How to Write a Fair SaaS Competitor Comparison Page
A competitor comparison page converts best when it says out loud where the other tool wins. That is not a soft opinion: Google's own documentation for review and comparison content tells authors to "Discuss the benefits and drawbacks of something" and to "Cover comparable things to consider, or explain which might be best for certain uses or circumstances" (Google Search Central). Yet when I audited six live vendor comparison pages on 11 August 2026, exactly one — Plausible's page on Google Analytics — carried a section headed "What GA4 does better." Everyone recommends fairness. Almost nobody ships it. That gap is the opening.
If you are about to write your first /vs/ page, this is the practical version: which page to build first, how to word the concession so it strengthens the pitch instead of dissolving it, what evidence you need behind every comparative claim, and how often to re-check it before an accurate page quietly becomes a false one.
Six live pages, one real concession
Start with what the field actually looks like, because it sets a low bar you can clear cheaply. Plausible's page is the exemplar. Its headings run through the argument first — what changed in GA4, why the lighter tool wins on speed — and only then arrive at "What GA4 does better," where the page states plainly: "If you need deep custom attribution modeling or enterprise-scale reporting with SQL access, GA4, especially with BigQuery, has more depth." Further down, under "Is Plausible right for you?", it disqualifies itself in a specific list: "Plausible is not the right fit for every use case. If you need session replay, user-level product analytics, retargeting or free hosted analytics, read this before you decide."
Two competitors of Substack manage one honest sentence each. Ghost's comparison page tells beginners: "If you're just getting started and you're looking for a simple platform to start with, Substack is great!" (Ghost). Buttondown is even more direct about the thing founders fear conceding — price: "To be fair, if you're only sending free newsletters, Substack is going to be cheaper than us" (Buttondown).
The rest of the sample concedes nothing at all. Ahrefs' page on Semrush is a strong differentiator list with no passage on where the other tool leads. PostHog's Mixpanel comparison has the best publishing hygiene I found — dated 28 January 2026, two named authors — but no "when Mixpanel is the better choice" section. And the ten pages in Powered by Search's showcase of the best SaaS comparison pages — Duda, ActiveCampaign, Shopify, Webflow, Asana, Zendesk, Klaviyo and others — contain no explicit acknowledgment of a competitor strength between them.
None of that is a failing on those companies' part; it is simply the industry default, and defaults are where differentiation is cheapest. If one honest paragraph is the thing your entire competitive set is missing, you can add it this afternoon. Which raises the prior question a careful founder should ask first: is the page worth building at all?
Why the vs page still earns its place in 2026
It is worth building because of where your buyer already is. In the TrustRadius 2026 B2B Buying Disconnect — 1,862 technology buyers and 444 vendors, surveyed in January 2026 — 83% of buyers shortlisted three or fewer products (HG Insights). The comparison is happening with or without you. The only question is whether it happens on a page you wrote or in a tab you never see.
The conversion data backs the intent read. In an analysis of 95 articles written for clients, Grow and Convert found comparison and alternative keywords converted at over 7.5%, and "versus" keywords at over 5% — the two highest-converting keyword categories they measured (Grow and Convert). For a founder used to blog posts converting under 1%, that is a different asset class. And the volume trap is worth naming: in the same write-up, six comparison articles targeting keywords with fewer than 20 monthly searches each collectively drove 149 organic signups. Your keyword tool will call "Yourtool vs Competitor" a 10-search-a-month term not worth writing. Ten searches a month from people comparing you to one named rival is not a traffic number, it is a pipeline number.
The ground underneath these pages has also shifted in your favour. Third-party review platforms used to absorb this demand and now absorb much less of it: SE Ranking's study of 30,000 commercial keywords, with AI Overviews captured on 1 December 2025 in the US, recorded organic traffic declines of 92.2% for TrustRadius, 89% for Capterra and 84.5% for G2 between 2024 and 2025 (SE Ranking). In the same dataset, vendor, e-commerce and other sites supplied 72.5% of all AI Overview links and appeared in 99.1% of AI Overviews, against review platforms at 8.5% of links. That layer is also consolidating: G2 announced on 29 January 2026 that it is acquiring Capterra, Software Advice and GetApp from Gartner (PR Newswire).
They still hold real estate — a March 2025 study of 1,485 software keywords found G2 in 6.26% of top-10 results, at an average CPC of $11.54 (Minuttia). Read that CPC as the rental price of traffic your own page could own outright.
