Comparison landing pages: how to build a “vs” page buyers actually trust
People who compare are close to buying. The page that wins them is the one that helps them decide, admits what the other options do better, and still makes your product the obvious choice for the right reader.
On this page 10 sections
- When a comparison page is the right format
- Anatomy of a comparison page that converts
- Choose criteria the reader cares about
- The concession: lose one criterion, honestly
- Count the cost over time, not the sticker
- Naming competitors: what the rules say
- Headlines for comparison pages
- Before you publish
- How do you keep a product comparison table verifiable?
- Are both delivery claims measuring the same stage?
Someone who clicks an ad that says "Red light mat, LED bed or panel?" has already done half of your work. They believe in the category. They are not asking whether to buy, only which one. A comparison page meets them at that exact moment.
Done badly, it is a table with green ticks in your column and red crosses in every other one. Readers have seen hundreds of those and trust none of them. Done well, it is the page that makes the decision for them, and often the last page they read before they buy.
That 0.72% is worth a second look: of the 1,525 unique English presell page titles in our research database, comparisons are the least used pattern of all1. Very few brands give buyers a fair comparison, even though buyers who compare are among the closest to a purchase.
When a comparison page is the right format
A comparison page works for one kind of reader: someone who knows the problem, knows that solutions exist, and is hesitating between them. In the language of customer awareness levels, they are solution-aware or product-aware.
You will recognize them by what they type, click and ask:
- "X vs Y", "best X for Y", "alternative to X"
- comments under your ads asking "how is this different from...?"
- support emails that start with "I was about to buy the other one"
- visitors who saw your product page, left, and come back through retargeting
For anyone earlier than that, a comparison page is the wrong first page. Someone who does not know their morning stiffness comes from the space between their knees has nothing to compare yet. Tell them the story first, with an advertorial, and keep the comparison for later.
There are three kinds of comparison pages, and they do not carry the same risks:
| Type | What it compares | Example | Main risk |
|---|---|---|---|
| Category vs category | Ways to solve the problem | A red light mat vs a spa LED bed vs a panel | Unfair category ranges |
| Your product vs the usual way | Your solution vs the habit it replaces | A refillable deodorant vs a plastic tube a month | Exaggerating the cost of the habit |
| Brand vs brand | Your product vs a named competitor | "Brand A vs Brand B" | Claims about the competitor you cannot prove |
The first two are the most common on presell pages, and the safest: you compare approaches, not a company that can answer back. Brand-versus-brand pages work too, but every sentence about the other brand has to be something you can prove.
Anatomy of a comparison page that converts
Megelin's red light mat page is a good model. It compares a full-length LED mat with the two other ways to get a full-body session: a spa LED bed and a panel on a stand.

Here is the order of its sections, and why each one is where it is:
- A headline that names every option and the method. "Red light mat, LED bed or panel: 8 weeks, 5 criteria, one verdict." The reader knows in one line what is compared, how, and that there is an answer.
- The verdict, immediately. "For three full-body sessions a week at home, the mat wins." People who compare want the answer. Making them scroll for it feels like a trick.
- The method. Who tested, for how long (8 weeks), with which criteria, set before the first session, and where the numbers come from. In this draft the tester, Mark, is a composite written as a placeholder, and the editor's checklist asks the brand to replace him with a real customer's test or relabel the section. On a published page, the test has to be real.
- The table. Five criteria, three options, a winner marked on every row, including the row the mat loses.
- One section per criterion, each ending with "Winner: ..." so a skimmer can read only the winners.
- "Who should choose what", with an honest profile for each option.
- Proof: the tester's notes and real customer reviews.
- The offer, the guarantee and the questions buyers ask before paying.
Notice what is not there: no list of twenty features, no attack on spa owners, no fake urgency. The page reads like a test report, and that is the point.
Choose criteria the reader cares about
The criteria are the real argument of a comparison page. Pick them badly and nothing else matters.
Start from the buyer's questions, not your spec sheet. The best sources are your reviews, your support inbox and the one- to three-star reviews of the alternatives: they say exactly what buyers fear. On the red light mat page, each of the five criteria answers a question a buyer would ask: how strong is the light at the skin, how much of the body does it cover, will I actually use it, what does it cost over time, and what can I adjust?
Keep them objectively measurable when you can. The FTC defines comparative advertising as comparing brands "on objectively measurable attributes or price"2. Dose in mW/cm², coverage in centimetres, cost over two years: all of these can be checked. "Feels premium" cannot.
Decide them before you test, and say so. "The criteria were set before the first session" is one of the most persuasive sentences on the Megelin page. It tells the reader the test was not built backwards from the result, which is exactly why it must describe a test that really took place.
Say where every number comes from. Megelin writes that figures for the mat come from the brand's specifications, while figures for beds and panels are category ranges, not one brand. That single line protects the page and the reader's trust.

