Help Me Choose Comparison Modal: +2.6% Revenue per Visitor on a Two-Model Product Page
[ SUMMARIZE WITH AI ]
[ FREE CRO TEARDOWN ]
Find the 3 biggest revenue leaks on your store.
Every day a conversion leak goes unfixed, you're paying for traffic that doesn't buy. Get a 5-minute Loom through your PDP, cart, and checkout, with mockups of the fixes. No pitch.
Get My TeardownWhen a product page sells two versions of the same thing, the shopper's first question is not "do I want this?" It is "which one?"
A sports equipment brand sells one of its core products in two models: a base model and a premium model. Same product family, different build, different price. On the product page, the model picker was two buttons with the model names on them. Nothing explained the difference.
We tested one small fix: a "Help me choose" link next to the model picker. Click it, and a comparison modal opens with both models side by side.
14 days. About 530 orders per variation.
Revenue per visitor rose 2.6%, at 62% probability to beat control, and conversion rate rose 2.6%, at 66%. On desktop, revenue per visitor rose 11.9%. None of that got close to 95%. This page explains what the result does and does not show, and why the brand shipped the help me choose link anyway.
Two Models, Two Buttons, Zero Explanation
Base vs premium is one of the most common setups in DTC. It is also one of the most neglected.
Brands put real work into the gallery, the reviews, the benefits copy. Then the most important decision on the page, which model to buy, gets two pill buttons and a price change. The shopper is left to guess what the extra money buys.
Guessing is expensive. Some people pick the cheaper one and wonder if they settled. Some open a new tab to look for a comparison. Some leave to think about it. None of that is an add-to-cart.
Our hypothesis was simple: if we explain the difference at the exact moment the shopper has to choose, fewer of them stall on the choice.
One Link, One Modal, Nothing Else Moved
Page: Product page for the two-model product, desktop and mobile
Platform: Intelligems
Test type: Control plus one variation
Duration: 14 days
Traffic: Split evenly, about 19,000 visitors in the variation
Control
The model row as it was: a "Model" label and two buttons, one per model.
Variation
The same row, plus a "Help me choose" link with a small info icon, sitting right next to the "Model" label. Clicking it opens a modal with two cards, one per model. Each card shows an image, the name, the price, and what makes that model different.
We also hid one option row below the model picker. It only ever had a single value and was already pre-selected, so it asked nothing of the shopper. That is a second change bundled in, and we flag it so nobody credits the whole result to the modal.
We did not invent a new interaction for this. The store already had a "Help me choose" fit guide on a different option, so the comparison modal reused the same link style and the same modal mechanics. It matched a pattern already on the store, and our team had a working, accessible modal to build on.
What the Comparison Modal Actually Said
The content is the part most teams get wrong, so here is the structure, with the specifics stripped out.
| Element | Base model card | Premium model card |
|---|---|---|
| Image | Close-up of the part that differs | Same crop, same size, so the difference is visible |
| Price | Shown | Shown |
| Build | Standard components | Upgraded components, named |
| Feel | How it feels in use | How the upgrade changes that feel |
Above the cards, one line on what both models share. That line matters. It tells the shopper they cannot make a bad choice, only a different one.
No feature checklist with 20 rows. No "most popular" badge pushing one side. Two cards, four differences, one close button.
A Small Lift, Pointing the Right Way
| Metric | Control | Help me choose | Change | Probability to beat control |
|---|---|---|---|---|
| Revenue per visitor | $11.14 | $11.43 | +2.6% | 62% |
| Conversion rate | baseline | +2.6% | +2.6% | 66% |
| Orders | 529 to 537 per variation | |||
Read this table honestly and it says one thing: the direction is positive, the size is not proven. At 62% and 66% probability, a 2.6% lift on about 530 orders per variation is too small to separate from noise.
Order value stayed flat. Revenue per visitor and conversion rate moved by the same amount, so the modal did not push people up to the premium model on average. It helped them pick, not upgrade.
7.9% of shoppers in the variation clicked "Help me choose", 1,500 out of about 19,000. Those clickers converted at 7.1%, with $27.62 revenue per visitor. Do not read that as the modal's effect. People who click a comparison link are already close to buying. It does tell you the link reached the right people.
Desktop Read It, Mobile Mostly Didn't Need It
The lift came from desktop. Revenue per visitor on desktop rose 11.9%, at 77% probability to beat control. Mobile came out flat, and mobile carries most of the traffic, so the overall number stays modest.
Our read: on desktop, the modal sits next to the buy box with room to breathe and the shopper reads both cards at a glance. On mobile, the same modal takes over the screen, and the decision it helped with was only one of several hurdles before checkout. The desktop orders are still a small sample, so treat the 11.9% as a direction, not a forecast.
Comparison Modal or Product Finder? Different Jobs
We have also published a need-based product finder case study. Both reduce choice overload. They are not the same tool, and picking the wrong one wastes a test.
| Product finder | Help me choose comparison modal | |
|---|---|---|
| Where it lives | Homepage or collection | Product page, next to the option picker |
| Choice it solves | Which product, out of many | Which version, out of two or three |
| Shopper's state | Browsing, not sure what fits | Already wants the product, stuck on the model |
| What it shows | Needs or use cases that route to products | The same product, side by side, with the differences named |
| Metric to watch | Product page reach, then revenue | Add-to-cart on that page, then revenue |
Rule of thumb: if the shopper does not know which product they need, build a finder. If they know the product and freeze on the version, build the comparison.
Shipping It as a Block the Team Can Edit
The brand called it a winner. Not a huge one, but a slight bump with nothing negative behind it, on a change that is cheap to keep and safe to deploy. We agreed. Waiting weeks more to push revenue from 62% toward 95% would have cost more than it could teach us.
Test code is not live code, so our team rebuilt it before it went to the live theme on 21 Jul 2026:
- Visible by default, with all test-only hide and reveal rules removed
- Rebuilt as a theme block the brand's team edits in the Shopify editor: link text, modal title, intro line, and both cards (name, price, image, description)
- Works on any product page that uses the same product section, so the next two-model product gets a help me choose modal by adding the block, with no developer
- The single-value option row we hid in the test stayed visible on launch, pending the brand's call, because it was never part of the modal itself
That last point is the useful habit. When a test bundles a small extra change, decide on it separately at launch instead of shipping everything by default.
Help Me Choose Modal Questions
When does a "Help me choose" comparison modal make sense?
When one product page sells two or three versions of the same product at different prices, and the option buttons only show names. If shoppers already understand the difference, for example sizes, a modal adds nothing.
Is a 62% probability to beat control enough to ship?
Not on its own. Here conversion and revenue moved in the same direction, desktop was clearly ahead, order value was flat, and the change was cheap to keep. That combination made shipping the better call. For a costly or hard-to-reverse change, wait for 95%.
Should the comparison modal push shoppers to the premium model?
Not in our test. The cards were neutral, with no 'most popular' badge, and average order value stayed flat. The win came from helping people decide, not from upselling them.
How is this different from a product finder or quiz?
A finder helps a shopper pick a product out of many, usually on the homepage or a collection. A comparison modal helps a shopper who already chose the product pick between its versions, right on the product page.
Why did desktop gain more revenue than mobile?
Desktop revenue per visitor rose 11.9%, while mobile came out flat. Mobile carries most of the traffic, which kept the overall revenue lift modest. Desktop orders were few, so treat that split as a direction.
Intelligems powered the test; the hypothesis, build, and analysis came from a CONVERTIBLES CRO engagement. Every published win lives in the full test archive, and a 30-minute call gets you three specific recommendations for your own product pages.