Nobody decides to stop engaging their audience

Part of a single argument: the engagement gap is a workload problem.

There is never a meeting where a team resolves to make its sessions duller. No one writes it down, nobody approves it, and if you asked at any point during the decline whether engagement mattered, everyone present would say yes and mean it.

The decline is made of individually sensible decisions, each defensible in the week it was taken. That is precisely what makes it hard to see, and harder to argue against. There is no bad decision to point at.

How it actually goes

Week one, someone builds the full version. Questions at the start, a poll in the middle, a proper follow-up. It goes well. People mention it.

Week four, the same person is behind on something else, so they reuse last month's deck and skip the poll. The session is fine. Nobody complains, because nobody knew there was going to be a poll.

Week nine, a colleague copies the deck to run their own version. They copy the week-four one, because it is the most recent. The poll is not in it.

Week twenty, the poll is gone from the team template entirely, removed by someone tidying up who noticed it had not been used in a while and reasonably concluded it was not wanted.

Nothing in that sequence is a mistake. Each step is the right call given that week's constraints and the information available to the person making it. The outcome is still a session that has quietly become a broadcast, and an organisation where the engaged version no longer exists to copy.

The ratchet only turns one way

The important asymmetry is that dropping the engaging element is cheap and restoring it is expensive. Dropping it takes a decision not to do something. Restoring it means rebuilding the activity, remembering what it was for, testing it, and reintroducing it to a team that has got used to its absence.

So the weekly pressure, which is small but constant, only ever pushes in one direction. The restoring force is large, rare, and requires someone to notice and care enough to spend a block of time on it. Over a year, a small constant force beats a large occasional one.

This is why teams that have drifted rarely drift back on their own, and why the recovery, when it happens, tends to look like a project with a sponsor rather than a return to normal.

Why nobody notices

Each individual session is only slightly less engaging than the one before. The difference is below the threshold at which anyone would remark on it, and there is no moment where the change is visible.

Feedback does not catch it either. Session ratings are famously insensitive to this, partly because attendees rate against expectations that decline in step with the sessions. If every session for eight months has been a person talking over slides, a person talking over slides is what a session is, and it scores adequately.

The people best placed to notice, the ones who built the engaging version originally, have usually moved on to other work by the time the drift is visible. Their successor inherits the current template and has no idea it was ever different.

The thing worth measuring

If you want an early warning, do not track engagement scores. They are lagging, noisy, and they move for reasons unrelated to what you did.

Track how long it takes one of your people to prepare an interactive session from scratch. That number is leading rather than lagging: when it goes up, the drift has already started, and it will show in the sessions a couple of months later.

It also has the useful property of being a number about your systems rather than about your people, which changes the conversation. Nobody feels blamed by it, and it points at something you can actually change.

The fix is structural, not motivational

The reliable intervention is not to ask people to try harder. It is to make the engaged version the cheap default and the flat version the one requiring a decision.

If the template arrives with the activity already in it, and keeping it costs nothing while removing it costs a deliberate act, then the weekly pressure now pushes toward engagement rather than away from it. Same people, same time constraints, opposite outcome, because the direction of the ratchet flipped.

This is unglamorous work. It is templates, defaults, and libraries rather than vision and enthusiasm. It is also the only version of the fix that survives a bad quarter, which is the only test that matters, because the whole problem is what happens to your intentions when the quarter goes bad.

What to do this month

Find the most recent version of your team's standard session. Compare it with the oldest version you can locate, ideally a year or more back. The difference is your drift, and it is usually more than people expect.

Then look at what was removed and ask, for each item, whether it was removed because it stopped being useful or because it was inconvenient at some particular moment. The second category is your recovery list, and it is normally short.

Put those items back into the template rather than into a plan. Plans decay at about the same rate as intentions. Templates get copied.