History and comparisons: how it works

In short

In almost every section, a bar at the top of the page lets you pick a period (7 days, 28 days, 3 months, 12 months or exact dates) and compare it with another: the previous period, the same period last year, or dates of your choice. But you can only compare with the past PivotSEO has already recorded. The longer your project stays active in PivotSEO, the more history you have, and the richer the comparisons become.

How the history builds up

• At every analysis, PivotSEO keeps a dated copy of your data: every week during the Sunday scan, and whenever certain pages are viewed.
• These copies pile up, week after week. Nothing is deleted: that is what lets you go back and compare.
• History starts the day your project is created — or the day a new feature appears in PivotSEO. Before that, there is nothing to compare.

What you can compare, depending on your project's age

• After a week: two weekly snapshots. You see what moved from one week to the next.
• After a month: one month against the previous one.
• After three months: a quarter against the previous one — enough to tell a real trend from a fluke.
• After a year: the same period last year. The fairest comparison, because it cancels out the season — a florist in February, a roofer in November.

The exceptions: what already has a past

A few data sets come with their own history, because Google keeps it on its side. They can be compared from day one:

• Google Search Console: the last 16 months are retrieved at once, query by query.
• Keywords' monthly volume: 12 months, published by Google.
• The Google Business Profile: views, calls and directions, as Google kept them.
• For everything else — citations in AI answers, mentions, sources, rankings, backlinks, external presence… — history builds up over time, from the day you sign up.

When a comparison is empty

• If the compared period falls before the history starts, the page says so: there is simply no snapshot at those dates yet.
• The comparison then reads « no data », with the exact dates (« no data between 1 and 28 August »), not « not available »: « not available » is kept for a value the source didn't measure, whereas « no data » says no measurement was taken at those dates.
• Today's reading describes the state at the close of the period: the situation on the last day, not an average.
• Search Console, GA4 and Bing: when the archive is missing days in either period, the comparison uses the average per archived day. A gap in the archive no longer reads as a traffic drop.
• PivotSEO never makes up a number to fill a gap. A « no snapshot for this period » isn't a failure: it's the history that hasn't caught up with the chosen period yet.
• Pick a more recent period — or come back in a few weeks.

Up-to-date comparisons, complete months

• Comparisons update as soon as the data changes: a Refresh, a new data pull or a calculation redone when a page opens shows up right away in the period comparison, with no more waiting of up to 12 hours.
• Saving the project settings (sector, market, Google listing, competitors, brand terms) also refreshes the comparisons that depend on them.
• Keyword history: Google's published monthly volume and the snapshot taken on the 1st of the month are now both kept; one no longer erases the other.
• Search history (Search Console): a month never rebuilt (May 2026, for example) is no longer mistaken for an ordinary snapshot that fell on the last day of the month. In the Archived snapshots view of the Search Console page, the Fill in missing months button rebuilds it from Google's free APIs.

Analysis history: reopen or remove

• Stored analyses (heatmap, audits, Core Web Vitals, brand perception, AI analysis of the period, brief versions…) have an Analysis history list: click a row to reopen it.
• Remove an analysis: click the row's ×, then confirm. It stays stored, but is no longer shown or compared.
• If you remove the analysis being displayed, the most recent remaining one takes its place.