Seeking Alpha in Growth

Growth as a search for mispriced channels

The growth organization Eric Glyman built at Ramp treats the job less as executing a known playbook and more as finding what the market has mispriced. George Bonaci, Ramp's VP of Growth, borrows the word "alpha" from investing, where it names excess return over a benchmark, and applies it to distribution: a growth team's unfair advantage is "something that is not saturated and that other people ideally are not doing."¹ The framing reclassifies growth work. Anything standard is, by construction, stale, because if a channel were obviously good it would already be crowded.

Three places the edge hides

Bonaci names three sources of alpha. The first is the unknown, doing things no one knows about yet, usually because they are new. The cited example is TikTok at launch, when no business-to-business brand was advertising there and the ad inventory barely existed; the teams that anticipated a B2B window opening before it saturated with consumer brands captured outsized returns. This is the acquisition side of Marketing Novelty Decay: reach the channel before the edge is competed away.

The second is the disbelieved, doing things everyone is convinced will not work. The worked example is direct mail, proposed years ago as "junk mail to people's homes, that absolutely won't work," which after a few iterations became one of Ramp's biggest channels.¹ The stated logic is that the disbelief of others is itself the moat, because a tactic that looked obviously good would already be saturated.

The third, and in this account the most defensible, is the cross-niche: borrowing from another vertical or geography. Peers in your own space already know the local playbook, so the richest vein is knowledge that is real but un-transferred. The example is WhatsApp, a minor marketing channel in the United States but a major one internationally, prompting the question of whether it could replace email. Any single cross-pollinated bet will probably fail, but run enough of them and one works that no competitor in your category is running.

How the edge is found

The framework maps those three sources onto three modes of learning. Learning academically, from books and past companies, is described as adapting old playbooks rather than copying them; the cited case is marketing mix modeling, a 1950s and 1960s method for measuring newspaper-ad impact now re-adapted to digital attribution. Learning from peers is called the weakest form of alpha, since someone else already knows it. Learning from other niches and geographies is called the richest, for the same reason it is defensible.

The anatomy of a good bet ties this back to the rest of Ramp's practice. Direct mail qualified on two axes at once: no one else was doing it, and it was "incredibly scalable with large sample sizes," since "there are very few channels where you can go tomorrow let's reach 200,000 people."¹ Contrarian plus high experiment throughput is what makes an alpha bet workable, which is why the framework presupposes the machinery of Growth as Experimentation. Once alpha is found, the discipline is to Saturate the Winning Channel, taking it to the asymptote fast before it decays.

The tensions the framing carries

Bonaci concedes several limits. Cross-niche borrowing has a low hit rate, so it only pays off under high experiment throughput, which not every team has. The examples also invite survivorship bias: direct mail and the TikTok-first-mover are remembered because they worked, while the disbelieved bets that stayed disbelieved and wrong are not counted. And the framework is more visible than any specific tactic, since a live alpha is by definition unshared; the tactics that get named in public are the ones already going stale.