Saturate the Winning Channel

The mistake is timidity, not recklessness

The common failure when a growth experiment works, according to the growth organization Eric Glyman built at Ramp, is caution. Asked what to do the moment an experiment works, George Bonaci, Ramp's VP of Growth, is blunt: "absolutely you should triple down." The common mistake is teams seeing a winner and politely "increase spend, double it, triple it," when the correct response is to take the channel to saturation as quickly as possible.¹ The stated conviction is that most teams reach the asymptote too slowly, and that "most things saturate more slowly than people expect," so a jump that sounds insane on paper, such as 10,000 dollars to 200,000 dollars of spend, is often less reckless than the gradual ramp it replaces.

The counterweight is that the crude version of the aggressive move does fail. Jumping from 10,000 to 200,000 overnight without instrumentation means "the 200,000 will not be nearly as efficient as that 10,000." The claim is not that scale is free; it is that the safe 10, 30, 50, 70, 200 ramp leaves value on the table when the curve could have told you to move faster.

Read the response curve, not your gut

The discipline that separates aggression from recklessness is measurement. The instruction is to graph results over spend or time and watch the shape approach an asymptote as incrementality decays. While the curve is linear, each unit of input still returns its expected output, so the guidance is to keep pushing. When the curve bends from linear into decay, "we're no longer getting an expected output given the input," and that inflection, not intuition, is where the conversation about marginal returns belongs.¹ The bend is the stop signal.

This connects the payoff step to the rest of Ramp's growth practice. Seeking Alpha in Growth finds the unsaturated channel, Prioritize on Confidence and Time-to-Results greenlights the bet, and saturation extracts its full value before Marketing Novelty Decay competes the edge away. The response-curve mechanic is also the same one underneath Performance Marketing as Arbitrage: scale spend until the arbitrage breaks down, seen from the platform side.

Why saturation rarely bites the way theory says

The textbook objection is that customer acquisition cost rises as you take share, since each incremental customer is harder to reach. Bonaci's practitioner answer is that this macro ceiling usually arrives far later than expected, because before it bites a growing company typically finds new products to sell that lift lifetime value, new geographies to enter, new channels that work, and cross-channel halo effects it was not fully attributing to acquisition cost.¹ The stated conclusion is that "it takes a really long time to get to the point where the macro effect of saturation is hitting your CAC." The asymptote, in this account, is per-channel and per-configuration rather than a hard wall on the business.

A related caution attached to the same framework is that lifetime value is false precision early. The acquisition-cost-to-lifetime-value ratio is called "a reasonable framework," but a number computed after six or twelve months is not real, and treating it as real is the error. The suggested substitute is a moving threshold, an agreed statement of what a customer is worth and what the team is willing to spend, revised as evidence accumulates rather than defended as a fixed truth.

The limits the framing concedes

The framework carries its own qualifications. Graphing the response curve assumes clean attribution and enough volume to see the bend; for noisy or low-volume channels the inflection is hard to read in time, which weakens the whole method precisely where teams most want a signal. And the optimistic view of acquisition cost, that new products and geographies defer saturation, is a growth-mode claim. Under a profitability mandate, the rising-cost textbook case reasserts itself, and the license to saturate fast narrows accordingly.