The Habit Path

What it is

Eyal's definition, at page 203, is that you are looking for a habit path, a series of similar actions shared by your most loyal users. The inversion is worth stating plainly: you do not design the habit first and then go and find users for it. You find the users who already have the habit and reverse engineer what they did.

That makes it an observational method rather than a theory. The path is not proposed in advance and then tested, it is recovered from behaviour that already exists in the product, and the design work is putting new users onto the same track.

In effect

Eyal's worked cases run the search rather than hand over a finished path. The Bible App had tens of millions of regular users well before its hundred millionth install, at page 190. Fitbod's Allen Chen, who built the app after reading Hooked in 2015, is described across pages 192 to 197 as having found what his most habitual users already did and having built the onboarding to put new users on the same track.

The method's logic is sound on its own terms. If a cohort of users has already converged on a sequence of actions, that sequence is evidence about what the product is actually for, and onboarding that reproduces it is a better bet than onboarding designed from an internal theory of the user.

What it does not say

It does not say the supporting numbers hold. The gym industry framing that travels with this argument, a figure of nineteen billion dollars a year and a quitting rate of roughly forty four per cent within six months, traces to a consumer self report survey and to a trade body attendance table from 2001 respectively. Neither is a research finding and neither is a United States quit rate. They are recorded here as untraced rather than repeated as data, and the argument does not need them.

It does not say a found path is a causal one. Loyal users share a sequence; whether the sequence made them loyal, or whether loyalty produced the sequence, is not something the method distinguishes.

It does not say the path generalises beyond the users it was recovered from. A pattern taken from a product's most devoted cohort is by construction a pattern from the people who stayed, which is the standard survivorship problem, and putting new users on that track is a hypothesis to test rather than a conclusion.

---