Activation Stage

Signup counts used to be the whole report. Twelve hundred signups this month read as a good month. What that line never showed was that more than half of those 1,200 people cleared the registration form and left without doing anything. Signing up says a person decided to try the product. It doesn't say they tried it. The two get reported as one number because they arrive on the same day, from the same campaign, in the same row of the dashboard.

The gap became obvious whenever ad spend went up. Double the traffic and signups roughly double too, yet three months later the number of people still around barely moved. The new arrivals were disappearing without ever really opening the product, and with nothing between the signup count and the retention stage number, there was no way to point at where the loss happened.

Activation puts a gate in that space. Between acquisition and retention it asks whether this person has tasted what the product does even once, and only the yes answers move forward. With that gate in place, "1,200 signups" becomes "540 activated out of 1,200 signups," and the 660 people in between stop being invisible.

Half the work in this stage is nailing down the definition. Without one sentence saying what counts as activated, the signup count flows straight through as the activation count, and every number after it inflates. Pagedeck uses the moment a person creates and saves their first document.

Three things make a good criterion.

1. It has to be an action that puts the user through the thing the product is for. In a collaborative document tool, logging in isn't that experience. Creating and saving a document is.
2. It has to be an action that actually separates the people who stay from the people who leave. If users who make three or more documents in week one are 60 percent present two months later while those who make two or fewer are at 18 percent, that split is the magic number and the criterion belongs near it.
3. It has to be something that can be recorded. "Understood the product" cannot be counted.

This gate runs on a short clock. Acquisition happens in a day and retention is confirmed over weeks, but activation usually resolves within minutes to a few days of signup. That's why teams also measure how long it took from signup to the action — the time to value. Two products can both sit at 45 percent activation and grow at very different speeds if one averages four minutes and the other three days. The three-day product has to survive everything that competes for attention in between — another tool, a meeting, the end of the week.

Read the number as a rate carried over from the gate before it, not as a standalone figure. Pagedeck's last quarter ran 10,000 visits, 1,200 signups, and 540 first documents saved — 12 percent through the first gate and 45 percent through this one. Forty-five looks acceptable on its own, but it means more than half the people who finished signing up never created a single document, and that is the largest drop-off anywhere inside the product.

Moving on requires the activated user to come back on their own. If 320 of those 540 return a week later, that gate runs at 59 percent, and those 320 are what the retention stage works with.

So the goal here isn't a bigger activation number but more activations that will return. Auto-creating a document during registration takes activation to 90 percent overnight, yet a document the user didn't make is no reason to come back next week. The first gate rises and the one behind it doesn't move, which is the shape every shortcut in this stage takes.

The work that moves this gate is onboarding — designing the first screens and prompts so a new arrival reaches that one experience without wandering. The point it aims at is the aha moment, when a user sees what the product does for them specifically.

Getting more people through the same traffic is conversion optimization, and splitting users into two groups to check whether a change actually worked is a growth experiment.

"Signups are up but activation isn't following. We've been stuck at 45 percent for three months."

Adding traffic has stopped producing anything.

"This sprint we only look at the activation gate. That's where we leak the most."

Of the five gates, this one was picked as the biggest loss.

"Lowering the activation bar to logging in just fixes the number."

Pushing back on loosening the definition to make the metric move.

At Pagedeck, Mia puts all five conversion rates on one screen each quarter and decides where to start. Last quarter came back as 12, 45, 59, 28, 53. Six hundred and sixty people dropped out at the second gate, more than every later gate combined. Noa opened the post-signup screen and found a blank document with a blinking cursor — nothing that tells a first-time visitor what to do.

Noa's proposed change was to show three example documents and let the user open one and edit it. Ravi argued for moving budget out of ads and into this work, on the arithmetic that converting people already inside costs less than buying more visits. Two weeks later the gate read 52 percent.

The most common error is treating activation as another word for signup. Both look like counts of new arrivals, so the confusion is understandable. But a signup is a person who has experienced nothing yet, and an activation is a person who has experienced the thing once. Collapse the two and the next gate's conversion rate reads lower than it really is, the team concludes retention is broken, and work goes into the wrong place. The people actually leaving left one gate earlier, and no amount of work on re-engagement emails reaches them.

Treating the activation rate itself as the target is the other frequent trap. Lower the bar and the number rises immediately: swap document-saved for logged-in and 45 percent becomes 95. The next quarter gets planned on top of that figure while the count of people who actually used the product hasn't changed at all. If the bar does have to change, record the date it changed and read the periods on either side separately.