There are two types of growth. The first runs on input: you spend on ads, you get customers. Stop spending, stop growing. The second runs on itself: customers bring more customers, data improves targeting, product improvements drive retention that funds acquisition. The second type is a loop.
Most businesses think they have loops. Most actually have funnels with wishful thinking attached.
What makes a loop a loop
A genuine growth loop has three properties. First, the output of one cycle becomes the input of the next — not just metaphorically, but mechanically. Second, the loop improves with scale: more users generate more data, more referrals, more content, more social proof. Third, the loop has a measurable multiplier: every customer acquired at the top produces some fraction of another customer downstream.
When that multiplier exceeds 1.0, you have compounding growth. When it's below 1.0, you have a funnel that needs constant top-of-funnel fuel.
Identifying your loop
The first step is tracing backward from your best customers to understand how they arrived. Not the channel — the mechanism. Were they referred? Did they discover you through content a previous customer created? Did a case study from your retention work drive a new enterprise inquiry?
Most businesses have a latent loop that's never been intentionally designed or measured. Making it explicit — naming the mechanism, instrumenting the handoffs, removing the friction at each stage — typically 2–4x the natural compounding rate.
The multiplier math
If your loop multiplier is 0.3, every 10 customers you acquire generate 3 more downstream. At a multiplier of 0.7, you generate 7. The difference in CAC over 12 months is not linear — it's exponential. A 0.7 multiplier business can sustain growth on half the acquisition spend of a 0.3 multiplier business at the same revenue target.
This is why two companies with identical products, identical teams, and identical budgets can diverge so dramatically in growth trajectories. The one with the better-designed loop doesn't just grow faster. It gets cheaper to grow over time.
Building the loop is a design problem. Measuring the multiplier is a causal one. Both matter equally.