SERP Analysis: Reading the Results Page Before You Write a Word
- Filed November 10, 2026
- Keyword research, guide 6 of 6
The results page for a query is the most direct evidence of what kind of answer a search engine currently shows for it. Reading it before writing tells you which format fits, which features compete for attention, what the ranking pages cover and where there is room for something better. It does not tell you why pages rank, and it is a snapshot that varies with location, device and time. Used as evidence about intent and format, SERP analysis prevents a costly content mistake: writing the wrong kind of page.
Before you look: control the conditions
Results change with location, language, device and personalization. Google’s description of ranking says its algorithms consider “the words of your query, relevance and usability of pages, expertise of sources, and your location and settings” (How Google Search ranks results). So:
- use a signed-out or private session;
- set the location to the market you serve;
- check mobile and desktop if both matter;
- record the date.
An annotated SERP analysis framework
Work through the page from top to bottom and fill in each field.
1. Features above and among the results
| Feature | What it suggests | Effect on organic clicks |
|---|---|---|
| AI Overview | Google judged an overview additive for this query | Some answers may be satisfied on the page |
| Ads | Commercial value | Pushes organic results down |
| Local pack | Local intent | Organic results below a map |
| Shopping units | Product purchase intent | Commerce pages compete with listings |
| Featured snippet | A concise answer is wanted | One result is highlighted |
| People also ask | Related questions | Shows adjacent intents |
| Video or image results | Visual format is useful | Text-only pages may underperform |
| Top stories | News-oriented query | Freshness matters |
| Forums and discussions | First-hand experience is valued | Personal perspectives compete |
Google’s AI features documentation notes that AI Overviews appear “only when our systems determine that it is additive to classic Search” (AI features and your website), so their presence is itself a signal about the query.
2. Page types in the organic results
For the top results, classify each page: product page, category page, comparison or review, how-to guide, definition or glossary, tool or calculator, forum thread, video, news article, government or institutional page.
If most results share one type, that is the format Google currently shows for the query. If they are mixed, the query has mixed intent.
3. Angle and depth
For each ranking page, note:
- the angle (beginner guide, expert reference, local, price-focused, comparison);
- what it covers and what it leaves out;
- its freshness (visible dates, current facts);
- who publishes it (brand, publisher, individual expert, institution).
4. Competitors’ site types
Are the ranking sites large brands, specialist sites, small publishers, forums? A results page made entirely of large brands and institutions is a different opportunity from one with small specialist sites. Third-party difficulty scores try to estimate the same thing from link data.
5. Gaps
List what a reader would still need after reading the top results: an unanswered sub-question, outdated information, missing local detail, no worked example, no first-hand testing. Gaps are where a new page can add something, provided you can fill them with real information.
6. Decision
Write a one-paragraph brief: the intent, the format, the angle, the gaps you will fill, and the features you must account for. If the honest conclusion is that you cannot produce a better result, say so and choose another query. If one of your own pages already ranks for it, decide first whether a new page would compete with it, the problem known as keyword cannibalization.
A complete annotated example
Imagine a company planning an article for the query “how to descale an espresso machine”, checked signed out from a US location on mobile. The observations are illustrative.
| Field | Observation | Implication |
|---|---|---|
| Features | AI Overview with steps; video carousel; "People also ask" about vinegar vs descaler and frequency | Clear steps alone are already on the page; video matters |
| Page types | Manufacturer support pages (3), how-to guides from coffee publishers (4), a forum thread (1), a video page (2) | How-to format; manufacturer instructions valued |
| Angle | Generic steps; few mention differences between machine types | Machine-specific guidance is a gap |
| Freshness | Mixed; two guides recommend products no longer sold | Current product references would help |
| Site types | Brands and coffee publishers | Coffee-specialist sites compete here |
| Gaps | Differences by boiler type; how to tell when descaling is due; water hardness; warranty implications of vinegar | Material for a better page, if sourced from manufacturer documentation |
| Decision | A how-to guide organized by machine type, with a short video, citing manufacturer instructions for each type, and a section on water hardness | Owns the "by machine type" angle |
What SERP analysis cannot tell you
- Why pages rank. Correlation between a page feature and its position is not a cause.
- Click distribution. It shows what is on the page, not how users click.
- Stability. Results change; recheck before and after publishing.
- Content gaps for competitor sites in general. Analyzing which topics competitors cover that you do not is a separate exercise from reading the results for one query.
Turning analysis into a brief
Carry the decision directly into the content brief: target query, intent, format, angle, required sections that fill gaps, features to account for (video, concise answer for a snippet, local details), and sources to use. A brief that skipped the results page is guessing at all of these. If the gaps add up to several distinct intents, plan them as a topic cluster rather than one page.