Culture of Doers

The organizing premise

Adam Foroughi describes AppLovin's operating model as one in which every person creates direct value, with no pure management layers and no processes that cannot survive a "why" test.¹ The premise he states is about clustering: "A players can exist in any role. But they won't exist in bulk if you have a bunch of B's, C's, and D's around them." In his telling culture is self-selecting, because A-players have options and leave mediocre environments, so a culture perpetuates in whichever direction it starts. A corollary he draws is that A-players cannot be sprinkled into a weak culture to fix it; the social dynamics either push them out or drag them down. This overlaps with the platform's A-Player Framework.

What it looks like in practice

Foroughi cites several concrete implementations. He describes an executive team of four, CEO, CTO, CFO, and General Counsel, with no COO, CRO, CMO, or CHRO, arguing each absent role would require someone who manages more than they build. He says HR was cut from roughly 70 to 80 people down to about 15, keeping individual contributors rather than "business partners" who facilitate process. He reports no scheduled one-on-ones, with feedback happening in real time, on the logic that anything needing a scheduled meeting to surface is already too slow. He describes no formal learning-and-development, on the premise that the best people arrive curious and self-teach, and no separate product organization, summarized as "either your product people become engineers or your engineers become product people, you don't need both."¹

The unifying test he attributes to a CTO, Giovani, is a single repeated question: why does this person, process, or role exist? He compares it to the pruning logic of asking why a product feature exists, applied instead to the organization. Things that cannot answer the question well, he says, get removed, which he claims made the organization faster.¹

The proactive reset

The most distinctive data point Foroughi offers is a 2024 reduction of roughly 40 to 50 percent of staff across most departments during a year of triple-digit growth, which he frames not as a response to financial distress but as a proactive reset.¹ His stated rationale was to rebuild the organization "as if we were building it knowing what technologies are available to us today," letting go of roles he judged automatable or slow to adopt AI. The effects he claims are faster AI adoption among those who remained, removal of the people most likely to resist it, and an immediate lift in the average quality bar. This aligns with his separate view that a lean organization is the design goal rather than a temporary constraint.

The stated remedy and its edges

Asked how to fix a culture that has already gone mediocre, Foroughi gives an unusually aggressive answer: "the only way to fix a culture like that is to go and fire 99% of people and just rebuild from the ground up."¹ His argument is that half-measures fail because the A-players have already left, so a partial culling leaves the best of a weak pool rather than a representative sample. He also states a permanent hiring preference against MBA and big-company profiles, favoring hunger and intrinsic drive, and describes promoting people several levels at once when they earn it.

Foroughi presents the model as effective at AppLovin's specific circumstances and does note tensions with adjacent views, including Brad Jacobs, whom he positions as focused on identifying and retaining A-players rather than mass removal of everyone else. The framework is stated as a personal operating conviction backed by AppLovin's results, and Foroughi himself flags that its most radical claim, tearing a culture down to rebuild it, is more extreme than the surrounding consensus.