To me, it's all about your level of curiosity and tenacity to get to the bottom of a question you have.

It could be about your website or digital app and examining the clues left by the data and the user experience.

For instance, an exit page is the last page a user visits before leaving your website.

To improve conversion rates, what I have often done in the past is examine these pages to figure out what made people leave.

So how do you do that?

First, identify the pages (Where is the problem).

Where do people leave your website? Could be form pages, could be checkout pages, or a page with a picture of a cat on it.

Doesn’t matter.

Just look for the ones where people leave in your analytics tools.

Second, what do the users do before leaving? (What is the problem)

Do they click around other parts of your website or app first? If so what are they looking at?

Third. So you know what they were doing before they left the site via the exit page.

Can you identify by the pages they visited what their intent was going to your site?

Why were they there, why is it happening? (Why points you to your testing options)

Did they find what they were looking for before they left?

Yes? Bingo. No worries. Your exit page is great.

No? Ahh. Now you might have a user experience problem.

Has that exit page got anything that might help them get to where they and you, need them to go next?

No?

Then that’s probably why they exited.

Action point?

Test the page with the most viable options of why your visitors might be there and move them further down the journey to what they want to achieve on your website.

A Real Life Example

An e-commerce website I worked on identified that its checkout page had a high exit rate, particularly on mobile devices.

The exit rate was almost double the rate of desktop visitors.

When we had a look with our mobile phones we discovered that the page took ages to load (10+ seconds).

So we cleaned up the code so it was responsive for mobiles and got that down to less than 2 seconds.

Conversion on mobile went to the same level as desktop.

TLDR: To become a data detective, find out where the problem is, what the problem is and why the problem happens.

Then fix it.