How do I test landing pages per ad without duplicating ads?
Short answer: test after the click, not inside Meta. Leave the ad as it is. When someone clicks it, send them to one of the pages you want to test. Then compare how each page sells for that ad.
Why duplicating ads is a bad test
The usual way is to duplicate the ad and give each copy a different page. It has three problems:
- The copies start from zero. No learning, and often none of the likes and comments the original has.
- It isn't a fair test. Meta puts spend behind the ad it already trusts. The copy often gets a small share, so each page gets very different visitors.
- It's slow. One ad and four pages means four ads to build, watch and clean up. For every ad you run.
How to test after the click, by hand
- Keep one link in the ad.
- Behind that link, split visitors between your pages. An A/B testing app can do this.
- Make sure a returning visitor sees the same page.
- Compare revenue per visitor, not just conversion rate. A page with fewer but bigger orders can win.
The catch: most A/B testing apps pick one winner for all your visitors. If your ads are different from each other, you need a winner for each ad.
Where PagePair fits
PagePair is a Shopify app that sends each Meta ad to the landing page that sells best for it. It sees which ad each visitor came from. Then it splits that ad's visitors between the pages that sell its product. The pages that make the most revenue per visitor for that ad get more of its visitors. You don't duplicate any ad. The ad's own page always keeps at least 5% of its visitors, so you can see how much more the other pages make.
When you don't need PagePair
- You only want to test one or two pages, once. A simple A/B test is enough.
- You run very few ads.
- Your ads send people to a store that isn't on Shopify.