Shopify PDP Redesign: Four Levels, and When It Pays

WRITTEN BY

Gary Culaev, Founder and CEO, Weblics Consulting

How to Redesign a Shopify Product Page: Four Levels of Change, and When the Fourth Is Worth It

The safest way to redesign a Shopify product page is to do most of it without a redesign. Take a typical PDP redesign brief apart line by line, and eight of its ten lines turn out to be small changes you can ship and test on their own. Only one needs a new template.

You look at your product page every day. Your customers see it once, usually on a phone, and usually not for long. After a year of daily looking, anyone would want to tear it down and start again.
‍
That feeling tends to turn into a brief. Something like this one, which is illustrative but probably won't look unfamiliar:

Product page redesign brief

1. Make it feel more premium and modern
‍
2. New hero image and gallery
‍
3. Add product video
‍
4. Rewrite the description, it's too long
‍
5. Move reviews higher up the page
‍
6. Add a comparison table against competitors
‍
7. Sticky add to cart button on mobile
‍
8. Show shipping and returns next to the button
‍
9. New fonts and colors
‍
10. Rebuild the page on a new layout

Every line on it is reasonable. The problem is shipping all ten at once.

In this guide

Why most PDP redesigns can't tell you what worked
The four levels of change
Taking the brief apart
Levels 1 and 2: what ships this week
Level 3: one section at a time
What template tests look like in our log
When a full redesign is worth it
FAQ

Why Most PDP Redesigns Can't Tell You What Worked

A redesign that ships ten changes at once gives you one result for all ten. If conversion moves, you can't say which change moved it, and if it drops, you can't say which change to undo.
‍
It's like changing your diet, your sleep and your gym routine in the same week. If you feel better by spring, you won't know which one to keep.
‍
In practice, the new page launches, conversion shifts a little, and the team spends a meeting arguing about whether it was the video, the gallery or the new fonts. Nobody can settle it, because nothing was compared against anything.
‍
The Nielsen Norman Group made the broader point back in 2015. In "Radical Redesign or Incremental Change?", NN/g's Hoa Loranger recommends incremental change as the default, because "radical changes have a higher chance of inadvertently breaking something critical for users."
‍
Yes, first impressions matter, and a page that looks dated can cost you sales. But on a phone, the first impression is made by the first screen: the image, the product name, the price, the button and whatever sits right next to it. That's a small part of the page, and most of it can change without a redesign.
‍
Some teams can name the change that moved their conversion rate. Others ship twenty changes at once and never find out which one did it. Everything below is about being in the first group.

The Four Levels of Change

Every change on a product page falls into one of four levels, from editing words in your admin to building a new template. Each level up costs more, and only the last one bundles so many changes that a test can't tell them apart.

Four levels of Shopify product page change, from admin edits to a new template, with who does it and what a test shows

The time ranges for levels 1, 2 and 4 are our rough estimates, not a benchmark. Level 3 uses Invesp's estimate for building a typical A/B test, which is 10 to 30 development hours.
‍
Level 2 is the one most stores underuse. According to Shopify's help center, you can add, rearrange, hide and delete sections and blocks in the theme editor without touching code. Which sections you get depends on your theme and its version, but a lot of what reads as "redesign" in a brief is really a drag and drop.

Taking the Brief Apart

Sorted by level, only one line of the ten-line brief needs a new template and two need a developer. Six can be done in the admin or the theme editor, and one isn't a change at all.
‍
Line 1, "more premium and modern," is a feeling you want the page to create, and the lines below it are how you'd get there. Take it off the task list and keep it as the test of whether the other changes worked.
‍
Lines 2, 3 and 4 are level 1. New photos, a product video and a shorter description are edits in the product admin, though the video may need a media block in the theme editor. Check your real-visitor speed report after adding video, because video is often the heaviest thing on a product page. How speed affects conversion shows where to look.
‍
Lines 5, 8 and 9 are level 2. Moving reviews up is dragging a block. Shipping and returns next to the button is a text or collapsible block. Fonts and colors live in theme settings. For where reviews belong on the page, see where to add reviews on Shopify.
‍
Lines 6 and 7 are usually level 3. A comparison table and a sticky mobile add to cart are components many themes don't include, so a developer builds them as sections. Some themes already have a sticky button in their settings, which drops that line to level 2.
‍
Line 10 is the only true level 4, and once the other eight have been tested, you may find you don't need it.
‍
So the ten-line redesign is really eight small changes and one rebuild, and the eight can be shipped and tested one at a time.

