Year-over-year (YoY) comparison sets this period's numbers beside the same period last year. It is the simplest way to separate real growth from seasonality: a December dip in B2B traffic looks alarming against November but normal against last December. YoY is standard in SEO reporting, where traffic follows strong calendar patterns.
How to calculate it
The formula is a percentage change, but the period choice matters as much as the math.
- Align by weekday for short periods: compare Monday to Sunday with the matching week, not the same calendar dates
- Watch moving holidays and events, such as a holiday that fell in a different week
- Use full months or quarters for board reporting to avoid partial-period errors
YoY change = (This period - Same period last year) / Same period last yearWhen YoY breaks down
YoY assumes the measurement was the same in both years. Tracking migrations (for example, from Universal Analytics to GA4), new consent banners, changed key events and site migrations all break that assumption. Annotate those dates and add a caveat when a comparison crosses them.
Search Console keeps 16 months of performance data in the interface, enough for YoY on recent periods. For longer history, export the data regularly through the API or bulk export.
Pairing YoY with shorter views
YoY tells you about the trend; month-over-month tells you about momentum. Use both, and read sudden month-over-month changes against what happened in the same month last year before reacting.
Example
A summer slump that was not one
Imagine Larchpoint, an education SaaS, sees organic sessions fall in July versus June. YoY shows July up against last July, with the same seasonal dip in both years. The team reports growth and keeps the content plan, rather than launching an emergency audit.
Common mistakes
Comparing across a tracking change without saying so
The difference may be measurement. Flag it in the report.
Using calendar dates for weekly data
Different weekday mixes distort short comparisons, especially for B2B traffic that drops at weekends.
Questions
Use month-over-month trends and compare against similar launches or pages, and start building the history you will need next year.
