Comparing date ranges in the Search Console performance report

Comparing two periods is how you turn a raw number into a decision. "Clicks fell" is not actionable; "clicks fell 18 percent against the same weeks last year, and only on the product pages" points at a fix. Search Console performance report has a built-in comparison mode that puts two values side by side and adds a difference column, so you do not have to export and diff the numbers yourself.

Open the compare view

In Performance, Search results, open the date filter at the top and switch to the Compare tab. You can compare a current range against the previous period of equal length, or against the same period last year, and you can also compare two values in a grouping dimension (for example two countries or desktop versus mobile).

When a comparison is active the chart overlays both periods, and the data table gains a Difference column showing the absolute and percentage change for each row. The Search Console API data window is about 16 months, so keep one end of that in mind when you choose a same-period-last-year range.

Pick the right granularity

The default daily view compares each day with the same weekday in the other period, which empties the comparison of meaning when the days are different (a Monday in one period versus a Sunday in the other). Google recommends switching to weekly or monthly granularity for comparisons, because aggregating by week or month cancels the day-of-week effect and gives a cleaner read of long-term trend.

• Use same period last year for seasonality-aware checks, so a busy month is compared with itself a year earlier rather than with an unrelated month.
• Use previous period (last 28 days versus the 28 days before) when you want a fresh baseline that excludes last year's campaigns.
• Use weekly or monthly to ignore weekday noise; keep daily view only when you are tracking a specific anomaly day.

Read the difference column

The difference column tells you which pages or queries moved. Sort by it to surface the rows with the biggest change, then judge the move:

1. Look at the direction: is the change real, or is it a small percentage on a tiny base?
2. Check whether the movement is broad (a site-wide shift suggests an algorithmic or technical cause) or narrow (one section suggests a page-level or content change).
3. Recall that impressions and clicks measure different things, and clicking through the performance report to the spikes guide shows you how to separate a real change from noise.

A common mistake is comparing a current range against "previous period" for a seasonal business. For a retailer, comparing December to November compares the busiest month with a quiet one, which makes the difference column misleading. Use the same period last year for seasonal checks and reserve previous-period for fast, seasonality-light reads. The period-first mental model from the performance report and the spike diagnostics article complete the loop once the difference column flags a row worth investigating.