One caveat before you over-index on any of this. A survey of 52 content marketers published in November 2025 found only 35.8% said comparison pages performed better than ever in 2024, while 49.1% saw no change and 15.1% reported declines (Intergrowth). Small sample, and the author says so. The framing that survives it comes from the same piece: comparison content exists "first and foremost to convince and convert," with organic traffic secondary. Build the page for the buyer who lands on it, not for the rank.
Google's documentation asks you to name who else might be better
Here is the part most comparison-page advice skips, and it is the strongest argument for writing an honest page rather than a triumphant one.
Google's reviews system covers "articles, blog posts, pages or similar first-party standalone content written with the purpose of providing a recommendation, giving an opinion, or providing analysis," and explicitly names "head-to-head comparisons" and "ranked recommendation lists" (Google). Your own vs page is in scope. It is assessed primarily at page level, but for sites with a substantial amount of review content, "any content within a site might be evaluated" — so a hub of thin comparison pages is a site-level bet.
Now read the second half of the guidance quoted at the top of this post again — "explain which might be best for certain uses." Google is asking a comparison author to state when the other product fits better.
The helpful content self-assessment adds the enforcement mechanism: does the content "clearly demonstrate first-hand expertise and a depth of knowledge (for example, expertise that comes from having actually used a product or service)," and does it "present information in a way that makes you want to trust it, such as clear sourcing"? The operational translation is blunt — you have to actually use the competitor's product, and you have to show your sources. A founder who has spent a week inside a rival's onboarding writes sentences no feature table can fake, and those sentences are what makes the page both rank and convert. For the paragraph-level mechanics of turning sourced claims into citable ones, our guide to getting cited by ChatGPT and AI Overviews goes deeper.
Which page to build first
"Comparison page" covers four different jobs, and confusing them is the most common structural mistake. A vs page answers "which of these two should I pick?" An alternatives page answers "what else is out there?" A [competitor] alternatives page, written from the rival's brand angle, catches dissatisfaction-driven switchers. And a category round-up answers "what are my options?" much earlier in evaluation (The Stacc).
Build three, not thirty: your top direct competitor's vs page, your largest "alternative to" opportunity, and one category comparison that frames your positioning. The vs page for your closest rival goes first, because it carries the clearest switching intent and the shortest path to a signup. And pick by sales conversation, not search volume — what you think matters much less than which two names your prospects keep saying on calls.
If you are the small player, there is a second move worth knowing. Instead of only writing "You vs Giant," target the query two larger rivals fight over — "Competitor A vs Competitor B" — and enter as a credible third option, because a three-way comparison that includes an unexpected third name is genuinely more useful to the searcher than a two-way one (Averi). It only works if you are honest about the two incumbents, which is the whole thesis in miniature. Then link to it from your main navigation rather than the footer — Teamwork.com reported a 54% sign-up increase after making that move (GetUplift). One company's anecdote, but a page nobody can find converts nobody.
How to concede without losing the deal
This is the section founders actually stall on. Admitting a weakness feels like handing over the deal. The research says the effect is real but dosed — and the dosing rules are specific enough to follow.
Eisend's meta-analysis of two-sided advertising found that the persuasive impact of message sidedness depends on how much negative information you include, where it sits, and whether you volunteered it — and it supports a curvilinear relationship between the amount of negative information and brand attitude (two-sided advertising meta-analysis). Curvilinear is the operative word: there is an optimal dose, and more candour is not linearly better. One specific, well-placed concession helps. A page that apologises in every section reads as a product that does not know what it is for.
Placement has its own evidence. Ein-Gar, Shiv and Tormala's "When Blemishing Leads to Blossoming" (Journal of Consumer Research, 2012) found that "people will be more favorably disposed to a product when a small dose of negative information is added to an otherwise positive description" — but with boundary conditions: the effect appears when the negative information follows the positive and is processed with low effort (Journal of Consumer Research). That boundary matters, because this study gets over-quoted. A buyer on a vs page is comparing on purpose — high-effort processing, exactly the condition where the effect does not fire. So don't treat "admitting flaws makes people buy more" as a law. Take the credibility argument from the sidedness literature, and take one practical rule from the blemishing work: concede after you have argued, not before.
Which is exactly Plausible's page structure. Argument, then "What GA4 does better," then "Is Plausible right for you?" with a concrete list of who should not buy. Notice that the self-disqualification is a feature list, not a vague hedge. "Not for everyone" persuades nobody. "If you need session replay, retargeting or free hosted analytics, this isn't it" tells the wrong-fit buyer to leave and tells the right-fit buyer that everything else on the page can be trusted.
