Comparison pages that convert without lying

The two least trusted genres of web page might be the competitor comparison ("why we're better than Brand X") and the plan matrix ("compare our tiers"). Both sit on extraordinarily high-intent traffic — someone typing "Brassfern vs alternatives" is holding a credit card in their mind — and both are usually squandered with checkmark theatre: a grid in which one column is mysteriously all green.

Here is the uncomfortable truth that makes these pages work: the page's job is not to win the comparison. It's to make the reader trust the comparison. Trust converts; victory laps don't.

The fair-dealing rules

Before design, the ground rules we hold ourselves and clients to:

Facts only, and dated. "Competitor X charges per seat (checked March 2026)" is defensible and useful. "Competitor X is clunky and outdated" is neither. Link to their pricing page where a claim matters.
Their trademark, their respect. Use their name to refer to them — nominative fair use — but never in a way that implies endorsement: no logo mashups, no lookalike branding, no ads on their brand terms promising things you won't show on the page the ad lands on.
Compare what you'd want compared. If your weakness is obvious from their site — price, a missing integration, platform coverage — it will be found. Better to put it on the page yourself, framed honestly, than to let the reader find it later and wonder what else you hid.
Update it or delete it. A comparison page referencing a competitor's 2024 pricing in 2026 is both a lie and a ranking liability. We put a visible "last verified" date on every vs page and calendar the re-checks. A stale comparison page is worse than none.

These rules are not purity for its own sake. They're the conversion strategy. Visitors arrive pre-sceptical; the only scarce resource on the page is believability.

Honest asymmetry is the engine

The single most persuasive block on any comparison page is the honest concession: "If you need deep enterprise procurement workflows and on-prem deployment, choose them. They're genuinely better at it. If you want a team that ships weekly and answers its own email, that's us."

This does three things at once. It self-selects bad-fit buyers out (cheaper for everyone — those deals churn or become support sinkholes). It makes every other claim on the page retroactively more credible. And it demonstrates that you understand the buyer's actual decision criteria rather than your own feature list. The whole technique is borrowed from conversion copywriting: voice-of-customer research tells you which trade-offs buyers are already weighing; the page should sound like their internal debate, answered.

Asymmetry applies to the matrix itself. If every "us" cell is a checkmark and every "them" cell is a cross, nobody believes a single cell. Some cells should honestly read "Partial" — including yours. Include a row where you lose. One conceded row buys belief for the eleven you win.

Anatomy of a comparison row people read

Most matrices fail at the row level. The fix is to write rows as jobs, not features:

Weak: "SSO/SAML" — ✓ / ✓ / ✗
Strong: "Your IT team can enforce single sign-on without an enterprise-tier call" — Us: yes, standard tier. Them: yes, enterprise only.

Three moves make rows readable. First, phrase the row label as a capability the buyer wants, not your internal feature name — nobody wakes up wanting "granular RBAC," they want contractors who can't see payroll. Second, answer cells with specificity, not glyphs alone: "Yes — standard plan," "Add-on, $X/mo," " via Zapier," "No." Third, group rows by the decision being made (getting set up, living with it daily, growing out of it) rather than by your org chart's taxonomy. Eighteen flat rows are a wall; four named groups with four rows each are an argument.

And a hard technical note: matrices are the most common casualty of small screens. Sticky first column, honest cell labels, and the decision between scrolling and reformatting — the full mobile treatment is in responsive table design. Test your matrix at 375px before you celebrate it at 1440px.

The us-vs-them page, block by block

The structure that has worked across SaaS and services engagements:

Who this page is for. One sentence of self-selection: "You're comparing us and X if you're a 10–50 person company that values speed over procurement depth." Readers who match lean in; readers who don't were never going to convert.
The one-line difference. Not "we're better" — the axis the products differ on. Two good answers for different buyers.
The matrix. Grouped, specific, honest cells, as above.
Where they win. The concession block. Short, sincere, specific.
Where we win. Now it lands.
The switching story. Migration path, data importers, a named human; switching cost is usually the real objection, and pretending it's trivial insults the reader.
Proof adjacent to claims. A testimonial about onboarding speed belongs next to the claim about onboarding speed — the proof proximity principle — not warehoused in a logo band at the bottom.

Plan matrices: the same rules, pointed inward

Comparing your own tiers tempts different sins. The big ones:

Fake differentiation. Rows invented to make the middle plan look fuller ("Priority color picker!"). Every row earns its place by changing a real decision, or it's padding the reader has to wade through.
The dishonest highlight. "Most popular" should mean most popular. If it's actually "most profitable," find an honest frame ("Best for growing teams") that is true.
Hidden gotchas. Limits that matter (seats, projects, history retention) belong in the matrix, not in a footnote at 11px. The pricing page research we summarised in what 30 pricing pages taught us about clarity is blunt on this: discovered limits are the top cause of checkout abandonment on plan selection.
Annual/monthly ambiguity. Show both numbers when the toggle flips, with the effective monthly maths visible. Anchor charm ("$49/mo billed annually") without the annual total is a small lie with a large support-ticket cost.

SEO without slime

"X vs Y," "X alternative" and "best [category] for [niche]" are legitimate searches with genuine intent, and honest pages can win them. The line we won't cross is the programmatic SEO playbook run wild: template pages for eighty competitors, each asserting superiority with no human ever checking a fact. If you can't verify the page quarterly, don't publish it. Three excellent, maintained comparison pages will outrank and out-convert thirty mouldering ones — our content clusters work keeps confirming it.

Structure mark-up honestly too: these pages qualify for FAQ structured data when they genuinely answer questions, and never for review stars you invented.

Measuring without fooling yourself

Comparison pages get read cold, so judge them on what they were built to do: assisted conversions (did visitors who saw the page convert at higher rates within the buying window?), scroll-through to the concession and switching blocks (skimmers bouncing at the matrix are telling you something), and sales-team feedback ("prospects arrive pre-argued" is the qualitative win). For plan matrices, tier-selection rate and downgrade/support-ticket rates beat raw click-through. Whatever you test, do it with the discipline in CRO experiments you can believe — honest pages deserve honest measurement.

Key takeaways

The job of a comparison page is to earn trust, not declare victory. Credibility is the conversion asset.
Fair dealing: dated facts, respectful trademarks, self-disclosed weaknesses, maintained pages.
Concede something real ("if you need X, choose them"). One lost row buys eleven won ones.
Write rows as jobs the buyer cares about; answer cells with specifics, not bare checkmarks; group by decision.
Plan matrices: no padding rows, no fake "most popular," no hidden limits, transparent billing maths.
Publish three verified comparison pages, not thirty decaying ones.

FAQ

Is it legal to name competitors on my site? Broadly yes, when you're factual, non-deceptive and don't misuse their trademarks — nominative comparisons are standard practice. Get counsel to review the template once; then the maintenance burden is editorial, not legal.

What if we genuinely win every row? You probably chose the wrong competitor or the wrong rows. Compare against the option your buyers actually weigh — often "do nothing" or "hire in-house" — and you'll find honest asymmetry fast.

How often should vs pages be updated? Quarterly verification at minimum, with the "last verified" date on the page. A competitor's pricing change is an alarm to update, an opportunity to rank, and — left alone — a slow-motion credibility leak.

Do comparison pages work for agencies and services, not just SaaS? Yes — arguably better, because buyers of services compare approaches and trade-offs, not features. The "where they win" block is where a services firm proves it has taste. We use the same anatomy on our own pricing and engagement thinking.