Four habits worth stealing from the big SEO suites
Semrush, Ahrefs and SE Ranking disagree about almost everything except four things, and those four are the ones worth taking. None of them require their software. All of them are things most sites still do not do.
One: treat a competitor's site as a to-do list
The single most useful feature any of them ship is the gap view: the queries where a rival appears and you do not. It works because it replaces a question nobody can answer — what should we write — with one anybody can act on: here are forty subjects three competitors cover and you do not.
You do not need a subscription to run a crude version of this. Take the three sites you lose to, list their commercial pages, and mark the ones you have no equivalent of. The crude version finds most of what the expensive version finds, because the gaps that matter are usually obvious once somebody writes them down.
Two: group queries into one page, not one page each
Every serious suite clusters queries before it recommends anything, because two phrasings that return broadly the same results are one page, not two. Splitting them produces a pair of thin pages that compete with each other and beat neither.
The grouping rule the suites use is comparison of the actual results, not similarity of the words. That distinction is the whole trick, and it is why keyword lists sorted alphabetically have never worked.
Three: measure the page, not the site
Site-wide scores exist to be screenshotted. Every decision that changes revenue is made at the level of one page: this page earns, that page does not, this one has slipped four positions on the query that pays. The suites that are genuinely useful let you pivot everything to a single URL; the ones that are not lead with a number out of a hundred.
Four: keep the history, and keep it honest
The reason people stay subscribed for years is the archive. Knowing that a page was at position eleven in March, that you rewrote it in April, and what happened in May is the only way to tell whether any of the work was worth doing. A tool with no memory can never answer the only question that matters.
Honest history means recording the change as well as the outcome. A ranking chart with no record of what you did to the page is a weather report.
The habit they all encourage that costs you money
All three sell attention by the metric. The reports are organised around traffic, rankings and scores, because those are the things the software can measure about everybody. So the work gets prioritised by search volume — and search volume is not value. A query with ninety searches a month from people holding a credit card outearns one with nine thousand from people writing an essay.
The fix is not a different tool, it is a different ordering. Put a number in pounds against each piece of work before you do it — even a rough one, built from your own conversion rate and average order value — and do them in that order. Most sites find the list rearranges completely.
What to do about it
- Write down the pages your three closest rivals have and you do not. That list is your next quarter.
- Group queries by whether they return the same results, then build one page per group.
- Judge everything at the level of a single URL. Ignore scores out of a hundred.
- Record what you changed alongside what happened, or you will never know which half worked.
- Order the work by estimated revenue, not by search volume.