Here is the difference between a table readers skip and a table they believe:
A table that loses trust
- Fifteen rows, most of them features only you have
- Green ticks in your column, red crosses everywhere else
- No source for any competitor figure
- Vague rows: "quality", "value", "results"
- The verdict hidden at the bottom of the page
A table that earns it
- Four to six rows, each a question buyers ask
- A winner on every row, and at least one row you lose
- A line saying where each figure comes from, and when it was checked
- Measurable rows: dose, size, time, cost over two years
- The verdict on the first screen, the table right after the method
The concession: lose one criterion, honestly
The most important line on the Megelin page is a heading: "The bed lights both sides at once. That one is not ours."

A concession does three things at once:
- It makes every other verdict believable. If the page admits the bed wins coverage, the reader believes it when the page says the mat wins on convenience.
- It helps the reader self-select. Someone who goes to a spa twice a month and wants both sides lit at once is told, kindly, that the bed suits them. They would have been an unhappy customer anyway.
- It shows you understand the category. Brands that never lose a criterion sound like they have never looked at the alternatives.
The concession must be real. A fake weakness ("our only flaw is that people love it too much") is worse than none. If your product truly wins on every criterion that matters, concede something smaller but true: a higher price, a longer delivery time, a learning curve in the first week.
Count the cost over time, not the sticker
When your product is a one-time purchase and the alternative is paid again and again, the sticker price is the wrong comparison. Megelin's fourth criterion is titled "Count the sessions, not the sticker", and it contains a small calculator: the price of one bed session, the number of sessions per week, and the two-year cost of each option.

The same logic works for any "buy once vs pay again" comparison: refills against disposable tubes, a water filter against bottled water, a home device against salon appointments. Try it with your own numbers:
Interactive
Buy once vs pay again
What the recurring option costs the buyer each month, and what your one-time product costs. Defaults: $45 spa sessions, three a week, against a $739 mat.
Use the buyer's real pace, not the most flattering one. If the payback only works at five sessions a week, the page should say who it is for.
Two rules keep this honest. Use a realistic pace, and show it: Megelin's calculator starts at the visitor's own numbers and lets them change everything. And keep the comparison fair: a panel is also a one-time purchase, so the page says so, and explains why a full-body routine would need several.
Naming competitors: what the rules say
You can name a competitor. Whether you should depends on whether you can prove everything you say about them.
United States. The FTC's policy statement on comparative advertising encourages naming or referring to competitors, but requires clarity and, if necessary, disclosure to avoid deceiving the consumer. It supports brand comparisons "where the bases of comparison are clearly identified", and it holds comparative claims to the same standard as any other claim: truthful and not deceptive2. Separately, section 43(a) of the Lanham Act makes anyone who, in commercial advertising, misrepresents the nature, characteristics or qualities of "another person's goods" liable to whoever is or is likely to be damaged, which in practice usually means the competitor3. In practice, the company you compare yourself with is the reader most likely to check your facts. And keep your brand visible: the FTC's 2024 rule on reviews and testimonials bans presenting a website the company controls as an independent reviewer of a category that includes its own products5.
European Union. Comparative advertising is any advertising that identifies a competitor or its products, "explicitly or by implication". It is permitted when it meets a list of conditions, including that it is not misleading, compares products that meet the same needs, objectively compares "material, relevant, verifiable and representative features" (price included), and does not discredit or denigrate the competitor4. The EU line on tone is stricter than the US one: a truthful but mocking comparison can still fall outside it.
A few practical habits follow from the rules:
- Compare like with like. Same use, same quantity, same period. A 30-day supply against a 90-day supply is not a price comparison.
- Date the data. "Prices checked in October 2026" costs one line and saves arguments.
- Do not borrow their brand. Their logo or product photo on your page can suggest an affiliation that does not exist.
- Write as if they will read it. They probably will.
Headlines for comparison pages
A comparison headline names the options and promises an answer. Fill in your own options to get ten starting points:
Interactive
Comparison headline generator
Name your product, two alternatives and your test. Every line updates as you type.
- Verdict{a|cap}, {b} or {c}: {time}, {n} criteria, one verdict
- Question{a|cap} or {b}: which one for {reader}?
- TestWe tested the {a}, the {b} and the {c} for {time}. Here is what won.
- Use caseFor {use}, the {a} wins. Here is how we tested it.
- Concession{a|cap} vs {b}: the honest comparison, including where the {b} wins
- Depends{b|cap} or {a}? It depends on one thing
- Cost{a|cap} vs {b}: compare the cost over two years, not the price tag
- Criteria{n} things that matter when you choose: {a}, {b} or {c}
- Profiles{a|cap}, {b} or {c}? An honest pick for each kind of user
- Before you buyThinking about the {b}? Read this comparison first
Keep only the lines your page can back up: a "we tested" headline needs a real test, with a method you publish.
