How do you improve user onboarding?

Activation experiments shorten the path to first value. The two moves that reliably work are removing steps between arrival and value, and getting the user to the value before asking for anything in return. Time to first value is the metric to drive down. Every step before the value is a place users leave, and most onboarding is longer than it needs to be.

What is time to first value?

Activation experiments shorten the path to first value. The two moves that reliably work are removing steps between arrival and value, and getting the user to the value before asking for anything in return. Time to first value is the metric to drive down. Every step before the value is a place users leave, and most onboarding is longer than it needs to be.

How do you increase activation rate?

Activation experiments shorten the path to first value. The two moves that reliably work are removing steps between arrival and value, and getting the user to the value before asking for anything in return. Time to first value is the metric to drive down. Every step before the value is a place users leave, and most onboarding is longer than it needs to be.

Activation and onboarding experiments · Product Experimentation

§The question: what is the shortest path from arrival to the first real value, and what stands in it?

§Activation was the first validation signal; now it is a growth lever. showed how many users reach first value; this chapter is about raising that number by making the path shorter, because the path is almost always longer than it needs to be.

§ 12.1Time to first value#

§The metric is time to first value: how long, and how many steps, from arrival to the activation moment. Wes Bush's -led growth makes this the central number, and the logic is the funnel's: every step before the value is a stage that leaks, so the fewer steps, the more users survive to the value.

§Drive it down. A user who reaches value in one screen activates at a higher rate than one who must pass through five, and the difference is not marginal, because leakage compounds across steps: five steps that each keep 80% keep only a third overall.

§ 12.2The two moves that work#

§Remove steps. Audit the path from arrival to value and delete everything that is not required to reach it. The account fields that can be collected later. The tutorial that explains what the user could discover. The configuration that could have a default. Samuel Hulick's onboarding teardowns return to this constantly: most onboarding is the team's need to explain, imposed on the user's need to get value, and the user's need wins.

§Deliver value before asking for anything. Let the user reach the aha moment before the sign-up wall, before the payment, before the profile. A user who has felt the value will complete the form; a user asked to complete the form first, on a promise, leaks. The reverse-the-order move, value then , is the single most reliable activation experiment.

Weak onboarding	Strong onboarding
Sign up, verify email, complete profile, take tour, then reach the product.	Reach the product and get one real result, then be asked to sign up to save it.
Configure settings before first use.	Sensible defaults, first use immediately, settings when the user wants them.

§ 12.3The anatomy applies#

§Each onboarding change is an experiment. : removing this step raises activation, because the funnel showed drop-off there. Metric: activation rate for the cohort that got the shorter path. Threshold and rule set in advance. And a control, because onboarding changes are where teams most often ship a change, see the next cohort do better for unrelated reasons, and credit the change.

§The instrumentation matters especially here: to know where users leave, you need the funnel step by step, which means events on each step of onboarding, from chapter 17. Without them, you are guessing which step to remove.

§ 12.4Watch, do not just measure#

§The number tells you the step where users leave; it does not tell you why, and the why decides the fix. A step with high drop-off might be confusing, might be asking too much too early, or might be where users realize the product is not for them, which is not an onboarding problem at all. The field sprint's discipline returns: watch a handful of new users go through onboarding, in the qualitative sense of chapter 4, because five screen recordings will show you the confusion that the funnel only locates.

§ 12.5What you leave with#

§A measured path from arrival to first value, step by step, with the leakiest steps identified. Experiments that remove steps and move asks after the value, each with a control. A qualitative read of why users leave at the worst step. And time to first value tracked over cohort as the activation number to drive down. The next chapter is the stage everything rests on: keeping them.

12. Activation and onboarding experiments