How to Label SERP Screenshots for a Technical SEO Audit

Label a SERP screenshot with its capture time and timezone, search engine, exact query, location or locale, device, capture source, and an evidence ID. Keep an untouched original, link it to the audit finding, and store longer context in an adjacent record rather than relying on a shortened filename. A screenshot is point-in-time visual evidence; it does not establish ranking or organic performance by itself.

1. Define what the screenshot represents

A SERP screenshot is an image of a search results page captured in a particular context. Its label should let a reviewer understand which query and search context the image represents.

Do not confuse it with a Google URL Inspection screenshot. Following a successful live test, URL Inspection may show a screenshot of the page as rendered by Google’s inspection tool. It is not a screenshot of search results. Google says screenshots are unavailable for the indexed URL and for an unsuccessful live fetch. Label this evidence explicitly as a Google URL Inspection live-test screenshot, and record the inspected URL, capture time, and live-test state. Google’s URL Inspection tool guidance explains the report and rendered screenshot.

URL Inspection contains index information and live-test information. A positive “URL is on Google” verdict does not guarantee that the page appears in search results. Keep the evidence type and what it can demonstrate clear in your audit. Google’s inspection documentation and Google’s guidance on inspecting a page describe these views and their limits.

2. Use a repeatable evidence record

Capture enough context that another auditor can trace the image and understand how it was produced. The exact query, time, location, device, and capture source are essential context; engine and evidence ID make the record easier to find and connect to the issue.

Field	What to record
Evidence ID	A stable identifier linked to the audit issue, such as `SEO-TECH-042-E03`.
Capture timestamp	Date and time with timezone or UTC offset. If you store both local time and UTC, make sure they agree.
Search engine	The engine whose results page was captured.
Exact query	The full query as entered. Do not rely on a shortened filename for the full value.
Location / locale	Country, language, and, when relevant, city or region. Note unknown or uncontrolled location.
Device	Mobile, desktop, or another device context; include viewport dimensions if known.
Capture source / mode	For example, manual browser capture, a named API, or Google URL Inspection live test. Include relevant settings if needed for reproducibility.
What it shows	A brief factual description of the observed result or SERP feature.
Related issue	The audit finding or issue ID and, where relevant, affected page or query.
File	The stored original image filename or controlled storage location.

The sample below illustrates a record format; it does not describe an actual capture or test.

Field	Illustrative value
Evidence ID	`SEO-TECH-042-E03`
Capture timestamp	`2026-10-03T14:22:00-04:00`
Search engine	Google
Exact query	`example query`
Location / locale	`US / en-US / New York, NY`
Device	Mobile; viewport noted if known
Capture source	Manual browser capture
What it shows	`Product rich result absent in this capture`
Related issue	`SEO-TECH-042`
File	`2026-10-03T1822Z__google__example-query__en-US-mobile__browser__SEO-TECH-042-E03.png`

3. Choose a filename and sidecar format

A filename should be stable, readable, and useful when images are listed together. One practical house convention is:

YYYY-MM-DDThhmmZ__engine__query-slug__locale__device__source__evidence-id.png

For example:

2026-10-03T1822Z__google__example-query__en-US-mobile__browser__SEO-TECH-042-E03.png

This is an editorial convention, not a filename standard mandated by Google. The example’s local time is 14:22 at UTC−04:00, corresponding to 18:22 UTC.

Keep the exact query and any long or sensitive context in an adjacent audit table or sidecar record. A shortened query slug can collide with another query, omit meaningful punctuation, or disclose information in shared storage. Use a stable evidence ID to connect the original image, any annotated copy, the sidecar record, and the report finding.

For example, a sidecar record can use these fields:

{
  "evidence_id": "SEO-TECH-042-E03",
  "capture_timestamp": "2026-10-03T14:22:00-04:00",
  "search_engine": "Google",
  "exact_query": "example query",
  "location_locale": "US / en-US / New York, NY",
  "device": "Mobile",
  "viewport": null,
  "capture_source": "Manual browser capture",
  "capture_mode": "SERP",
  "observation": "Product rich result absent in this capture",
  "related_issue": "SEO-TECH-042",
  "file": "2026-10-03T1822Z__google__example-query__en-US-mobile__browser__SEO-TECH-042-E03.png"
}

Adapt fields to the audit. Do not fill unknown values with guesses: mark them unknown or leave them blank, and distinguish an uncontrolled capture condition from a controlled one.