Before you publish
Comparison page checklist
11 items
A good comparison page is generous: it would help the reader even if they chose the other option. That generosity is exactly what makes them choose yours.
How do you keep a product comparison table verifiable?
Give every comparison row an evidence rule before filling the cells. Fictional example: a table comparing travel mugs might use stated capacity, published dimensions, and an explicitly described carrying situation. Those are different kinds of information. A manufacturer specification, an observation from a defined test, and an editorial preference should not look like interchangeable checkmarks.
For a missing competitor detail, write that the detail was not verified in the material reviewed. Do not convert an empty field into a failure. Record the product version and the source date in your working notes so a later reviewer can reproduce the comparison. If two products are suitable for different situations, explain the trade-off rather than inventing a single winner for every buyer.
An interactive product quiz can route readers between plausible use cases, but it cannot supply missing product evidence. The advertorial disclosure review helps keep a commercial verdict recognizable as advertising. Separate those two decisions: the origin of the recommendation and the support for each comparison row.
| Review point | Useful evidence | Avoid |
|---|---|---|
| Specification | Name version and source | Turn an unverified field into a failure |
| Observed behavior | Describe the test conditions | Present a preference as a measurement |
| Verdict | State the relevant use case | Invent a universal winner |
Are both delivery claims measuring the same stage?
Compare the same fulfillment stage before adding a delivery row to a “versus” table. A preparation window, a carrier transit estimate and an arrival estimate describe different parts of the journey. Shopify’s fulfillment terminology separates these stages; a short dispatch promise cannot fairly beat a longer end-to-end estimate simply because its number is smaller.6
| Claim found on an option | Meaning to establish | Information still needed |
|---|---|---|
| “Ships within the stated window” | Preparation or dispatch | Applicable transit information |
| A carrier transit estimate | Movement after the relevant handoff | Preparation time and applicable destination |
| An estimated delivery date | An arrival estimate with conditions | Source, item, destination and service |
| “Free shipping” | A charge condition | Timing and any eligibility threshold |
Use the published information for the actual item and destination. If one option does not provide a comparable arrival estimate, label that field as unverified instead of filling the cell with a guess. Record the date of the source and the condition beside the claim. An observed checkout estimate for one address does not establish the same date for every customer.
This also changes the verdict. An option with a lower shipping charge can still have a less suitable arrival estimate for the reader’s task. Present those as separate criteria and let the reader choose which matters. Do not turn a missing estimate into evidence that a competitor is slow.
The presell shipping-promise review helps separate dispatch from arrival and creates a short note for the person who owns the offer. Use it before repeating the comparison in an ad or the final product handoff. The review records a clearer claim; it does not establish a conversion gain.
Frequently asked questions
Can I name a competitor on my landing page?
In the US, the FTC encourages truthful comparisons that name competitors, as long as the basis of comparison is clear and nothing is deceptive, and competitors can sue over false claims about their products under the Lanham Act. In the EU, comparative advertising is allowed when it meets the conditions of Directive 2006/114/EC, including comparing verifiable features and not denigrating the competitor. Many brands compare against a category ("spa LED beds") rather than one named brand, which avoids having to prove facts about a single competitor.
Should the comparison table be on the first screen?
The verdict should be. The table can come right after the method, because readers trust a table more once they know how it was filled in. On mobile, keep the table to three or four columns or switch to stacked cards.
How many criteria should I compare?
Four to six. Each one must matter to the buyer, be measurable or at least observable, and be decided before the test. If you only pick criteria you win, readers notice.
Can I send cold traffic to a comparison page?
Usually not. People who do not know the category yet have nothing to compare. Use a story, a discovery article or a quiz for them, and keep the comparison for retargeting, for people who searched for the options, or for ads that name the choice in the first line.
What if my product is more expensive?
Do not hide it. Compare the cost over the time the buyer will actually use the product, or concede price openly and show what the extra money buys. A page that hides the price comparison loses the readers who came for it.
Sources
- Presell page titles in our research database (1,525 unique English titles), Ads collected from the public Meta Ad Library. Ads started April 11 to October 11, 2026
- Statement of Policy Regarding Comparative Advertising, 16 CFR 14.15, Federal Trade Commission (1979)
- False designations of origin, false descriptions, and dilution forbidden, 15 U.S.C. 1125 (Lanham Act, section 43), U.S. Code
- Directive 2006/114/EC concerning misleading and comparative advertising, European Parliament and Council (2006)
- Federal Trade Commission announces final rule banning fake reviews and testimonials, Federal Trade Commission (August 14, 2024)
- Order fulfillment and shipping terminology, Shopify Help Center (Accessed October 30, 2026)



