Raise activation.

Don’t start with an onboarding redesign. First, make sure the activation event does predict further value, find the segment and step with the loss, and then figure out the cause: an obscure promise, excess effort, technical error, or lack of a suitable scenario.

What is activation and why raise it

Definition

Activation is the first valuable experience that a user has, and if they have taken a key action (e.g., created the first project, placed an order), then activation has taken place, most often it is a transition from “registered” to “understood and tried the value of the product.”

Why activation is important

High activation predicts retention. The higher the conversion of new users to activated users, the more likely they are to stay and pay. If there are gaps in activation, it is useless to spend money on attraction, people leave without even realizing what your product can do.

Case from practice.

Adding clues alone does not guarantee activation: they only help where the user is already motivated but does not understand the next step.If the problem is in the promise of value, a technical error or the wrong segment, the extra layer of learning will increase the noise.

Quick Actions to Increase Activation

Check the key path: whether it breaks

First, check how the user is going through the first steps. At the start, there are often bugs, non-obvious places or technical failures.

Example: Two screens after registration, the payment form hangs indefinitely, the user can not complete the activation.

What do you do? Go through the beginner’s journey completely yourself and ask your colleagues to try different devices. If you find stoppers, fix them first.

Take the extra steps off the start.

The fewer fields and complex solutions, the better, and every extra screen, optional question, or complex requirement kills a portion of the conversion.

Example: the team removes fields that are not needed before the first result and saves them when the user understands the reason for the request. The test compares not only the completion of the registration, but also the quality of the subsequent scenario: less data at the input should not impair the personalization or success of the operation.

** Recommendation:**
Leave only what is needed to get the first value.

Include transparent tips and welcome scripts

Most users don’t read the instructions, and the tips and visual markers immediately show what’s important to do.

Example: Illuminated the button “Try now” immediately after registration, activation on this step increased.

How do you do that? Add learning steps only at key points, instead of longread, make one-line calls and pop-up prompts.

Tests and hypotheses that work quickly

A/B test the activation step option

The text, position, and visual priority of a CTA should be changed when there is a hypothesis that the action is not understood or invisible. Local click-through growth does not equal activation growth: check whether the user has reached the result and whether random actions have increased.

Remove paywall or registration for the test period

Often people drop a product on a paywall or sign-up screen: Let’s try the main script first, ask for an email or pay after.

Add a small bonus at the start.

Free test period, discount, points are an easy way to motivate you to take the first step.

Example

In a mobile note-taking app, a new user can create the first entry from a short template without setting folders and tags, which shortens the time to a result; separately checks whether people return to their own note after deleting test content.

Typical errors and traps

Long onboarding

Don’t make the user watch a 5-screen tutorial before they try anything.

Require e-mail or card before the first action

Collect data only when the user has already seen the value.

No feedback on the result

The user has done an action – show that it worked. No answer – part will leave, deciding that it doesn’t work.

What and how quickly to measure

The main metric is activation

In standard products, it is measured by the formula: Number of activated / Number of registered (or started onboarding)

Activation is considered successful if the user has reached a key event (for example, saved the first file, created a task).

Tracking in analytics

Set up an event in Amplitude, Mixpanel, Yandex.Metrica, divide the path into stages and watch where people fall off most often.

FAQ: Quickly on the main thing

What is an activation in my product?

This is the step in which the user gets the first valuable result – finishing an important scenario.

How long does it take to activate a new user?

On average, 1-3 days from the date of registration, but depends on the script and the product.

Do I need to ask for user data before activation?

Only the minimum necessary for the first action is desirable.

How quickly can you see the effect of change?

Most often, you can see an increase in the metric after 2-7 days after the change is launched.

Where to look at best practices for activation?

Learn about Growth Design (growth.design/case-studies), Amplitude and Lenny’s Newsletter.

What other metrics do you track besides activation?

Often watch retention of the first day, the path to payment, the behavior of segments of new users.