Search intent is the goal behind a query: to learn something, to reach a specific site, to compare options, or to buy. Google ranks the page format that satisfies that goal, so the fastest way to read intent is to study what already ranks and match that format before you build a single link.
Most pages that fail to rank aren’t badly written. They’re answering a different question than the one being asked. You can spend two months on a 3,000-word guide, point good links at it, and still sit on page four — because every result above you is a product category page and yours is an essay.
That’s an intent mismatch. It’s the cheapest ranking problem to diagnose and one of the most expensive to ignore, because no amount of authority fixes a page that’s in the wrong format. Here’s how to read intent properly, and what to do when you’ve already got it wrong.
What is search intent in SEO?
Search intent is what the searcher is actually trying to accomplish when they type a query. Google’s ranking systems are built to satisfy that goal rather than to reward the longest page or the one with the most backlinks. Two pages targeting the same keyword can land in wildly different positions purely because one matched the intent and the other didn’t.
The useful thing about intent is that it isn’t a guess. Google has already run the experiment, millions of times, on every query you care about. The results page is the published answer. If eight of the top ten results for a term are comparison listicles, the searchers behind that query have told Google — through clicks, dwell, and refinements — that a listicle is what satisfies them. Your beautifully argued opinion piece is arguing with data.
Intent also isn’t static. It drifts. A query that returned tutorials two years ago might return software pages now, because the market changed and the population of people typing that phrase changed with it. That’s why intent needs periodic re-checking rather than a one-time classification at the keyword research stage.
What are the four types of search intent?
The four standard categories are informational, navigational, commercial investigation, and transactional. Informational queries want knowledge, navigational queries want a specific destination, commercial investigation queries want help choosing, and transactional queries want to act now. Most keyword sets contain all four, and they demand completely different pages.
Informational. The searcher wants to understand something. “Why are my tomato leaves curling”, “how long does concrete take to cure”, “what is search intent”. These SERPs are dominated by guides, explainers, and increasingly by featured snippets and video. The page that wins is usually the one that answers the question in the first hundred words and then earns the rest of the reader’s attention.
Navigational. The searcher already knows where they want to go and is using search as a shortcut. “Gmail login”, “ahrefs pricing”, “royal mail track parcel”. You almost never want to target someone else’s navigational query — the brand owns it and will keep owning it. The exception is your own brand terms, which you should absolutely control.
Commercial investigation. The searcher intends to buy eventually but is still comparing. “Best CRM for small teams”, “wordpress hosting comparison”, “gas boiler vs heat pump running costs”. This is the most commercially valuable content category and the one most businesses handle worst, because they publish a sales page when the searcher wanted an honest comparison that includes options they don’t sell.
Transactional. The searcher wants to complete an action now. “Buy standing desk”, “emergency plumber Bristol”, “link insertion service pricing”. These SERPs favour product pages, category pages, service pages, and local packs. A blog post rarely wins here, no matter how thorough it is.
How do you read intent from the actual SERP?
Open an incognito window, run the query in the right country, and inventory the top ten results by page type rather than by domain. Count how many are guides, listicles, product pages, tools, or videos. Whichever format holds the majority is the format Google currently rewards, and that’s your brief.
Pay attention to the furniture around the results too. A shopping carousel means transactional. A “People also ask” block stuffed with definitional questions means informational. A local pack means the query has a geographic component whether or not the searcher typed a city name. Video carousels near the top usually mean the task is easier to show than to describe — recipes, repairs, software walkthroughs.
| Intent type | What you’ll see on the SERP | Page format that tends to win |
|---|---|---|
| Informational | Featured snippet, People also ask, long-form guides, Wikipedia, video carousel | Structured guide or explainer with a direct answer up top |
| Navigational | One brand dominating, sitelinks, knowledge panel, brand’s own subpages | The brand’s own page — don’t compete unless it’s yours |
| Commercial investigation | Listicles with numbers in titles, “best” and “vs” phrasing, review sites, comparison tables | Comparison post or ranked listicle with real selection criteria |
| Transactional | Shopping ads, product carousel, category pages, local pack, prices in titles | Product, category, or service page with pricing visible |
| Mixed | Two or three formats splitting the top ten roughly evenly | Hybrid page, or two separate pages targeting each variant |
One more signal worth reading: title tag phrasing. If seven of ten titles start with a number, the market has decided this is a list query. If they all include a year, freshness is being rewarded and a page you published in 2021 and never touched is quietly bleeding position. Titles are cheap intelligence and almost nobody reads them systematically.
Why does intent mismatch stop a good page from ranking?
Because relevance is judged against the query’s expected outcome, not against the topic in isolation. A page can be topically perfect, technically clean, and well linked, and still be the wrong kind of thing to hand someone. Google has better options and takes them.
This is why intent mismatch survives link building. When we look at a page a client says “won’t rank no matter what we point at it”, the problem is a format mismatch surprisingly often — a service page fighting a SERP full of tutorials, or a blog post fighting a SERP full of ecommerce categories. Adding authority to a mismatched page mostly buys you a slightly better position among the wrong results, and even that erodes. If you want the honest version: links amplify a page that already fits, and they can’t manufacture fit that isn’t there.
