--- type: Leaf title: Reading Search Console as a weekly record description: Pull clicks, impressions and position by page, indexing state for the URLs that matter, and crawl statistics each week into a file you keep. resource: https://nationaldigital.com.au/headless-cms/ai-visibility/search-console-weekly-record/ tags: - headless-cms - Measurement - Technical SEO - AI Search Visibility - Search Console - Google Search Console weekly report - Search Console page dimension - indexing report - crawl stats - Search Console API export timestamp: '2026-10-15T06:00:00.000Z' language: en-AU publisher: National Digital author: Cameron Young authorRole: CEO readingTimeMinutes: 6 wordCount: 1181 datePublished: '2026-10-15' dateModified: '2026-10-15' pillar: headless_cms canonical: https://nationaldigital.com.au/headless-cms/ai-visibility/search-console-weekly-record/ --- _Home — Content & Experience — AI visibility — Reading Search Console as a weekly record_ # Reading Search Console as a weekly record Search Console shows a rolling window and forgets. Pulled by page on the same day each week into a file you own, it becomes a series that shows what the site is being shown for and what is not indexed. **Quick answer:** Search Console pulled by page on a fixed weekday, with indexing verdicts for key URLs and crawl statistics, becomes a series the business owns; the query view is incomplete and most indexing reasons are engineering findings. ## About this page - Written by Cameron Young, CEO - 6 min read - Updated 15 October 2026 _Measurement — Technical SEO — AI Search Visibility_ ## What should a weekly read of Search Console actually record? Impressions, clicks and average position by page, the indexing state of every URL the business cares about, and the crawl statistics. Pull them on the same day each week into a file you keep, because the console shows a rolling window and forgets. **Additional Context** The page dimension is complete. The query dimension omits rare queries for privacy, so a series built by page is the one that reconciles. _High confidence · Verified 15 Oct 2026_ *The console as a series* ## Build the series by page The Performance report can be grouped by query or by page, and the two do not add up to the same total. Google omits rare queries from the query view to protect privacy, so a site with a long tail of specific questions sees a large share of its impressions attributed to no query at all. The page view is complete. Build the weekly series by page, then group pages by the service or the location they serve, and the totals reconcile from week to week. By page also answers the questions that matter in practice. Which pages are shown and never clicked. Which page lost impressions the week a template changed. Which new page has started to be shown at all. Each of those is a page with a URL, and a URL is something an engineer can open. ## Indexing state as a signal The page indexing report lists the reasons a URL is not in the index, and many of them are engineering findings with the diagnosis already written. Discovered and not yet crawled points at crawl budget or internal linking. Crawled and not indexed carries no reason from Google and usually needs a look at the page's content and duplication. Duplicate without a user-selected canonical points at a template that generates two URLs for one page. Excluded by a noindex tag points at a tag nobody remembers adding. Each one is recorded weekly against the URL, so a regression appears the week it happens. The same mechanisms decide whether a migration keeps its rankings, which [SEO preservation in a migration](/headless-cms/wordpress-migration-to-headless/seo-preservation/) covers. ### Analytics beside it Analytics sits beside the console in the weekly read, and the honest use of it is narrow. Engaged sessions by landing page show which pages hold a visitor. Key events are whatever the business configured, and a key event is not a lead until the inbox or the CRM says one arrived. The record keeps the two apart: what the analytics tool counted, and what the business actually received that week. Measured together over months, the gap between them is one of the more useful numbers the record produces, and it is a number the business owns. ## The engineering that follows A page shown and never clicked gets its rendering and its structured data checked; whether its title should change is a marketing decision the record informs. A page that dropped the week a template shipped gets the template diffed. A URL excluded by canonical gets the canonical generation fixed in the CMS. A sitemap that still lists URLs that redirect gets the sitemap job repaired. These are tickets with evidence attached, and the method for writing them is in [turning findings into engineering tickets](/headless-cms/ai-visibility/findings-to-engineering-tickets/). The weekly console pull is one of the sources in [AI Visibility Management](/services/ai-visibility-management/), kept alongside the [weekly crawl](/headless-cms/ai-visibility/crawl-health-and-core-web-vitals/) so that what Google reports can be read against what the site actually serves. ## How to keep Search Console as a weekly series The pull that turns a rolling dashboard into a record the business owns. - Total time: Half a day to set up, then minutes a week - Difficulty: intermediate - Audience: it_manager **Tools:** - Search Console property with a service account - The Search Console API - The CrUX API or a manual report export - A CSV repository 1. **Grant a service account read access** _(Half an hour)_: Add a service account to the Search Console property with read permission so the pull runs without a person's login. 