Levels 1 and 2: What Ships This Week

Level 1 and 2 changes can go live within days and be tested on their own, so start with the ones that touch the first screen on a phone.
‍
That's where the first impression lives: the main image, the product name, the price, the add to cart button, and the shipping and returns line beside it. If the brief's complaint is that the page looks dated, this is the part the complaint is about, and the fix is usually a mix of new photos, tighter copy and a reordered block or two.
‍
This week, you could move shipping and returns next to the button, which is line 8, and start a test. Three months from now, on the full redesign plan, you'd still be waiting for the launch. At four weeks a test, the one-change-at-a-time path gives you three answers before the redesign gives you its first.
‍
For the full list of what belongs on a product page, see our product page optimization checklist.

Level 3: One Section at a Time

A new section is worth a developer's time when a level 1 or 2 test has shown there's something to gain in that part of the page, or when shoppers keep asking a question the page can't answer.
‍
The comparison table is a good example. If customers keep emailing to ask how your product differs from a competitor's, the page is telling you what it's missing. Build the table as one section, test it, and keep it if it wins.
‍
At 10 to 30 development hours, and $65 an hour, the rate we used in our breakdown of what one A/B test costs, one section comes to roughly $650 to $1,950 before design and QA. That's where the money starts, so let a cheaper test earn it first.

What Template Tests Look Like in Our Log

In our own testing log, five level-4 product page tests ran in the past year, and the only test that won with confidence was the one whose new templates removed things.
‍
On a pasta brand in June 2026, we tested the live product template against two simpler versions, one stripped further than the other, on about 2,400 visitors and 200 orders per version. Both simpler templates beat the control on conversion, by 22% and 27%, with a 99% or better probability of being ahead, and revenue per visitor rose about 20% on each. The brief for that page could have been the ten lines above. What won was subtraction.
‍
The three template tests that added or rearranged came back the other way. A jewelry brand's rebuilt page was 35% behind on 900 visitors per version, too few to read. A subscription food brand's new template was 5% behind on about 4,200 visitors per version. A menswear brand's best of two new layouts was 9% ahead. None of the three reached significance, and none shipped.
‍
The closest thing to the full brief above was a food brand's product page rebuilt from scratch in July 2026: two new templates against the old page, new visitors only, six weeks, about 4,000 visitors and 140 orders per version. One came back 5% ahead on conversion, the other level, and neither reached significance. Six weeks of design and development produced a page that neither won nor lost, and nobody could say which of its changes had done what. That's the level-4 result in its purest form.
‍
Five tests is a log, not a law. It does say where to start: before the new template, find out what the old one can lose.

Five Weblics template tests: the two simpler templates won on conversion, the five that added or rearranged did not

Lifts are against each store's live template. Only the two simpler templates cleared a 99% probability of being ahead, so the other five bars show what the test produced rather than a real effect.

When a Full Redesign Is Worth It

A full product page redesign is worth it when the page can't do what you need even after the cheaper levels, and NN/g lists five situations where a radical redesign is justified.
‍
Adapted to a product page, they are:
‍
• You've made incremental changes for a while, and the gains have become tiny or stopped.
‍
• The theme is so outdated that changes you need, such as a proper mobile layout, can't be made.
‍
• Years of patches have left the page a tangle of sections that fight each other.
‍
• Conversion is severely low across the whole site, not on one element.
‍
• Side by side, your page is far behind your direct competitors.
‍
If one of those is true, redesign. Then ship the new page as a test. Tools such as Intelligems and Shoplift can test one Shopify template against another, so the old page keeps running for part of your traffic and the new one has to beat it. A redesign tested that way is still one result for many changes, but at least you'll know whether the whole package won.
‍
Knowing the four levels isn't enough to act on, though. You still need to know which changes are worth making at levels 1 to 3. Our product page optimization checklist is the list we start from at Weblics before anyone budgets for a rebuild.
‍
A redesign isn't the big version of optimization. It's what you do after the small versions stop working.