There’s a related trap worth naming. Some pages rank briefly after a burst of links, then slide back over a few weeks. That pattern usually means the page got a positional trial from Google, failed to satisfy the people who clicked it, and got quietly demoted. The links did their job; the page didn’t. If you’re weighing spend, it’s worth understanding what separates a link worth paying for from one that isn’t before you spend more on a page that can’t hold the position anyway.
How do you fix a page that targets the wrong intent?
You have three options: reformat the existing page to match the dominant SERP format, split it into two pages serving two different intents, or repoint it at a different keyword where its current format already fits. Rewriting in place is usually fastest, but only if the URL doesn’t carry conflicting signals from its title, slug, and existing links.
Reformatting is the common case. A guide that needs to become a comparison gets a criteria section, a table, and honest coverage of alternatives. A service page that needs to become informational gets the sales copy moved below the explanation. Keep the URL, keep the links, change the shape. If the slug itself screams the wrong intent — /buy-widgets/ trying to rank for “how widgets work” — publish a new URL instead and use the old one for the query it actually deserves.
Splitting is right when a page is trying to serve two masters. You’ll recognise it by the internal contradiction: half the page teaches, half the page sells, and neither half is as good as a page that only did one job. Split it, connect the two with a clear internal link, and let each one compete on its own terms. If the split creates overlap risk, a quick pass over your internal link structure will tell you which URL Google currently treats as the canonical answer for that topic.
Use this before you rewrite anything:
- Run the target query in an incognito window, in the right country and language
- Classify all ten organic results by page type, not by domain authority
- Note every SERP feature present: snippet, PAA, video, shopping, local pack
- Check whether titles carry numbers, years, or price signals
- Open the top three results and record what the first screen actually delivers
- Compare that against your page’s first screen, honestly
- Decide: reformat, split, or repoint — then write the brief before touching copy
- Record current position and impressions so you can tell whether the change worked
That last item matters more than it looks. Intent rewrites are the kind of change people make and then never measure, which is how teams end up rewriting the same page three times in a year. If you’d rather have someone else do the SERP classification and rewrite briefs across a whole site, that’s the sort of work our page-level optimization service exists for.
What do you do about mixed-intent SERPs?
A mixed-intent SERP is one where no single format holds a majority — say four guides, four product pages, and two videos. Google is hedging because the query population genuinely splits. Your job is to pick the segment you can win rather than trying to satisfy everyone in one page.
In practice there are three workable responses. Pick the underserved half: if the informational results are thin and the commercial ones are strong, write the better guide. Build a hybrid: lead with the explanation, then transition into the options, which works well for queries like “X vs Y” that carry both learning and buying motives. Or accept the SERP is a poor target and spend the effort on a cleaner query nearby, which is often the right call when the mix is a symptom of low search volume and unclear demand.
What doesn’t work is splitting the difference badly — a page that’s 60% guide and 40% pitch, satisfying neither group. Mixed intent rewards commitment to one shape with a graceful exit toward the other, not a blend.
How often should you re-check intent on important pages?
Once or twice a year for stable topics, and immediately after any noticeable ranking drop that isn’t explained by a technical change. Intent drift is slow but real, and it’s a common cause of a page that used to rank well quietly sliding out of the top ten over six months.
The check takes five minutes per page. Run the query, count formats, compare to your notes from last time. If the mix has shifted — more product pages than there used to be, a video carousel that wasn’t there before — the SERP is telling you the audience changed. Rewrite or accept the decline. A page that’s been drifting for a year is also worth checking against whether the links pointing at it are still live, since decay often has two causes stacked on top of each other.
Key takeaways
- Intent is the goal behind the query; the SERP is Google’s published answer about which format satisfies it.
- Classify the top ten by page type, not domain — the majority format is your brief.
- Intent mismatch is the most common reason a well-written, well-linked page won’t rank.
- Fix by reformatting, splitting, or repointing — and record baseline metrics before you start.
- On mixed-intent SERPs, commit to one format and serve the other with a clear exit link.
- Re-check intent on money pages once or twice a year; intent drifts, and pages decay with it.
Can a page rank for two different intents at once?
Sometimes, but usually only on low-competition queries. On a competitive SERP, a hybrid page tends to lose to two specialised pages — one built for the informational variant, one for the commercial. If a single URL is picking up both, check whether the traffic is actually converting or just accumulating.
Does matching intent mean copying the top result?
No. Match the format, not the content. If the top five are all comparison tables with selection criteria, build a comparison table with better criteria. Copying the substance gets you a page with nothing to recommend it over the original, which is a worse position than being different and useful.
How does intent affect anchor text choices?
Commercial and transactional pages attract more exact-match anchors naturally, which makes them the pages most likely to end up with a skewed profile. Our own working target is roughly 40-50% branded and only 5-10% exact-match, and commercial pages are where that balance slips first.
My page matches intent but still isn’t ranking. What now?
Then you’ve eliminated the cheapest cause, which is progress. Next look at whether the page is genuinely more complete than the results above it, whether it’s internally linked from anywhere that matters, and whether the domain has enough authority to compete on that SERP at all. Under roughly DR 20, most competitive queries are out of reach regardless of page quality.