Member disengagement leaves a trail. Here's how to spot it before renewal season.

You probably know which members are engaged. The harder question is whether you’d know, right now, which ones are silently on their way out.

Engagement tends to be measured at its most visible points: event registrations, renewal completions, qualification, volunteering, new joiners. What sits between those moments, the gradual drift, the missed webinar, the newsletter that stopped being opened, rarely gets the same attention. And that’s precisely where the signal is.

The challenge isn’t usually a lack of information. It’s that the information lives in different places: attendance in one system, email behaviour in another, web access somewhere else entirely. Nobody assembled the full picture because nobody had a reason to, until it was too late to act on it.

The moment that’s already too late

Renewal is not the moment a member decides to leave. It’s the moment they confirm a decision that’s been forming for months. By the time an invoice goes unpaid, the decision to leave has already been made. The organisation is simply the last to know.

The window for intervention sits well before that point, when the member is still reachable, still open to a conversation, still connected to what the organisation offers. The closer you get to renewal, the narrower that window becomes. A generic reminder email at 30 days is not retention work. It’s confirmation that no retention work happened.

The MemberWise Digital Excellence Report 2026/2027, drawing on responses from circa 480 UK membership professionals across professional bodies, trade associations and chartered institutes, found that the inability to measure member engagement has become the sector's number one challenge. Nearly four in ten organisations don’t measure engagement at all. This has been a pattern for some time.

What disengagement actually looks like in your data

Not all CRM data is equally useful for spotting who’s at risk. Job titles and employer records matter for segmentation. Payment history matters for finance. For early warning, the signals that count are behavioural: what a member is doing, or increasingly, not doing.

The most reliable indicators are shifts in pattern rather than the absence of activity. A member who attended regularly and then stopped carries a different kind of risk from one who has never attended at all. The change is the signal. Four or five behavioural signals taken together tend to give a clear enough picture: event registration and attendance gaps, email engagement trends over time, web login frequency, learning content interactions for professional bodies, and dropped committee or volunteer roles. Active participation is one of the strongest predictors of renewal, so its absence is worth watching.

None of these signals is conclusive on its own. A member who missed one event might have had a diary clash. A member who missed four, stopped opening emails, and hasn’t logged in for six months is telling you something quite specific.

Why the picture is hard to see

The data exists. In most membership organisations, it’s been accumulating for years. The difficulty is that it’s spread across systems that weren’t designed to talk to each other, which means assembling a complete view of any one member’s engagement requires someone to go and find it. That takes time teams rarely have, and it tends to happen only when a concern has already been raised.

The members who drift quietly, without flagging themselves as at risk, are exactly the ones this approach misses. They don’t raise concerns. They just stop engaging. When engagement data from the website, the email platform and the member portal flows into a single member record, the question changes from how to find out whether a particular member is at risk to which members the system is already flagging. That’s a different kind of work, and a more manageable one.

What to do before the renewal window closes

The aim here isn’t a sophisticated churn prediction model. For most membership organisations, the more useful step is a small number of consistent behavioural triggers that prompt a human conversation before the renewal window closes.

A process that flags members who haven’t engaged across two or more channels in 90 days gives a team a working list to act on each month. It doesn’t need to be a long list to be a valuable one. Catching five or ten members per month in the window where re-engagement is still possible, and making a well-timed call or sending something genuinely relevant to them, will have a more measurable effect on renewal rates than a standard reminder sent three weeks before expiry.

Across the sector, the guidance on timing is consistent. Starting renewal communications 90 days before expiry, rather than the standard 30-day prompt, gives a team meaningful time to act on the members who need more than a reminder. That lead time is only useful, though, if you have a view of which members need more than a standard message.

Behavioural data gives you that. It’s the difference between treating every renewal the same and knowing, in advance, which relationships are worth a phone call.

Where AI comes in

For organisations already working in the Microsoft ecosystem, this kind of identification work is exactly what Copilot in Dynamics 365 is built for. Rather than building a report (which may evolve and change), a membership team can ask Copilot in plain language to surface members who haven’t engaged across specified channels within a defined period. It returns a list. Someone makes a call.

The important qualification is that Copilot works with the data it can see. If event attendance sits in a separate system that doesn’t connect to Dynamics, Copilot can’t include it. The same applies to email engagement, portal logins, or any other signal that lives outside the connected environment. Copilot amplifies good data infrastructure. It doesn’t substitute for it.

This is why the data and integration question comes first. Once the data is in one place and flowing correctly, identifying at-risk members becomes something the system does for you rather than something your team has to search for.

The cost of leaving it as it is

Acquiring a new member costs significantly more than retaining an existing one. For example, an organisation with a steady 85% renewal rate might feel reasonably comfortable. When applied to a membership of two thousand, that’s three hundred members to replace each year, all of whom need finding, attracting, onboarding and re-engaging before they contribute at the level of the members who left.

There’s a quieter cost that rarely appears in the retention figures: the members who were close to lapsing and didn’t, because someone reached out at the right moment with something relevant. Those conversations only happen when you can see the signal early enough to act on it. Without that visibility, they don’t happen at all.

Where to start

If behavioural signals aren’t currently part of how your team reviews member health, the starting point isn’t a full system overhaul. It’s three questions about what your current data can tell you.

1. Can you see, against a member record, whether someone attended the last event they registered for? 
2. Can you see email engagement history alongside membership status?
3. Can you identify, right now, which members haven’t logged into your website in the last three months?

If yes, well then, you have a foundation to build from. If the answer is no, or it depends on which system you look at, the gap isn’t in your team’s approach to retention. It’s in how your systems share information.

If you’d like to talk through what this looks like for your organisation, we’re happy to start with your current setup and work from there.

Get in touch today to discuss how we can support your next project.

  Find us on LinkedIn and connect!