📣 Your dashboard is lying to you

That's the AI paradox hiding in your CX stack. Tickets close. Customers leave. And most teams don't see it coming because they're measuring the wrong things.

Efficiency metrics look great on paper. Handle time down. Containment rate up. But customer loyalty? That's a different story — and it's one your current dashboards probably aren't telling you.

Gladly's 2026 Customer Expectations Report surveyed thousands of real consumers to find out exactly where AI-powered service breaks trust, and what separates the platforms that drive retention from the ones that quietly erode it.

If you're architecting the CX stack, this is the data you need to build it right. Not just fast. Not just cheap. Built to last.

See the data

👋 Hey friends!

Let me tell you about a trap almost every PM falls into at least once.

And then six months later you realize the thing you were measuring had almost nothing to do with whether your product was actually working.

That’s not a data problem. That’s a metric design problem. And it’s way more common than anyone admits.

This week I want to break down what a good metric actually looks like, how to tell the difference between one that’s meaningful and one that just makes you feel productive, and how to choose the right one for any decision you’re facing.

Here’s the simplest way I know to tell them apart.
A vanity metric makes you look good. A meaningful metric helps you make a better decision.
Vanity metrics tend to:

Meaningful metrics tend to:

A real example: in the Dashboards tool at HubSpot, it’s tempting to track how many dashboards users create. The number grows. It looks healthy. But creation isn’t the job. Using the dashboard to make a real decision is the job.

The moment you shift to tracking return visits and downstream actions, you see a completely different picture.
Same data. Different question. Totally different insight.

A good metric has three qualities.

It measures outcome, not activity.
“Features shipped” is activity. “Time to value for new users” is an outcome. One tells you your team is busy. The other tells you your product is working.

It’s actionable.
If the metric moves in the wrong direction, you should know immediately what you’d do next. If you have to think about it for more than 10 seconds, it’s a reporting metric, not a decision metric. Those aren’t the same thing.

It connects to something a customer actually cares about.
You should be able to draw a straight line from your metric to a real human outcome. If that line is blurry, you’re probably measuring a proxy for the thing that matters rather than the thing itself.

This is where most PM advice falls short. Telling you what a good metric looks like is easy. Helping you pick the right one for your specific situation is the hard part.
Here’s the framework I use.

Step 1: Start with the job to be done.
What is the customer actually trying to accomplish? Not what your feature does. What the customer’s life looks like when your product is working perfectly.

Step 2: Find the behaviour that proves it’s working.
What would a customer do differently if your product was solving that job well? That behavior is your metric. Not a proxy for it. The actual behaviour.

Step 3: Ask what would make this metric lie to you.
Every metric has a failure mode. If you reward signups, people game signups. If you track time on page, you might be measuring confusion, not engagement. Know the ways your metric can mislead you before you commit to it.

Step 4: Define success before you start.
Write down what a good result looks like before you run the experiment or ship the feature. If you wait until after to decide what the numbers mean, you’ll find a way to make anything look like a win.

Most teams don’t have a data problem. They have a metric design problem.
They’re measuring the wrong things carefully, and wondering why the insights never quite land.
Fewer metrics. Better questions. Ruthless honesty about what the numbers actually mean.
That’s the difference between a dashboard that makes you feel good and one that actually helps you build better products.

See you next Friday,
Stef

P.S. If you’re an aspiring PM trying to figure out how to talk about data without a formal analytics background, that skill is more learnable than you think. It starts with asking better questions, not knowing more tools. Book a free session with me on ADPList and we’ll work through it together. https://adplist.org/mentors/stefanie-brown