4. Preserve the original and link it to the finding

Save an unedited master image and retain its original filename or evidence ID.
If callouts, crops, or redactions are useful, create a separate annotated copy. Record the relationship to the master and describe any changes that affect interpretation.
In the report, state what the image demonstrates, the affected query or page, its evidence ID, and the recommended action.
Keep the exact context record alongside the image or in the audit’s evidence table, with access appropriate to any sensitive query or client data.

Describe only what is visible in the capture. A SERP image records one view at one moment; pair it with appropriate search performance or ranking data when the audit question is about performance. Do not describe the image alone as proof of rank or organic outcomes.

5. Make repeated captures comparable

For before-and-after or cross-market review, keep these capture axes consistent where possible:

Exact query and search engine
Location and locale
Device and viewport
Capture source and mode
Timestamp conventions and any repeat schedule
Storage, export, and evidence-ID practice

If a setting differs, record it and call out the difference in the report. Do not present captures with different query, location, device, or method as like-for-like evidence. For recurring or cross-market workflows, check whether a capture service actually provides the location routing, scheduling, overlays, or report exports your audit needs. For example, ScreenshotCenter describes support for captures across 80+ countries, scheduled captures, timestamp and URL overlays, and PDF reports; these are vendor-described features, not independent test results. A manual capture may be sufficient for a single example.

6. Capture a reproducible SERP image

Choose and record the exact query, engine, location or locale, device, and viewport as relevant.
Capture the results page and record the actual timestamp and time zone.
Save the original image and assign an evidence ID tied to the relevant issue.
Complete the sidecar or audit-table record, including the capture source and what is visible.
Review the filename and metadata for consistency before adding the image to the report.

For a browser-based capture, make sure the screenshot includes the relevant part of the results page and that the device and viewport are recorded. Browser automation can standardize viewport and capture steps, but it does not automatically make query, location, personalization, or result conditions identical; record the context the method controls and note what it does not.

7. Automate screenshot capture when the audit needs it

ScreenshotNeo is a website screenshot API and MCP server from Yorker Media. It can capture a page from one GET request and offers options including device and viewport controls, custom headers and cookies, wait conditions, CSS selectors, and image or PDF output. For repeatable page captures in an audit workflow, use its options to record the capture configuration alongside your evidence metadata. A website screenshot API captures a URL; it does not by itself reproduce a search query or guarantee a particular SERP, location, or personalized result. Preserve the query and search context separately, and confirm the chosen capture method is appropriate for the page and access conditions.

ScreenshotNeo documents its API at https://screenshotneo.com/docs/. The examples below use a target URL you control. Store your API key securely; do not commit it to source control or expose it in public client-side code.

cURL

curl -G "https://api.screenshotneo.com/v1/shot" \
  -d access_key=YOUR_API_KEY \
  --data-urlencode url=https://example.com \
  -o shot.webp

Python

import requests

r = requests.get(
    "https://api.screenshotneo.com/v1/shot",
    params={"access_key": "YOUR_API_KEY", "url": "https://example.com"},
    timeout=90,
)
r.raise_for_status()
with open("shot.webp", "wb") as image_file:
    image_file.write(r.content)

Node.js

const q = new URLSearchParams({
  access_key: process.env.SCREENSHOTNEO_API_KEY,
  url: 'https://example.com',
});
const res = await fetch(`https://api.screenshotneo.com/v1/shot?${q}`);
if (!res.ok) throw new Error(`Screenshot request failed: ${res.status}`);
const image = Buffer.from(await res.arrayBuffer());
await import('node:fs/promises').then(fs => fs.writeFile('shot.webp', image));

Request cURL from Python or Node

The cURL command above is the direct command-line example. Use the Python or Node.js examples when capture is part of an audit script or service, and preserve the requested URL and capture settings in the same evidence record as the output file. For SERP evidence, verify that the method and target actually represent the search results page and context required by the audit.

Options relevant to evidence capture

ScreenshotNeo supports full-page capture with lazy images loaded; CSS-selector element capture; dark mode; device presets and custom viewport; retina scale; PDF output and page settings; custom CSS and JavaScript; clicking an element; hiding selectors; waits for a selector, delay, or network idle; blocking ads, trackers, requests, or resource types; custom headers, cookies, user agent, authorization, timezone, and geolocation; transparent backgrounds; resizing; caching with a chosen TTL; signed links for public image tags; async jobs with signed webhooks; bulk capture up to 100 URLs per call; a usage API; and an OpenAPI spec. Check the API documentation for parameter names and accepted values before adding options to a request. ScreenshotNeo also accepts parameter names used by other screenshot APIs to make switching easier.

