All signal, no noise: The rules of engagement

Engagement is easy to measure and surprisingly hard to define. Most teams start with the usual culprits: how often customers open the app, how long they stay, what they click. These numbers are useful, but incomplete. High activity doesn’t automatically mean your customers are better off.

Engagement is more accurately described as meaningful interaction. It’s whether customers can successfully resolve issues, understand their bills, and manage their energy in ways that reduce support calls, prevent payment issues, and increase retention.

That definition changes what “good” looks like in practice.

Take something as simple as a prompt in the app. A message that says “Check your usage” can easily be ignored. The same message, shown just before a bill lands or right after someone has already opened their bill once, often performs very differently.

For example:

“Want to review your usage?”
“Your next bill is coming up. Want a quick look at how your usage is tracking?”

Same feature. Same intent. Different outcome.

What makes the difference is not the metric you track, but the choices you make around timing, effort, progress, and value. Those are the levers teams actually control.

What makes a good engagement metric?

Not all metrics are created equal. Meaningful engagement metrics tend to answer three questions.

1. Does this action reflect real intent or progress?

Look for actions that suggest learning or commitment, such as setting alerts or goals, viewing personalised usage insights, or completing a self-service journey.

2. Does it correlate with better outcomes for the customer and the company?

In practice, this often shows up in patterns like app usage shortly after a bill is issued correlating with fewer support calls, or goal-setting linking to on-time payment.

3. Is it trackable, explainable, and improvable?

You should be able to measure it consistently, explain why it matters, and influence it through design. If you cannot change it through your product decisions, it is not a very useful metric.

At this point, many teams hit the same wall. They can see which behaviours matter, but translating that into better engagement in the app is harder than it sounds.

This is where the design patterns come in.

For instance, one pattern we see repeatedly is that reducing perceived effort matters more than adding new features. Framing a task as “3 quick questions” instead of “complete your profile” leads to very different completion rates. Showing progress toward a goal tends to work better than points or rankings. Leading with a concrete outcome beats explaining the feature behind it.

When you step back and look across an app, the same types of decisions show up again and again. How you ask for information. When you prompt someone to act. Whether you frame something as effort or progress. Whether the value is obvious before the ask.

On their own, each decision feels minor. Together, they shape whether customers follow through or quietly ignore what you put in front of them.

What we started doing was collecting these decisions as we saw them in practice. Not as abstract principles, but as concrete patterns. The kinds of choices teams make when they are trying to get customers to complete a step, understand a bill, or come back at the right moment.

That collection eventually became a short, practical guide. Four patterns that show up consistently in energy apps where engagement leads to action, each illustrated with real example copy taken from live customer experiences.

It is designed to be skimmed, shared, and used as a reference when reviewing flows, notifications, or onboarding journeys.

For those who want to go further, the same patterns are explored in more depth in an accompanying email course, looking at how they play out over time across onboarding, billing cycles, and seasonal moments.