In short
Learning goal: by the end of this lesson you can check a product page against the questions shoppers ask before they buy, and use heatmaps to see which parts of the page they actually look at.
The product page is where a shopper decides. Everything before it is getting them there; everything after it is not losing them. On a new store, product pages are usually built once, from the theme's defaults, and never looked at again. This lesson is the second look.
What to measure at this step
| Metric | Definition | Role |
|---|---|---|
| Revenue per visitor to the page | Revenue from visitors who viewed the page ÷ those visitors | Primary goal: decides whether a change helped |
| Orders per visitor to the page | Orders ÷ visitors who viewed the page | Primary goal when order size won't change |
| Add-to-cart rate | Visitors who added the product ÷ visitors who viewed it | Diagnostic: explains a result, never decides it |
| Variant selection | Share of visitors who choose a size or color | Diagnostic: shows whether options are clear |
| Scroll reach and clicks | How far visitors scroll, what they click | Diagnostic: shows what's seen and what's ignored |
| Refund rate | Refunded orders ÷ orders | Guardrail: a page that oversells gets more returns |
Why add-to-cart can't decide: a page can get more people to add to cart by hiding things they'll find out later (shipping cost, delivery time, size limits). Those carts then stall at checkout. More adds, same orders. The example below shows why it's still tempting to use.
The checklist
Work through it on your best-selling product, on a phone.
Photos
- [ ] The first photo shows the product clearly, on its own, at a size you can read on a phone.
- [ ] At least one photo shows it in use or in a hand, for scale.
- [ ] Each color or variant has its own photo, and choosing the variant changes the photo.
- [ ] Photos load quickly on mobile data. Very large image files slow the whole page.
Price
- [ ] The price is next to the product name, not below the description.
- [ ] Any crossed-out price is a price you actually sold at. In Shopify the crossed-out price is the compare-at price, and Shopify only shows a sale when it's higher than the price.
- [ ] If the price depends on the variant, it updates when the shopper chooses one.
- [ ] Shipping cost, or the free-shipping threshold, is stated near the price.
Delivery
- [ ] An estimated delivery time is under the Add to cart button: "Ships in 1 to 2 days, arrives in 3 to 6".
- [ ] In November and December, the last order date for delivery before the holidays.
Baymard's research lists "delivery was too slow" as a reason for 20% of abandoned purchases among shoppers who weren't just browsing (page updated 22 September 2025). A shopper who can't find the delivery time leaves to look for it, and may not come back.
Returns
- [ ] A one-line returns summary near the button ("Free returns within 30 days"), with a link to the full policy.
Baymard lists an unsatisfactory returns policy as a reason for 13% of abandoned purchases (same page).
Reviews and trust
- [ ] Reviews are real and recent. If you have few, show the ones you have; don't invent or buy them.
- [ ] Contact details or a chat option are easy to find. Baymard lists distrust of the site with card details as a reason for 19% of abandoned purchases (same page).
The button
- [ ] Add to cart is visible on the first screen on a phone, or a sticky button appears as the shopper scrolls.
- [ ] No popup or chat bubble covers it.
- [ ] If you've turned on express payment buttons (Shop Pay, Apple Pay, Google Pay), they appear on the product page and work. Shopify's accelerated checkouts let returning customers skip the cart and go straight to checkout from the product page.
Use heatmaps to see what's ignored
A checklist tells you what should be on the page. A heatmap tells you what shoppers actually see. Two views matter:
- Scroll reach: how far down the page visitors get. If most phone visitors never reach your returns note, it might as well not be there. Move it up.
- Clicks: what visitors tap. Taps on a photo that doesn't zoom, or on text that looks like a link, are signs of a shopper looking for something the page doesn't give them.
Heatmaps describe behavior; they don't prove that a change helps. Use them to choose what to fix or test, not to declare a winner.
Test ideas, and why they'd work
| Idea | Mechanism |
|---|---|
| Delivery estimate under Add to cart | Answers the gift buyer's first question without leaving the page |
| Returns line next to the price | Lowers the perceived risk of buying something untried |
| Product-in-hand photo second | Answers "how big is it?" before the shopper reads the description |
| Sticky Add to cart on phones | Keeps the next step in reach after scrolling through photos |
A product page gets 1,500 visitors a week. 2% of them order, and 8% add the product to the cart. The owner wants to test a delivery estimate under the button and is choosing a goal.
- Orders as the goal: detecting a 20% lift (2% to 2.4%) at 95% confidence and 80% power needs 21,109 visitors per version, about 29 weeks of this page's traffic.
- Add to cart as the goal: detecting a 20% lift (8% to 9.6%) needs 4,921 visitors per version, about 7 weeks.
Computed with sampleSizePerVariant from Benchmyrk's statistics engine. The store is illustrative.
Seven weeks is tempting. But a delivery estimate could raise adds for a reason that doesn't help orders (shoppers adding to cart to see the shipping cost, then leaving). If you test on add-to-cart, you've learned about adds, not sales. For a fix like a delivery estimate, which answers a real question honestly, the better choice at this traffic is often to add it and watch the funnel (lesson 1 shows how), not to test it.
A 10-minute exercise on your own store
- Open your best-selling product page on your phone.
- Without scrolling, note whether you can see: the price, the button, the delivery time, the returns line.
- Scroll to the bottom and count how many screens it takes. Note where reviews and the returns policy sit.
- Go through the checklist above and tick what's there. Circle the first unticked item in the Delivery or Returns sections.
- Fix that one item today. Note the date, so you can look at the funnel for the weeks before and after.
Do it in Benchmyrk
For product page work before the sale, see the fix list in Black Friday for a small Shopify store.
Check yourself
- A change raised add-to-cart by 15% and orders didn't change. What might have happened? Answer: the page may be getting adds it doesn't turn into orders, for example by hiding shipping costs until the cart. Add-to-cart is a diagnostic; orders or revenue per visitor decide.
- The heatmap shows most phone visitors never scroll to your returns policy. What do you do? Answer: move a one-line returns summary near the price or button. That's a fix, not a test.
- Is a crossed-out "was" price fine if the product has never sold at that price? Answer: no. A crossed-out price should be a price you actually sold at. Rules on reference prices vary by country; part 5 of the small-store Black Friday series covers the EU and US guidance.