When growth depends on cross-side participation rather than a linear acquisition funnel, see the Platform Business Model guide for network-effect and multi-sided-market economics.