How Rankpax works

Rankpax turns search data and website checks into plain-English recommendations. This page explains where every number comes from, how recommendations are chosen and what the limits are — so you can judge how much weight to give them.

Search data

• Source. Keyword demand, difficulty, intent, related searches, Google results and the searches a website ranks for come from DataForSEO, a licensed search data provider.
• Country. All research and rank checks use Google in the United Kingdom, in English.
• Estimates, not traffic. Monthly searches are averages of the last 12 months. They are estimates, not your website's visits.
• Freshness. Keyword data is reused for up to 30 days and Google results and competitor data for up to 7 days, because these change slowly. Every report shows when its data was collected and whether it's a fresh collection or a cached result. Refreshing a report collects new data and uses one search.

Difficulty and intent

• Difficulty is the provider's 0–100 estimate of how hard it is to reach the first page. Rankpax describes it as: under 20 easier to rank, 20–39 moderate, 40–64 competitive, 65+ very competitive.
• Intent is the provider's classification of the main purpose of a search, shown as Learning (informational), Comparing (commercial), Buying (transactional) or Navigating (looking for a specific site).

How the opportunity is chosen

Every keyword report highlights one more specific opportunity. From the related searches with at least 50 estimated monthly searches, Rankpax prefers those with lower difficulty than your original search and ranks them by monthly searches ÷ (difficulty + 10)^1.5. This favours searches with real demand that a smaller website has a fair chance of winning. It's a starting point: always check the search fits what you actually offer.

How next steps are chosen

The three next steps come from simple, consistent rules rather than guesswork:

• plan a page for the opportunity above, when there is one;
• if your website already appears in the results, improve that page; otherwise match the search's intent (a guide, a comparison or a product/service page);
• match the format of what already ranks, based on whether the top five results look like guides or shop pages;
• for very competitive searches (difficulty 60+), start with more specific searches.

Competitor keyword gaps

Rankpax compares the top 300 searches (by demand) that a competitor ranks for with those your website ranks for. Only searches where the competitor is in the top 20 are shown. Each is labelled Missing (your site wasn't found in that data — not a position of zero), Weak (you rank below 20) or Shared (you both rank in the top 20). Searches for the competitor's own brand name are left out.

Website audits

• Crawling. RankpaxBot starts from your homepage and sitemap, follows links on your own domain, respects robots.txt, checks up to four pages at a time and stops at your chosen page limit. Each page has 12 seconds to respond; up to 5 redirects are followed. Only public websites can be checked.
• What's analysed. The HTML your server returns. Content added later by JavaScript isn't analysed.
• Checks. 17 rule-based checks with fixed thresholds — for example titles under 15 or over 65 characters, pages taking over 1.5 seconds to respond, or fewer than 150 words. The full list: http error, broken link, missing title, duplicate title, title length, missing meta description, duplicate meta description, missing h1, multiple h1, noindex, redirect chain, slow response, missing viewport, missing alt, thin content, missing lang, not https.
• Priority. High priority means visitors or search engines hit the problem directly (errors, broken links, missing titles, no HTTPS); medium and low are improvements.
• Rechecks. A recheck inspects only the affected pages. A finding is marked verified only when every affected page passes.

Rank tracking

Once a week Rankpax checks the top 100 organic Google results in the UK for each tracked search and records where your domain appears. If it isn't there, it's recorded as not found in the top 100. If a check can't run, that week is left as a gap — Rankpax never fills in or estimates missing positions. Positions can differ from what you see personally because of location, device and personalisation.

Content briefs

Briefs are drafted from your research — the search's intent, related searches and the pages currently ranking — using fixed rules for structure. They're a starting point for you or your writer; check facts and claims before publishing.

Allowances

Every metered action reserves its allowance first and is charged only if it succeeds; failed searches are restored automatically. Audit pages that can't be fetched are held, not charged, and can be retried without a second charge. There are no automatic overage charges. Full details are on the pricing page.

What Rankpax doesn't do

• It doesn't guarantee rankings, traffic or revenue.
• It doesn't change your website — you make every change yourself.
• It doesn't yet include Google Search Console data (your real clicks and impressions).

Questions about our methods? Contact us.