For evidence work, record settings that can change the captured output, especially viewport or device, geolocation, cookies, user agent, wait behavior, blocked resources, custom CSS or JavaScript, and cache use. If you redact or alter the output with custom CSS or JavaScript, disclose that in the capture record and keep an unmodified original where possible.

Or skip the browser setup

ScreenshotNeo can capture a URL with one API call. Cookie banners, popups, and chat widgets are removed before the shot. Bot checks, blank pages, and failed loads are never billed; response headers indicate the page verdict and billing status. An MCP server lets AI agents use screenshot tools. The free plan includes 1,000 screenshots a month with no card; paid plans start at $5 for 3,000. Use the API only where a URL capture suits the evidence requirement, and keep the exact SERP query and context in your audit record.

curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://example.com -o shot.webp

See the ScreenshotNeo API docs for request options. Sign up free for 1,000 screenshots a month, with no card required.

8. Troubleshooting labels and evidence

Problem	Likely cause	Fix
A reviewer cannot tell which query the image shows.	The filename has only a shortened query slug, or the exact query was never recorded.	Add the exact query to the sidecar or audit table and link it by evidence ID.
The timestamp seems inconsistent.	Local time and UTC were mixed or the timezone was omitted.	Store an explicit offset or timezone; check any UTC filename against the recorded local time.
Two captures appear different, but the cause is unclear.	Location, locale, device, viewport, or capture method changed without being recorded.	Compare the context fields, record differences, and qualify the comparison.
An annotated image is questioned as an original.	The edited copy replaced the master or its changes were not disclosed.	Retain the unedited master, identify the annotated derivative, and note crops, callouts, or redactions.
A URL Inspection image is being cited as a SERP.	The rendered-page screenshot was mislabeled as search results.	Label it as a URL Inspection live-test screenshot and state the inspected URL and live-test state.
The report claims a ranking based only on a screenshot.	A point-in-time visual observation is being treated as performance data.	Limit the statement to what is visible and use appropriate search performance or ranking data for performance claims.
A capture API returns an unexpected page or no useful image.	The target, access conditions, waits, or capture configuration may not fit the page.	Check the requested URL and API response, review wait and access settings, and record any custom headers, cookies, scripts, or blocked resources.

9. Performance, reliability, and cost considerations

Manual capture is often enough for one-off evidence. Repeated audits benefit from a stable naming convention, a reusable metadata form, and scripted or scheduled capture where consistent settings and volume matter. Automation can reduce repetitive steps, but it does not remove the need to preserve the exact query, locale, device, timestamp, and source.

For reliable evidence, retain the original and its metadata together, verify that files open, and keep evidence IDs unique within the audit. If an automated capture supports cache controls, async jobs, bulk requests, or signed webhooks, choose them based on freshness, volume, and delivery needs; document settings that could affect whether an image is new or reproducible. Do not infer a service’s latency, success rate, or cost from feature descriptions.

ScreenshotNeo plans include Free with 1,000 shots per month and no card, Starter at $5 for 3,000, Growth at $15 for 15,000, Pro at $39 for 60,000, Scale at $99 for 250,000, and Business at $249 for 1,000,000; yearly billing gives two months free, and every feature is on every plan. Its billing model charges only for clean shots: bot checks or CAPTCHAs, blank pages, timeouts, failed loads, and cache hits cost nothing, with page verdict and billing status reported in response headers. Choose based on expected capture volume and required options; a screenshot service’s page capture remains distinct from evidence of a specific SERP unless the actual capture method supports that context.

10. Quick audit checklist

Original image retained unchanged.
Unique evidence ID links image, record, and finding.
Timestamp includes timezone or UTC offset.
Search engine and exact query recorded.
Location or locale, device, and viewport recorded where known.
Capture source and relevant mode or settings recorded.
Observation is factual and limited to what the image shows.
Annotated derivative is separate and linked to its master.
Comparisons disclose changes in capture conditions.
SERP and URL Inspection evidence are labeled distinctly.

FAQ

Is there an official Google filename format for SERP screenshots?

No filename format is specified in the guidance summarized here. Use a consistent house convention and retain the full context in metadata.

Should I put the whole query in the filename?

Not necessarily. Use a readable query slug in the filename and preserve the exact query in the adjacent record, especially if it is long or sensitive.

Can a URL Inspection screenshot prove a page appeared in search results?

No. It shows a page rendered in a successful live test, not a SERP or a user’s search results.

Do repeated screenshots prove a ranking change?

They show what was visible in each capture context. A screenshot alone is not a ranking or organic performance measurement.