Colquhoun.AI
← All guides

Research Guides

Search Console Comparisons: A Before-and-After Worksheet

· Colquhoun.shop editorial

A page edit went live, and Google Search clicks increased. Before calling the edit a success, check whether the two reports cover the same audience, pages and length of time. A useful comparison starts with a reproducible view and ends with a specific next question.

This worksheet is for people reviewing their own Google Search Console data. It includes a fictional example and a short reporting template. It does not describe a Colquhoun.shop integration, report on a real business or promise a ranking improvement.

Define the question before opening the chart

Choose one question that the available evidence can answer. For example: “Did recorded Google Search clicks to these five guides change between two completed four-week periods?” That is narrower and more testable than “Did our SEO work?”

Write down the five page addresses before looking at the results. If you later add a newly published guide to the second period, label that as a different comparison. Otherwise, an apparent improvement might simply reflect a larger collection of pages.

Google's Performance report overview distinguishes clicks, impressions, click-through rate and average position. Pick the metric that fits the question. Clicks can help you examine search traffic, but this report alone does not establish sales, customer satisfaction or profit.

Copy this comparison record

Keep this record beside the chart or export. It is a working template, not a required Google format.

Write “none” for unused filters. An empty field is ambiguous: it could mean no restriction or a setting nobody recorded. Save separate records for separate questions rather than overwriting the original setup while exploring.

Choose periods you can explain

For a first review, two consecutive completed 28-day windows are often convenient because each contains four of every weekday. This is a practical starting point, not a statistical threshold. A seasonal business or infrequently searched page may need a different window and more context.

For example, September 7 through October 4, 2026, and October 5 through November 1, 2026, each contain 28 days, running Monday through Sunday. A promotion in one period still makes the periods different in an important way. Record it.

Avoid comparing a partial current month with a full previous month. A year-over-year comparison can add context for recurring seasons, but check moving holidays, campaigns and changes to the page set before treating it as a clean control.

Google's data-status guidance explains that recent figures can be preliminary and change. Use completed periods for this worksheet, and record the check date. If an incident requires immediate attention, examine recent data separately and label the uncertainty instead of postponing the incident response.

Keep the view consistent as you investigate

Set the date comparison and check the entire filter row before copying any values. Google documents multiple filters and date comparisons, including the fact that clicking a table row applies a filter. An exploratory click can therefore change what your next screenshot represents.

First record the overall result for your chosen scope. Then inspect a page or device segment and save it as a separate view with its own settings. Return to the baseline before opening another segment.

For a five-page review, compare the same five individual page rows across both periods. Do not substitute a site-wide property total for those page-level figures. Google's aggregation explanation describes different counting at property and page level. Record the filters and aggregation rather than assuming every chart represents the same scope.

Work through a fictional result

Suppose a report filtered to the same five guides, with all other settings unchanged, shows these totals for the two completed periods:

Clicks increased by 12.5%, calculated as 100 divided by 800. Impressions increased by 50%, calculated as 10,000 divided by 20,000. Both statements can coexist with a lower click-through rate because impressions increased faster than clicks.

Calculate each period's overall click-through rate from its total clicks divided by its total impressions. Do not take a simple average of daily percentages when the days have different impression counts. For help labeling rates and changes, see Read Traffic Changes in the Right Units.

The figures justify a limited conclusion: “This view recorded more clicks and impressions, with a lower click-through rate.” They do not show that a title change caused the increase, that every page improved or that the additional visitors became customers. This example contains no average-position data, so it supports no conclusion about a position change either.

Preserve missing information

Do not force a query table to equal the headline total. Google's data-discrepancy guidance explains that anonymized queries, table limits and aggregation can produce differences. Query filters can exclude anonymized data, so a filtered view is not a complete list of everyone who searched for you.

Record what the report displays before exporting. Google's export guidance warns that unavailable values shown as a tilde or dash become zeros in the download. Keep a separate note identifying those entries. A downloaded zero should not automatically become a claim that nobody clicked.

If a value is genuinely zero in the earlier period, report the counts rather than dividing by zero to create a percentage change. If it is unavailable, say unavailable. Those are different situations and lead to different next steps.

Choose one follow-up, then write the conclusion

Use the comparison to narrow the investigation. If one guide accounts for most of the click decline, inspect that guide and its recorded changes before rewriting the entire site. If clicks grew while the overall click-through rate fell, inspect page and query patterns before assuming the wording became less effective.

Google's traffic-drop investigation guide identifies several possible explanations, including technical problems, changing demand, search-system changes and reporting issues. A date match alone cannot tell you which explanation applies.

Finish with three short statements: what the same-scope data shows, what it cannot establish and what you will inspect next. For the fictional example: “Clicks rose from 800 to 900 across equal completed periods. We have not attributed the increase to the page edits. Next, we will compare the five page rows to locate where the additional clicks occurred.”

Keep the comparison record with that conclusion. A colleague should be able to reopen the same view, understand its limits and continue the investigation without guessing what the chart meant.

This guide describes the built-in Web Reader and Keyword Density behavior verified for this publication. Available data depends on the page response; these tools do not measure visits or search rankings.