There is a further trick in how you frame it: the same fact serves both vendors. ClickUp treats feature richness as the advantage; Basecamp treats simplicity as the advantage (GetUplift). Your "we don't have as many integrations" is also "we have less to configure." State the fact, then say who each side of it suits — that is not spin, it is the segmentation your page exists to do.
The reason all of this matters more in 2026 than it did in 2022 sits back in the TrustRadius data: 63% of buyers used AI during the purchase journey, and 94% of those buyers fact-check the AI's responses at least sometimes. Your claim about a competitor's pricing will be checked, probably in the next tab, probably within a minute. A comparison table that is 3% flattering to you is not a marketing choice any more; it is a discoverable error. As one practitioner guide puts it, "Overclaiming is the fastest way to lose credibility with both the reader and with AI engines" (Firebrand Marketing).
The evidence file behind every claim
Naming a competitor is not the legal risk. Not being able to back up what you said about them is.
The FTC's policy statement on comparative advertising "encourages the naming of, or reference to competitors," holds that "industry self-regulation should not restrain the use by advertisers of truthful comparative advertising," and opposes any code that imposes a higher substantiation standard on comparative claims than on ordinary ones (16 CFR 14.15). Truthful comparison is explicitly protected ground. The exposure is substantiation — and civil penalties under 16 CFR 1.98 run up to $53,088 per violation for penalties assessed after 17 January 2025 (Luthor's FTC comparative advertising guide).
Practically, that means building an evidence file organised by claim type, each entry dated:
- Price claims — the competitor's plan, billing period, contract terms, any discount window, plus the date you checked and a screenshot.
- Performance claims — test method, sample size, environment, and the benchmark date. Reproducible or don't publish it.
- Feature claims — the competitor's current capability at a named plan level, since "they don't have X" is usually wrong at enterprise tier.
- Rating or ranking claims — source, sample, date, and whether it is representative.
- AI capability claims — what the system actually does, where a human is still involved, and the limits.
Three traps catch first-timers specifically:
Switcher testimonials are regulated. The FTC's Consumer Reviews and Testimonials Rule took effect on 21 October 2024 and prohibits fake or AI-generated reviews, reviews from people with no actual experience of the product, positive reviews bought with contingent compensation, and — the one that catches small teams — insider testimonials from officers, employees or their relatives without clear disclosure of the relationship, with penalties up to $51,744 per violation (Goodwin). Your co-founder's quote about how much better your tool is needs a disclosure, or it needs to go.
Star ratings you host are not eligible for rich results. If the entity being reviewed controls the reviews about itself, Google's review snippet documentation says those pages are ineligible for the star review feature; ratings "must be sourced directly from users," and you must not aggregate them from other websites (Google). Quote a third-party rating with its source and date instead.
Check the licence before you benchmark. Many commercial software licences carry a DeWitt clause. Microsoft's SQL Server EULA states: "BENCHMARK TESTING. You must obtain Microsoft's prior written approval to disclose to a third party the results of any benchmark test of the software" (Brent Ozar). If your page includes a speed test against a paid competitor, read their terms first — open-source projects generally have no such clause, commercial ones often do.
The realistic worst case is not a lawsuit anyway. Any company or non-governmental organisation can file a challenge with BBB National Programs' National Advertising Division, where the Fast-Track SWIFT process decides a single well-defined issue within 20 business days (NAD). A dated evidence file is the entire defence — an hour to build, minutes to maintain.
If you draft with AI assistance, this is where the assistant's habits matter most: a fabricated or drifted citation on a page that names a real company is a different order of problem than a broken link in a listicle. Magic Share checks every cited source before it hands over a draft and ships a verification report with it — the same discipline you would want from a research assistant here. Reviewing by hand instead? Our 20-minute fact-checking checklist covers the same ground manually.
Keeping it true: the update cadence
An evidence file has a shelf life. As the same FTC guide puts it: "Comparative claims drift when competitor prices change, features ship, rankings update, or benchmarks age. A campaign that was accurate in January can become misleading by June."
SaaS pricing in particular moves fast. Nearly 40% of SaaS companies had not revisited their pricing structure in the last 18 months, per OpenView's 2024 benchmarks quoted in a 2025 write-up (Monetizely; OpenView no longer serves the original). Flip it: roughly 60% did change pricing inside 18 months. If your page compares three competitors, the odds all three pricing tables are still right a year from now are poor.
A workable rhythm for a solo founder: monthly spot-checks of competitor pricing and any major feature shipping, quarterly full reviews of every comparison page, and a same-week update whenever a competitor ships something big or changes pricing.
