Discussion
  • Search Console
  • SEO Analytics

Are you using Search Console custom annotations as a change log, and has it actually made diagnosis faster?

Published 1 reply

Discussion starter: an SEO measurement question, posted with an answer from Niraj Raut to open the thread. If you have dealt with this on a site, reply with what you saw, especially where it differs.

Disclosure: Niraj Raut, who posted this answer, runs the SEO consultancy linked at the end of it.

1 reply

  • admin
    Admin
    Original poster

    Yes, but as a pointer on the chart, not as the change log itself. Custom annotations speed up one specific step of diagnosis: answering “did anything change around this date?” without digging through Slack, tickets and memory. On sites where several teams or an agency touch production, that step can eat much of the first hour. They do nothing for the harder steps, which are working out where the change happened and proving what caused it.

    No public study measures whether annotations shorten diagnosis, so any claim either way is anecdotal. You can measure it for your own team, and the constraints of the feature make a strong case for keeping the real log somewhere else.

    What the feature does, and where its limits bite

    Google launched custom annotations in the Performance reports in November 2025. You right-click the chart, choose a date and add a note. Google’s suggested uses include infrastructure changes, SEO work, content strategy shifts, and external events such as holidays. The details that matter for a change log:

    Constraint Detail What it means in practice
    Length Up to 120 characters per note Room for a label and a ticket reference, not an explanation
    Volume Up to 200 annotations per property Fine for major events; a team deploying daily would run out
    Editing Notes cannot be edited, only deleted and re-added Typos and changed rollout dates mean deleting and redoing
    Retention Removed automatically after 500 days Roughly matches the 16 months of Performance data, but your history disappears with it
    Visibility Everyone with access to the property sees them Good for shared context; keep sensitive details and names out
    Scope Stored per property, shown on Performance charts A Domain property and its URL-prefix properties need a decision on where notes live

    The length, cap, edit and retention details come from launch coverage. Do not assume annotations travel with API data or exports; the launch coverage describes them only on the Performance report charts.

    Where diagnosis time actually goes

    A change log is to traffic diagnosis what a medication list is to a doctor. It diagnoses nothing, but a doctor who has to phone three relatives to find out what the patient took last week loses the first hour before examining anyone.

    Detect: clicks or impressions moved
    ↓
    What changed around that date? (annotations help here)
    ↓
    Where: segment by page type, query group, country, device, search type
    ↓
    Why: form hypotheses from the segment pattern
    ↓
    Verify: logs, URL Inspection, rendered HTML, rollback or test

    Annotations shorten the second step and help with handovers: a new team member, an agency, or an executive looking at the chart sees the context without asking. They also help you avoid chasing phantom problems. When Google stopped supporting the &num=100 results parameter in September 2025, many sites saw desktop impressions fall and average position improve in Search Console, while clicks were affected far less, according to Brodie Clark’s analysis. A dated note on that change saves every future analyst from diagnosing a reporting artefact as a ranking event. The same goes for the March 2026 and May 2026 core updates (March, May), and for knowing that AI Mode data has been counted in Performance totals since June 2025.

    For a solo operator who ships every change personally, the gain is smaller, because the change log already lives in one head. The value grows with the number of people who can change production.

    Four ways annotations can mislead you

    • A note next to a dip invites a causal story. An annotation shows timing, not cause. Confirm with segmentation: a template change should hit that template first, not every page equally.
    • Time zones shift the date. Search Console dates are in Pacific Time. A Tuesday-morning release in Sydney or Perth falls on Monday in Search Console. Annotate the Pacific date, or the note will sit a day after the effect it explains.
    • Effects lag the change. On-page and template changes need recrawling before they show up, so the movement can appear days or weeks after the note. A robots.txt block or an accidental noindex can act faster. Hourly data from the Search Analytics API can narrow the onset for fast-moving problems like those.
    • An incomplete log produces false negatives. “Nothing was annotated, so it wasn’t us” is an expensive mistake. A patchy log is worse than none, because it creates confidence.

    A two-layer setup that holds up

    Keep the system of record in a shared sheet, database or ticket system, ideally fed automatically from deployment tooling. Use Search Console annotations as a short, consistent pointer to the significant entries.

    Fields for the master log

    • Date and time with time zone, plus the Pacific date
    • Change type: template, content, internal links, redirects, robots or sitemap, rendering, CDN or firewall, tracking, business event, external
    • Scope: URL pattern, template, country or subfolder
    • Expected effect, metric and likely lag
    • Owner, ticket or pull request link, rollback date
    • Whether a Search Console annotation was added

    A consistent note format inside 120 characters

    For example: DEPLOY | /products/* canonical to parent | JIRA-4812 or EXTERNAL | Google core update started | see log #214. A fixed prefix makes the chart scannable, and the ticket reference points to the full context.

    What earns a Search Console note

    Changes with plausible search impact at template or section level, migrations and redirect batches, robots.txt and sitemap changes, CDN or firewall rule changes that could affect Googlebot, Google updates and reporting changes, outages, and major business events. In Australia that includes end of financial year sales, Click Frenzy and Black Friday. Single-article edits belong only in the master log.

    Team situation Role of Search Console annotations Primary change log
    Small site, one person ships changes Can be the log, within the 200-note cap Optional
    Agency working with client developers High value as shared context both sides see Shared sheet, with logging as a release checklist item on the client side
    Enterprise with frequent releases Major events only Deployment log joined to performance data, for example through the BigQuery bulk export

    How to tell whether it is saving time

    Treat it as a process change you can measure. For the next 90 days, record every traffic anomaly worth investigating, and for each one note the time from detection to the first credible hypothesis and whether the eventual cause was already in the log.

    • Metric: log hit rate, the share of anomalies whose confirmed cause was logged before anyone started investigating.
    • Target: 80% or more of internally caused anomalies already in the log (a working threshold, not a benchmark). If the hit rate is well below that, the problem is capture discipline, not tooling.
    • Secondary metric: median time to first credible hypothesis, compared with a rough baseline reconstructed from the previous six months of incidents.
    • Next step: if the hit rate is high but time has not improved, the bottleneck is segmentation, so invest in prebuilt segment views (for example in Looker Studio), query groups or a BigQuery export instead of more annotation.

    Because you choose the date when you add a note, you can also backfill major past events from the last 16 months. That gives the chart useful context from day one, and the exercise shows how many past incidents nobody can explain.

    The practical rule: make the master-log entry a release checklist item, and add a Search Console note only when the change clears the section-level bar above. The speed gain comes from that logging discipline, not from the chart. Annotations make the discipline visible to everyone who opens the report, which is worth the few seconds each note takes.

    Need help with this on your own site?

    Niraj Raut works with businesses in Nepal, Australia, the UK, Europe and the US on technical SEO, ecommerce SEO, local SEO and AI search.

    See SEO services

Add a reply

Replies are open to approved members. Log in or apply to join.