Measuring UGC in Canada carries one extra complication: a national campaign usually runs in two languages, and comparing them against each other tells you almost nothing useful.

The numbers that decide

Cost per purchase and three second retention. The first tells you whether the campaign works, the second tells you whether the hook works. Everything else diagnoses. A creator whose videos have high plays and no purchases has made entertaining content, which is not what was bought.

The numbers that only diagnose

Views, reach, likes and shares. They explain why something happened, they do not tell you what to do next. A brand that reports on reach is reporting on the part of the campaign that costs nothing to inflate.

Do not compare English against French

The two audiences differ in size, in cost per impression and in competitive density. A French campaign with a higher cost per purchase is not failing, it is running in a smaller market. Compare each language against its own previous results and against its own control.

Keep a control running

Leave one known piece running while testing new ones. Without it, a drop in performance is impossible to attribute: it might be the creative, or it might be the season, the competition or the platform. The control separates the two.

What survives the campaign

The winner is not a video, it is a lesson about which promise moves your buyer. Write it in one line after each test. In a bilingual market, note whether the winning promise was the same in both languages, because sometimes it is not, and that is one of the most valuable things you can learn here.

Keep reading

Find creators for your campaign →

Frequently asked questions

Which two numbers decide?

Cost per purchase and three second retention.

Can I compare English and French?

No, compare each against its own history.

Why keep a control?

To separate creative from season and competition.

What should I record after a test?

Which promise moved the buyer, in one line.