2. **Pull the Performance data by page** _(Automated)_: Use the Search Console API to fetch clicks, impressions and position by page for the previous seven days, on the same weekday every week. 3. **Pull the indexing state for the URLs that matter** _(Automated)_: Inspect the service, location and product URLs the business cares about and record the verdict and the reason for each. 4. **Record crawl statistics and Core Web Vitals status** _(Manual export, or the CrUX API)_: The Crawl Stats and Core Web Vitals reports are not in the Search Console API, so export them from the console on the same day, or pull field data per URL group from the CrUX API, and keep both so a change is dated. 5. **Store every pull as a row in a file you own** _(Automated)_: Append to plain CSV in a repository. Never overwrite, because the series is the asset and the console will not give last year back. **Summary:** Give a service account read access, pull Performance by page and indexing verdicts for key URLs on a fixed weekday, keep crawl and Core Web Vitals status, and append every pull to a CSV series. *Key Takeaways* ## What a weekly Search Console series is for Search Console pulled by page on a fixed weekday, with indexing verdicts and crawl statistics, becomes a series the business owns. Most indexing reasons are engineering findings with the mechanism already named, and the record keeps analytics counts apart from real enquiries. - **Build the series by page, because the query view is incomplete** Rare queries are omitted for privacy, so only the page dimension reconciles week to week, and a page is something an engineer can open. _Critical_ - **Indexing reasons are findings with the diagnosis attached** Duplicate without a chosen canonical, crawled and not indexed, excluded by noindex: each names the mechanism and the URL. _Critical_ - **Keep analytics counts and real enquiries in separate columns** A key event is what a tool counted. A lead is what the inbox received. The gap between them over months is a number the business owns. _Important_ - **Pull on the same day and never overwrite** The console shows a rolling window. The question in six months is answered only by the rows kept today. _Important_ ## Questions about Search Console as a record ### Why do the query totals not match the page totals? Google omits rare queries from the query dimension to protect privacy, so part of the impressions in the page view appear under no query at all. The page view is complete, which is why the series is built from it. ### How far back does Search Console keep data? The Performance report holds a rolling window of months. Anything older is gone unless it was exported. A weekly pull into a file the business owns is the only way to answer a question about last year. ### Is a key event in Analytics a lead? No. A key event is an action the business chose to count, and the count can include tests, bots and people who did not go on to enquire. The record keeps the analytics count and the enquiries the business actually received as two separate columns. ### What access does a weekly pull need? Read access for a service account on the Search Console property and on the Analytics property. No write access, and no person's login. ## Want your Search Console history kept as a series from this week? - [Book a scoping call](https://calendly.com/cameron-at-national-digital/discovery) ## Related reading - [Recording what AI assistants say about your business](/headless-cms/ai-visibility/ai-assistant-answers/) - [The sources AI assistants cite when they describe you](/headless-cms/ai-visibility/sources-assistants-cite/) - [Testing whether an AI agent can use your website](/headless-cms/ai-visibility/assistant-task-tests/) - [Measuring your place in AI Overviews and search features](/headless-cms/ai-visibility/ai-overviews-and-serp-features/) - [Keeping Business Profile, Bing and your site in agreement](/headless-cms/ai-visibility/business-profile-and-bing-consistency/) - [Crawl health and Core Web Vitals as a standing measurement](/headless-cms/ai-visibility/crawl-health-and-core-web-vitals/) ## Related **Parent:** - [AI visibility](/okf/headless-cms/ai-visibility.md) **Related:** - [Recording what AI assistants say about your business](/okf/headless-cms/ai-visibility/ai-assistant-answers.md) - [The sources AI assistants cite when they describe you](/okf/headless-cms/ai-visibility/sources-assistants-cite.md) - [Testing whether an AI agent can use your website](/okf/headless-cms/ai-visibility/assistant-task-tests.md) - [Measuring your place in AI Overviews and search features](/okf/headless-cms/ai-visibility/ai-overviews-and-serp-features.md) - [Keeping Business Profile, Bing and your site in agreement](/okf/headless-cms/ai-visibility/business-profile-and-bing-consistency.md) - [Crawl health and Core Web Vitals as a standing measurement](/okf/headless-cms/ai-visibility/crawl-health-and-core-web-vitals.md) - [Turning visibility findings into engineering tickets](/okf/headless-cms/ai-visibility/findings-to-engineering-tickets.md) # Citations - [Google Search Central - Search Console overview](https://developers.google.com/search/docs/monitor-debug/search-console-start) — What Search Console reports about a site's presence in Google Search. - [Search Console Help - Performance report](https://support.google.com/webmasters/answer/7576553) — How clicks, impressions and position are counted, and the dimensions the report can be grouped by. - [Search Console Help - Core Web Vitals report](https://support.google.com/webmasters/answer/9128668) — How the report groups URLs by status using field data from the Chrome User Experience Report.