Two cheap habits make that cadence visible. Put a "last verified" date on the page and attach a byline — of the pages I audited, only PostHog's carried both, and Ahrefs' one dated third-party citation ("as verified by Cloudflare," data as of July 2026) does more for trust than a whole column of green checkmarks. Both are what Google's helpful-content questions about sourcing and expertise are looking for.
The honest problem is that this is a calendar problem, not a writing problem, and calendars are what founders drop first. Whatever runs it — a recurring task, a diff-watcher on their pricing page, or an agent that wakes on a schedule and drafts the refresh — what matters is that the re-check leaves your head.
When not to build one
Three situations where the honest answer is "not yet."
Don't mass-produce them. Generating /vs/[competitor] pages for all forty names in your category is the pattern Google's spam policies describe: scaled content abuse is "when many pages are generated for the primary purpose of manipulating search rankings and not helping users," and doorway abuse is "when sites or pages are created to rank for specific, similar search queries" (Google). Three researched pages beat forty templated ones. We covered where that line falls in our read of Google's scaled content abuse policy.
Don't publish before your product can survive the comparison. A page written too early reads as a desperate sales move, and it hands a bigger rival free awareness inside your own niche — a real risk when your positioning is "radically different from X" rather than "better than X" (Sebastien). The build condition is simple: do prospects actually search for this comparison, and can you win a specific segment of it on the facts?
Don't expect AI answer engines to hand you the traffic. A University of Toronto paper on generative engine optimization reports that AI search engines "exhibit a systematic and overwhelming bias towards Earned media (third-party, authoritative sources) over Brand-owned and Social content," across ChatGPT, Perplexity and Gemini (arXiv:2509.08919). A benchmark of 100 cybersecurity vendors across 250 buyer-intent prompts and six AI platforms, run September 2025 to January 2026, found 73% of tested vendors received zero ChatGPT citations for category recommendations (GrackerAI).
If you don't yet have an audience to carry that traffic, how to promote blog posts with no audience yet covers the distribution steps that come before search and AI answers.
That does not contradict the SE Ranking figures earlier — different engines, different query classes. The honest position: your vs page is a strong bet in Google AI Overviews, a weak bet as your only play in ChatGPT, and reliably useful the moment a buyer lands on it from any source. Build it for the human doing the last comparison before they decide.
One last thing worth saying plainly: nobody has published a clean A/B test of a comparison page with and without a "when to choose them" section. The case for fairness rests on Google's documented preferences, the message-sidedness literature, and the fact that 94% of AI-using buyers verify what they are told. Strong chain — but not a measured lift, and treat anyone quoting one with suspicion.
Measuring this on your own site is harder than it sounds when traffic is thin, so how to measure blog attribution on low-traffic sites walks through telling whether a post actually drove signups.
FAQ
Should I link to my competitor's website from a comparison page? Ghost's and PostHog's pages don't, and no ranking rule forces you either way. The argument for linking is verifiability: if you cite their pricing page as the source for a price claim, a reader who can check it in one click is a reader who believes the rest of your page. If you'd rather not pass the link, cite the URL as plain text with a checked date.
How many comparison pages should a small SaaS start with? Three. Your closest direct competitor's vs page, your biggest "alternatives to X" opportunity, and one category comparison. Pick them from the names that come up in sales conversations rather than from search volume — comparison keywords with fewer than 20 monthly searches drove 149 signups in one documented set of six articles.
What if the competitor is genuinely better than us on most things? Then you are not ready to publish, and the page would tell your prospects so. The tactic works when you win decisively for an identifiable segment. Find that segment first, name it in the page, and let everyone else read your self-disqualification list and leave happy.
How often do I need to update a comparison page? Spot-check competitor pricing and major feature changes monthly, do a full review quarterly, and update within days when a competitor ships something significant. Roughly 60% of SaaS companies changed pricing inside an 18-month window, so a page left alone for a year is very likely wrong somewhere.
The short version
The comparison page is the highest-intent page you will ever write, and what makes it work is the thing most vendors leave out: a clear, specific statement of who should buy the other tool. Argue your case first, concede once, name the segment you do not serve, source every comparative claim with a date, and put the re-check on a calendar. That is closer to editorial discipline than to copywriting.
If keeping that discipline across a growing set of pages is the part you know you will drop, that is the problem Magic Share was built for: an agent that researches, drafts, checks every cited link, and then waits for your approval — nothing publishes on its own. Your first three posts are free — a fair place to test whether the discipline holds.