A presell page sits between your ad and your product page. Its job is to educate and build trust before people see the product. It doesn't take the order, so the click on its button is not a sale.
That changes what you measure. The number that decides whether a presell page works is what it costs you to get a customer through it, compared with sending the same ads straight to your product page.
The click on the button is not the result
Founders tell me they get 4-9% click-through rates on their ads and still don't convert. The clicks show that people wanted to know more. So you look past the ad, to the page.
The same thing can happen one step later. A landing page can send lots of people on to the product page, and those people can still leave without buying. A higher click-through to the product page is not a win if acquiring a customer becomes more expensive.
So read the numbers in this order:
- Purchases, cost per purchase and CAC for traffic that went through the presell page.
- Average order value and refunds, because a page that explains a bundle can change both.
- The click-through from the landing page to the product page, to find where people drop.
CAC is ad spend divided by the customers you acquired. Cost per purchase is ad spend divided by purchases. If your ad platform and your store disagree on purchases, pick one source and use it for both sides.
What each number tells you
| Number | Where to find it | What it tells you | What it can't tell you |
|---|---|---|---|
| Ad click-through rate | Your ad platform | Whether people wanted to know more | Whether the page sells |
| Landing page click-through to the product page | Page visits vs store sessions with that UTM tag | Whether the page made people want to see the product | Whether they bought |
| Product page conversion for that traffic | Store analytics, filtered by UTM tag | Whether the product page and offer matched what the page sold | Which part of the landing page did the work |
| Purchases, cost per purchase and CAC | Ad platform and store orders | Whether the presell page pays for itself | Why it won or lost |
| Average order value and refunds | Store orders | Whether the page changed what people bought, and if they kept it | Whether a lower CAC came from a smaller order |
Set it up so the two sides are readable
Do this before you spend anything:
- Give each page version its own URL. If you test two versions of the presell page, they need two URLs, not one URL you edit halfway through.
- Tag every ad link with UTM parameters. Use the same source and medium on both sides and a different campaign or content value for each destination. That is how your store tells the traffic apart.
- Check the button's destination. Tap the main button and confirm it opens the product, variant and price the page sold. The message match worksheet walks through that check.
- Click through once yourself. Open the ad link on a phone, logged out, tap through to the product page and confirm the visit shows up in your store analytics with the right tags.
If the presell page lives on a different domain from your store, check this last step carefully. A visit that loses its tags on the way to the store gets counted as direct traffic, and the presell side looks worse than it is.
Run it as a split, not a before-and-after
Duplicate one ad that's already spending. Change only the link: one copy goes to the presell page, the other to your product page. Same audience, same budget, same dates.
A before-and-after comparison mixes the page with everything else that changed that week: the season, the offer, the creative, the budget. If the page, the offer and the copy all change in the same week, the result describes the combined change.
The homepage vs landing page post covers the split step by step.
Read the result honestly
Let both sides run until each has enough purchases to compare. A result with only a handful of orders on each side is uncertain. Write it down as uncertain instead of calling a winner. Stop early for a broken page or a tracking problem, not because one side leads after a few orders.
When I tested a Landra page against our old product page on Pow (same ads, same product), CAC dropped 46%. That's one test on my own brand, not a promise. CAC is the number I'd read first on yours.
When the numbers disagree, find where people drop
If the presell page loses on cost per purchase, the diagnostic numbers tell you where to look:
- Low click-through from the landing page to the product page. The page isn't answering what the ad promised, or the answer comes too late. Check the opening against the ad first.
- High click-through but few purchases. People liked the page and then left on the product page. Check that the button opens the product, variant and price the page sold, and that the offer's conditions were stated before the button.
- More purchases but a lower order value. The page may be selling a smaller order than your product page. Compare what each side's buyers actually bought.
Fix one thing, then run the split again. Keep a short record each time: what you changed, the dates, spend, purchases, revenue and refunds for each side, and the result (winner, uncertain or tracking problem). If you write a brief before you build, the landing-page brief has a place for the ad, the audience and the destination.
Building the page to test
If you need a presell page to test against your product page, Landra turns a product URL into a finished advertorial, listicle or presell page, with the copy, layout and images done for you (14-day trial, plans from $19/mo). We recommend checking your page to make sure we got everything right, including where its button goes, before you send ads to it.
Frequently asked questions
What is a good conversion rate for a presell page?
There is no single number to aim for, because the presell page does not take the order. Compare cost per purchase and CAC for traffic that went through the presell page with the same ads sent straight to your product page.
Should I judge a presell page on click-through to the product page?
No. A page can send more people to the product page and still cost more per customer. Use that click-through to find where people drop, and use cost per purchase and CAC to decide.
How do I track a presell page in Shopify?
Tag each ad link with UTM parameters and give each page version its own URL. Then check that a visit from the ad shows up in your store analytics with those tags before you spend.
How long should a presell page test run?
Until both sides have enough purchases to compare. A result with only a handful of orders on each side is uncertain. Stop early only for a broken page or a tracking problem.
