CPR Finder · Laerdal Medical × AHA

List view first: +7% mobile conversion

I own CPR Finder’s product analytics in PostHog. It’s the team’s feedback loop: track, find, test and measure, all live.

DraftPublished Updated 1 min read

This is a working draft. I’m still filling it in.

  • +7%

    Mobile conversion after making the list view the default

  • 50%

    Of visitors click through to a distributor to book

On this page
  1. One loop, start to finish
  2. Measuring what learners actually do
  3. Data for every team
  4. What I learned
  5. Keep reading

One loop, start to finish

  1. Track

    How mobile learners get from search to booking.

  2. Find

    Mobile learners converted better from the list view than from the map.

  3. Test

    We made the list the default on mobile.

  4. Measure

    Mobile conversion went up 7%.

With the list first and the map second, mobile converted about as well as desktop.

I also A/B test the landing page itself, like which kind of photo gets learners to search for a station.

CPR Finder on a phone: nearby training sites in a list, with a Book now button
The list view, now the default on mobile.

Measuring what learners actually do

I pushed back on learner targets that didn’t match how people actually behave. Now I track the share of visitors who click through to a distributor to book: about 50%.

Data for every team

  • I build dashboards for everyone from Marketing to impact managers, each showing the numbers behind that team’s decisions.
  • I get data on learners who struggle from our data team and go through it with RQI Partners’ impact managers, so distributors can see where learners get stuck.
  • I ran a data quiz on learner behavior, personas and the US market, so the whole team decides from the same picture.

What I learned

  • Taste vs test. An A/B test gets us to a decision faster than going back and forth on opinions.
  • I judge my work by how it does live, and I let the data change my decisions.

Topics: data, experimentation, growth. Also as plain markdown.