SERP stands for search engine results page. It is the page you get back after typing a query into Google, Bing or any other search engine, and it holds every result the engine decided to show you, organic listings, ads, images, videos and everything in between.
That definition is the easy part. The useful part is learning to read a SERP, because what sits on the page decides whether a keyword is worth targeting at all and what ranking for it would actually be worth.
What a SERP is
A SERP is generated fresh for each query. Nothing on it is a fixed list that the search engine keeps lying around, which is why the same query can return a different page an hour later.
The organic results are the listings a search engine ranks on relevance rather than payment. Everything else on the page, including ads and the various boxes and panels, competes with those organic listings for the reader’s attention.
People use SERP loosely to mean “the rankings”, as in “we moved up the SERPs”. That is fine in conversation, but it hides the thing that matters most, which is that your position is only one part of what the page looks like.
What Google actually puts on a results page
Google documents its own result elements publicly, and the names it uses are worth knowing because they are the names in its documentation and reports.
- Text result. The standard blue link listing, built from an attribution, a title link and a snippet. Source: Google Search Central.
- Attribution is itself made of a favicon, a site name and a visible URL, which splits into domain and breadcrumb.
- Rich result. A listing enhanced with structured data, which Google calls rich attributes when they appear inside a text result.
- Image result and video result, each with their own thumbnail and attribution, and an upload date in the case of video.
- Sitelinks group. The cluster of extra links that sometimes sits under a listing.
- Exploration features, which is Google’s term for the related searches group and the related questions group, the one most people call People Also Ask.
Two things in that documentation matter more than the list itself. Google states that appearance varies by device, country, query language and other factors. It also says exploration features are generated automatically and that you cannot control what appears in them.
Google is also explicit that no single listing carries everything: in its words, you will not see a text result that includes all of the possible visual elements.
Why no two SERPs look the same
Because the page is assembled per query, the layout you see is not the layout your client sees, and neither may be what the rank tracker recorded.
Ads are a good example. Google says the top ad location is generally above search results, but how many ads appear there is decided by an auction with a minimum Ad Rank threshold, so a commercial query can push the first organic listing well down the page while a research query leaves it near the top. Source: Google Ads Help.
Device changes it again. The same page that shows an organic result above the fold on a desktop monitor can bury it on a phone, where the fold arrives after a couple of boxes.
How to read a SERP before you target a keyword
This is the part most keyword research skips. A keyword tool reports volume and a difficulty score, and neither one tells you what kind of page Google wants to rank.
Open the results and count what type the top ten actually are. The answer decides whether writing an article is even the right move.
- Mostly working tools? The intent is transactional and an article will not displace them. A query where ten of ten results are live checkers or calculators needs a tool, not prose.
- Mostly one brand’s own pages? The query is navigational. People searching it want that company, and you are competing for scraps.
- Mostly service and pricing pages? The searcher wants to buy, and a guide will sit outside the intent.
- Mostly articles and guides? Now volume and difficulty are worth looking at, because an article can win.
Then look at who those articles belong to. If every result is a large brand, note it and decide whether you have the authority to enter. If a small site is sitting in the top ten, that is the single most encouraging signal on the page.
Doing this takes two minutes and it prevents the most expensive mistake in content planning, which is writing a good article for a query that was never going to rank one. It is also the check to run before you build any SEO forecast, since a forecast built on unwinnable keywords is wrong before the arithmetic starts.
What position one is actually worth now
The number one organic result draws around 27% of clicks, according to Backlinko. At the same time, around 60% of Google searches now end without a click to any website, according to SparkToro. Both figures and their sources sit in our SEO statistics roundup.
Those two numbers describe the same page from opposite ends. Ranking first is still worth far more than ranking fourth, and ranking first on a query that resolves on the results page is worth much less than it used to be.
So judge a keyword by the shape of its SERP, not only by its volume. A 500-search query on a clean page of blue links can send more traffic than a 5,000-search query answered in a box at the top.
SERP features you can influence, and ones you cannot
Treating every feature as an opportunity wastes effort, because some of them are not yours to win.
- Rich results are the most directly influenceable, since they are driven by structured data you add to your own pages.
- Sitelinks are generated by Google from your site structure and internal linking, so you influence them indirectly at best.
- Related questions and related searches cannot be controlled at all, by Google’s own statement. They are still useful, because they tell you what Google associates with the query.
- Ads are not an SEO lever. You can only account for the space they take.
The practical move is to read the uncontrollable features as research rather than targets. The related questions box is a free list of what readers actually want answered, and answering those questions inside your page is useful whether or not you ever appear in the box.
How to track where you rank
Search Console reports your own average position for every query you appear on, using your real impressions rather than a simulated check. It is the most accurate source you have for your own site, and it is free.
Third-party rank trackers add things Search Console does not, such as competitor positions and which features appear on the query. They check from a chosen location on a schedule, so they are a sample rather than a record of what your visitors saw.
Use both and expect them to disagree. Our guide to SEO tracking tools covers the options, and striking distance keywords explains how to find the positions worth chasing first.
Frequently asked questions
What does SERP stand for?
SERP stands for search engine results page. It is the page a search engine returns after someone enters a query, containing the organic listings along with ads, images, videos and any other elements the engine chooses to show. The page is assembled fresh for each search rather than pulled from a stored list.
What is the difference between a SERP and organic rankings?
Organic rankings are only one part of a SERP. The results page also holds paid ads and various boxes and panels, all competing for the same attention. This is why being first in the organic rankings and being first on the page are not the same thing.
Why do SERPs look different for different people?
Google states that result appearance varies by device, country, query language and other factors. The number of ads shown is also decided by an auction rather than being fixed, so a commercial query can push organic listings much further down than a research query does. Your rank tracker checks from one chosen location, which is why its view can differ from yours.
How do I know if a keyword is worth writing about?
Open the results and look at what type of page occupies the top ten. If they are mostly live tools, one brand’s own pages, or service and pricing pages, an article is unlikely to rank no matter how good it is. If they are mostly articles, and especially if a small site appears among them, the keyword is worth considering.
Can you control which SERP features you appear in?
Only partly. Rich results come from structured data on your own pages, so those are the most directly influenceable. Sitelinks are generated by Google from your site structure, and Google says related questions and related searches are generated automatically and cannot be controlled.
What is the best way to track SERP positions?
Search Console is the most accurate source for your own site, because it reports average position from real impressions rather than from simulated checks, and it is free. Third-party trackers add competitor positions and feature data, but they sample from a set location on a schedule. Using both and expecting some disagreement between them is the realistic approach.