How to reduce churn in the first 30 days

Most churn talk is about renewals, price and the competitor down the road. But a lot of the users who leave never got far enough to judge any of that. They signed up, poked around, got stuck somewhere in setup and quietly stopped coming back.

That kind of churn happens in the first 30 days, and it is mostly an onboarding problem. The good news is that it is also the kind a product team and a customer success team can do the most about, because the cause sits inside your own app.

Why early churn is an onboarding problem

A user who has never got real value from your product has nothing to lose by leaving. They haven’t built the report, taken the first booking or moved their team over. Cancelling costs them nothing.

So in the first month, the question isn’t “are they happy?”. It’s “did they get there?”. Two numbers tell you:

• Activation rate. Users who reached your activation event, divided by users who signed up in the same period, times 100. The activation event is the first action that shows a user got what they came for, such as the first project with a task in it.
• Time to value. How long a new user takes to reach that event, measured from sign-up. Take the median across a month’s sign-ups, and track the share who never got there alongside it.

The two move together: the longer the path, the more people stop on it. You can work out both with the free activation rate calculator and time to value calculator.

For example, say 400 people signed up last month and 140 reached the activation event within 7 days.

Example

sign-ups in the month

400

reached the activation event in 7 days

140

activation rate

35%

never got there, the churn risk

260

Those 260 are where most first-month churn comes from. Your renewal process won’t reach them, because they never got to the point of having an opinion.

How to spot it before they leave

You don’t need a churn model to see early churn coming. You need to know where in setup people stop.

1. List the steps between sign-up and your activation event, in order.
2. Count how many users reach each one. Most product analytics tools can show this as a funnel.
3. Look for the big drop. There is usually one step, sometimes two, where far more people stop than anywhere else.

For example, a funnel might look like this:

Signed up100%

Created a workspace82%

Connected their data41%

Invited a teammate36%

Activated35%

Show as a table

Signed up	100%
Created a workspace	82%
Connected their data	41%
Invited a teammate	36%
Activated	35%

In that example, “connected their data” is the problem. Polishing the welcome email won’t help much. Making that one step easier will.

Watch your support queue too. The “how do I…” questions that arrive in the first week are the same steps, told in your users’ words.

What to change in the first session

The first session decides whether there is a second one. Aim to get each new user to one real result before they close the tab.

• Ask less at sign-up. Every field you can ask later, ask later.
• Start them somewhere useful. Sensible defaults, a sample project or a template beat an empty screen.
• Help at the step where they stall, not on a tour of everything. If your funnel says the data connection loses half your users, that is where help belongs.
• Let them say what they want. A new user often knows the goal (“set up my first project and invite Maya”) but not the path. Meeting them at the goal skips the learning.

What to change in the first week

The first week is about the second and third result, and about pulling in the people who make the product stick.

• Nudge from what they did, not the calendar. An email that says “you created a project, here’s how to invite your team” beats one sent on day 3 to everyone.
• Get a teammate in. Products used by one person are easy to drop. An invite is often the step that turns a trial into a habit.
• Answer the first questions fast. A stuck user on day 2 who waits a day for a reply may not come back to read it.

What to change in the first month

By the end of the month, the users who stay should be using the product for real work. This is where customer success earns its keep.

• Review the users who didn’t activate. Not to chase them, but to learn which step stopped them. Feed that back into the first session.
• Check time to value after each change. If you shortened setup, the median should drop. If it didn’t, the change didn’t reach the step that mattered.
• Turn repeated questions into product changes. Ten tickets about the same setting is a design problem, not a support one.

1. Step 1First sessionOne real result before they leave›
2. Step 2First weekA second result and a teammate in›
3. Step 3First monthLearn from who didn’t activate

The customer onboarding guide covers each stage in more depth.

Where an in-app AI onboarding agent fits

Most of the advice above comes down to one thing: fewer steps between sign-up and first value, and help exactly where people stall. An in-app AI onboarding agent goes at that directly, by doing the steps for the user.

Stiro is one. It lives inside your product as one script tag. A new user says what they want set up, typed or spoken, in their own words, and Stiro clicks, types and navigates your real interface to do it while they watch.

• It meets users at the goal. “Connect our Slack workspace” or “invite the design team as viewers” is a sentence for them and a few clicks for Stiro.
• There are no flows to record. It reads the live screen, so when you change setup, it works from what is on the page now.
• The user keeps the final say. Anything that sends, spends or deletes, such as sending an invite, is set up and handed back for them to press.
• It says so when it can’t. If the request isn’t possible in your app, it stops, says why and changes nothing.
• Stuck goes to your team. A signed-in user can message your team from the same panel, with what they asked already attached, so a first-week question gets to customer success instead of going unasked.

We tested it cold in three open-source apps it had never seen, and it passed 44 of 44 everyday requests, each checked in the app’s own database. The benchmark lists every task and reply.

Start with the step that loses the most people

Find the one setup step where your new users drop, and make it easier first. If that step is something users could just ask for, AI onboarding with Stiro lets them ask, and the free plan includes 50 runs a month on localhost and a staging address, so you can try it on your own setup flow before your users see it.