SERP Analysis Template: Free CSV and Decision Worksheet

The CSV fields

Field group	Columns	Why it matters
Query context	query, query_cluster, review_date, country, language, device	Makes the capture reproducible and prevents unlike SERPs from being mixed
Result identity	rank, domain, url, page_type	Separates a company from the exact page competing for the query
Result-page interpretation	search_intent, serp_features, title_promise, visible_freshness	Records what the result page appears to reward without treating it as a permanent rule
Evidence	evidence_type, authority_context, internal_support	Distinguishes original proof and supporting signals from word count
Diagnosis	gap_type, observed_weakness	States the reason for action in a reviewable form
Action	candidate_action, impact_1_5, confidence_1_5, effort_1_5, risk, owner	Converts research into an owned decision
Measurement	measurement_start, measurement_end, notes	Prevents the team from declaring success immediately after publishing

Step 1: define the query cluster

Start with one exact query and a small set of variants that share the same job. Do not combine a software-comparison query, a definition query, and a pricing query merely because they contain the same noun.

Record the cluster label in plain language, such as:

• choose an AI SEO platform
• compare AI SEO pricing
• learn how to inspect a SERP
• find a SERP analysis worksheet

The label is an editorial working definition, not a claim about a search engine's internal classification.

Step 2: capture a clean result-page context

Record:

• exact query spelling
• date and time if volatility matters
• country and language
• desktop or mobile
• city when local results matter
• whether the browser was signed in or personalized
• whether the result came from a live browser, an API, or a vendor database

Different tools can return different positions because their settings and collection times differ. The template preserves those settings so later comparisons are meaningful.

If you need a tool, the SERP analyzer tools guide separates live snapshots, history, location controls, authority context, exports, and execution.

Step 3: record URLs, not only domains

A business competitor and a search competitor are not always the same. One SERP may include:

• vendor product or pricing pages
• editorial comparisons
• documentation
• community discussions
• video pages
• marketplaces or directories
• tools that answer the query directly

Record the exact URL and classify the page type. Repeated page types often reveal more than a list of famous domains.

Step 4: record the promise and proof

For each result, summarize the visible title promise in your own words. Then record the evidence available on the page:

• first-hand test or observation
• original dataset
• screenshots with a date and method
• worked example
• official documentation
• named expert review
• customer evidence with clear attribution
• no clear supporting proof

Do not mark a page as better simply because it is longer. Google recommends helpful, reliable, people-first content and warns against producing content primarily for search-engine manipulation. Use Google's people-first content questions as a quality check.

Step 5: classify the gap

Use one primary gap for the candidate action:

Gap	What it means	Typical response
Intent	The page solves a different job	Reframe, map a different owner, or create a distinct page only when justified
Content	The page misses a necessary answer, decision, or example	Add the missing information with evidence
Authority	The result set is dominated by stronger, repeatedly cited sources	Improve proof, expertise, distribution, and relevant links
Technical	Crawling, indexation, rendering, canonicalization, or eligibility blocks the page	Fix the technical defect before rewriting
Snippet / CTR	The page appears but attracts too few useful clicks	Improve the title, description, promise, or result appearance without changing the core intent

If two gaps matter, keep one primary and record the second in notes. A template with every gap selected is not a diagnosis.

Step 6: choose one candidate action

Write the action as a production change, not a topic:

• add a verified pricing methodology and dated evidence
• split a mixed-intent section into its correct owner page
• add contextual internal links from three relevant pages
• fix a canonical mismatch and validate the preferred URL
• improve a title after confirming a CTR problem
• defer the query because the authority gap is not realistic

Then score impact, confidence, and effort from 1–5 and assign a risk label. The numbers are prioritization inputs, not scientific probabilities.

Step 7: define the measurement contract

Before shipping, record:

• the baseline date window
• the exact page or pages changed
• the primary metric
• guardrail metrics
• the earliest review date
• the owner
• the rollback or correction condition

For an indexed page with impressions, the primary metric may be qualified clicks or qualified page-one impressions. For a technical fix, it may be successful validation and consolidation. For a new page, the first milestone may simply be discovery and indexing—not an immediate traffic claim.

What not to infer from one template

One SERP capture cannot prove:

• permanent rankings
• causal ranking factors
• exact search volume
• a vendor's paid-data accuracy
• that copying the current winners will rank
• that a recommendation will increase traffic or revenue

The worksheet makes the observation reviewable. It does not turn an observation into certainty.

Frequently asked questions

How many results should a SERP analysis include?

Record enough results to identify repeated page types, promises, and evidence patterns. Ten organic results are a practical manual starting point, but the correct depth depends on the result page and the decision.

Should I include ads and AI features?

Record any visible element that materially changes the user's path, but keep paid and organic results distinct. The SERP feature analysis guide provides a focused inventory method.

Is this a content brief?

No. It is the evidence layer before a brief. A content brief should be created only after the team confirms the page type, intent, and realistic gap.

Can I use the template with a SERP API?

Yes. Record the provider, collection settings, and timestamp in notes, and keep raw provider output separate from normalized interpretation.

How often should I repeat the analysis?

Repeat it before a material page decision and after the agreed measurement window. For ongoing tracking, use the SEO competitive benchmarking workflow.

Use the template to make a decision

A finished worksheet should end with one of four outcomes: update, create, fix, or defer. If it ends with a larger spreadsheet and no owner, the analysis is incomplete.