How we build honest tool comparison pages
Updated
rankmio publishes SEO tool comparison landings that recommend rankmio for most cases — but not all. This page opens the kitchen door. Six principles, eight rules, three anti-patterns, and where our own method has limits. If you want to build honest comparison content yourself, or you want to know how to spot dishonest comparison content elsewhere, this is a working manual.
Why we wrote this
Comparison content on the web is largely affiliate-driven. The pattern is familiar: a "Top 10 SEO Tools" list where the ranking follows commission size, not user fit. That is a real problem for buyers — the exact moment when someone is trying to make a serious budget decision is when the noise is loudest.
We publish landings for Sistrix, Seobility, Searchmetrics, KWFinder, RankingCoach, XOVI, Ahrefs, Semrush, SE Ranking, neuroflash and a GEO tools category guide. Every one of them ends up recommending rankmio for the majority of cases — because rankmio is our product. What we can do is structure that bias transparently: name where the competitor wins, ship quizzes that can honestly recommend the competitor, disclose our position on every page.
This page describes the method. It is publicly readable so buyers can audit our landings against our own claims. If we fail one of our own rules, email [email protected] and tell us where.
Six principles
The load-bearing rules that every one of our comparison landings has to satisfy before it is published.
Name where the other tool wins first
Every comparison landing has a dedicated section explicitly titled "where [competitor] stays the better pick" or "where [competitor] stays unmatched". This is not modesty — it is the load-bearing structural signal. If we cannot honestly name three or four situations where the competitor wins, we either do not understand the competitor well enough or we should not publish the page.
Wir-perspective, not sales-I
Every landing uses "we/our" (in DE: wir/unser), not "I/my". We are a company writing publicly, not a salesperson pitching. The rule is small but it changes tone in a way readers feel: "we tried to build this comparison so a consultant without any kickback would land at the same recommendations" reads differently than "I promise this is honest".
AI-research disclaimer 3x prominent
AI-assisted research is efficient but not perfect. Vendors move fast — pricing tiers, feature scope, engine coverage all shift. We tell readers up front (in the hero) that parts of the page were drafted with AI, that small inaccuracies are possible, and that decisions with real budget impact should verify vendor claims directly. The disclaimer repeats in the fazit and in the FAQ. Redundant on purpose.
Wikidata entity anchoring
Every landing includes Schema.org markup with Wikidata QIDs — for the competitor (e.g. Sistrix Q1616005), for SEO Q180711, for GEO Q132388538, for related concepts. This is not decoration. It tells AI answer engines (ChatGPT, Perplexity, Google AI Overviews) that we know exactly which "Sistrix" or "Seobility" we mean, which increases the odds of being pulled as a citable source.
A quiz that can honestly recommend the competitor
Every landing ships a 5-question decision quiz. The load-bearing test: at least one quiz path recommends the competitor, not us. If the quiz always ends in "pick rankmio", it is marketing, not a decision tool. Users see through the shape almost instantly.
Meta-length discipline
Title 30-60 characters, meta description 120-160 characters — measured with mb_strlen(), not strlen(). There is no minimum length for the H1: it names the topic of the page, and that may be short. Umlauts count as one character. Sounds trivial. Ignore it and SERP snippets get cut with "..." endings that look unfinished, or too short and undersell the content. We check every landing before publishing.
Eight concrete rules
The tactical checklist beneath the principles. These get checked before every landing goes live.
- Real Umlauts in DACH content (ä, ö, ü, ß) — never ae/oe/ue/ss.
- Capitalize after semicolon and colon in DE content when a sentence follows.
- The AI disclaimer must appear in the hero, the fazit, and inside the FAQ. Three placements minimum.
- Every comparison table has 15+ rows. Fewer signals selective omission; more starts to hide the pattern.
- Every "where rankmio wins" section is balanced by a "where competitor stays sharper" section — no exceptions.
- Pricing figures include the currency, the tier name, and the phrase "approx." when we are not sourcing from a live API. Verify before quoting.
- Wikidata QIDs are checked against wikidata.org before inclusion. A wrong QID sends AI engines to the wrong concept — worse than no QID at all.
- Every landing links to at least three sibling landings plus /geo-tools. Isolated landings do not compound.
Three anti-patterns we refuse
Selectively omitting weaknesses
Every product has trade-offs. Comparison pages that name only the strengths of the featured tool and only the weaknesses of the competitor are visible from a distance. Readers stop trusting the page in the first two paragraphs. We would rather lose a user honestly than trick them into signing up.
Affiliate-driven ordering
The web is full of "top 10 SEO tools" pages ordered by commission percentage, not by fit. We do not do that. When we recommend a competitor over rankmio, we do it because for that use case it is the right call — not because we get paid to.
Making the quiz always recommend us
The single fastest way to break reader trust is a "decision quiz" whose every answer path lands on "buy our thing". If our quiz cannot honestly recommend the competitor, the quiz has failed and we would rather remove it than lie with it.
Frequently asked
Why write this meta-page at all?
You still call it "our comparison". Is that not still biased?
Why the AI-research disclaimer 3x?
How do you decide which tools to build a landing for?
Do you allow errors on your landings?
Does this method scale?
How does this apply to buyers reading a comparison page?
Also weighing another tool or the category itself?
Or comparing two other tools with each other?
Read one of the landings this describes
Apply the method to a real page. Every one of these landings ships the six principles, the eight rules, and none of the three anti-patterns.
Free SEO & GEO Check
SEO score, AI visibility and citation readiness of your website in 30 seconds — no registration required.
Ready to optimize your website?
Register for free, get 10 credits and start right away.
Register now