Leading and lagging indicators

Membergate Support - April 30, 2025

By the time your monthly churn figure arrives, the members it counts have already gone. By the time your quarterly revenue is in, the quarter is over. These are the numbers that matter most to your business, but they all report on the past. If they are the only numbers you watch, you will always be reacting to problems weeks or months after they began.

The fix is to watch two kinds of numbers side by side. Lagging indicators tell you what happened. Leading indicators give you an early hint of what is about to happen, while there is still time to change it.

What makes an indicator leading or lagging

A lagging indicator measures an outcome. It is the result you actually care about, but it is only known after the fact. For a membership, lagging indicators include:

Total paying members
Recurring revenue
Churn and renewal rates
Lifetime revenue from a group of members

A leading indicator measures an earlier activity that tends to move before the outcome does. It is less important in itself, but it arrives in time to act on. Examples include:

New members reaching an early point of value, such as finishing a first lesson
How many members log in or take part each week
Attendance at live events
Email sign-ups, trial starts and visits to your pricing page
Support complaints and requests to cancel or pause

The same figure can be leading for one outcome and lagging for another. Trial starts are an outcome for your marketing, but a leading indicator for next month's new members.

Pair every outcome with an early signal

For each lagging figure you care about, choose one leading indicator that plausibly drives it:

Renewals and churn: early engagement from new members. The idea of an activation moment is exactly this kind of signal.
New members: email sign-ups, trial starts and pricing page visits.
Revenue per member: interest in higher levels or add-ons, such as visits to the upgrade page.
Cancellations: complaints, unresolved problems and long gaps between logins. Tracking support requests as a business metric gives you one of these for free.
Word of mouth: how likely members say they are to recommend you, collected by asking the recommendation question.

Test your leading indicators against your own history

A leading indicator is only useful if it genuinely moves before the outcome. Before relying on one, check it against your own past data: did groups of members who scored higher on the early signal go on to have better outcomes?

Take a hypothetical resource membership for homeschooling parents, run by a former teacher named Tamsin. The figures are invented to illustrate. Looking back over a year, she finds that new members who download at least two lesson plans in their first two weeks renew at their first renewal about 80% of the time, while those who do not renew about 40% of the time. The early downloads reliably come before the renewal outcome.

Now she can use the signal. Her January group of 200 new members included 140 who downloaded two plans in the first two weeks. Her April group of 200 included only 90. Using her own rates, she can estimate the renewals each group will produce:

January: 140 × 0.8 = 112, plus 60 × 0.4 = 24, for about 136 renewals.
April: 90 × 0.8 = 72, plus 110 × 0.4 = 44, for about 116 renewals.

Tamsin sees this gap in mid-April, not months later when the April group's renewals come due. She traces it to a redesign of her welcome page that buried the lesson plan library, fixes it, and has time to send the April group a personal note pointing them to the plans.

Traps to avoid

Vanity signals. Social media followers and total page views are easy to watch and rarely predict anything about members. Choose signals that sit close to the value members pay for.
Gaming the signal. If Tamsin forced every new member to download two plans before seeing anything else, the indicator would rise without any real change in value. Once a leading indicator becomes a target, check that it still predicts the outcome.
Too many signals. Pick one leading indicator per outcome. Ten early signals produce ten alarms and no action.
Stale relationships. What predicts renewal can change as your membership changes. Retest your indicators once or twice a year.

Give each kind of number its own rhythm

Leading indicators change quickly, so look at them weekly, when there is still time to respond to a dip. Lagging indicators move slowly and are noisy from week to week, so review them monthly and judge them over a quarter. A useful weekly check takes ten minutes: three or four leading figures compared with the previous few weeks, and a note of anything unusual to investigate.

Setting up your indicators

List the three lagging outcomes that matter most to your business.
For each, choose one leading indicator you believe drives it.
Check each pairing against your own history: do members who score higher early do better later?
Drop any leading indicator that does not predict its outcome, and try another.
Review leading indicators weekly and lagging indicators monthly.
When a leading indicator dips, investigate and act that week, then watch whether the outcome holds.