Local SEO
How to Compare Local SEO Performance Across Two Periods Fairly
A report that says “clicks fell 18%” may be mathematically correct but still misleading if one period includes a festival, a closure, or more days tha
By Chamko Research Team Updated 2 min read
Quick answer
A report that says “clicks fell 18%” may be mathematically correct but still misleading if one period includes a festival, a closure, or more days tha
A report that says “clicks fell 18%” may be mathematically correct but still misleading if one period includes a festival, a closure, or more days than the other. A fair comparison starts with period design.
Step 1: Choose comparable windows
For a quick operational review, compare equal-length complete periods. Twenty-eight days against the previous 28 days avoids unequal weekday counts better than comparing an incomplete month with a full month. For a seasonal business, also compare with the corresponding period from the previous year when reliable historical data exists.
No single comparison is perfect. A prior-year comparison may include changes in competition, pricing, tracking, or the website. A previous-period comparison may be more sensitive to seasonality.
Step 2: Preserve metric definitions
Use the same source and same measure in both periods. Do not compare Business Profile views in one month with website sessions in the next. Do not silently change the URL group, country filter, device filter, or report attribution setting.
| Question | Better comparison |
|---|---|
| Did Search visibility change? | Search Console impressions and clicks for the same property/page group |
| Did profile actions change? | The same Business Profile action metric and equivalent date windows |
| Did website visits change? | The same analytics session definition and landing-page segment |
| Did qualified leads change? | The same CRM qualification rule and source period |
Step 3: Calculate change transparently
Percentage change = (current period − comparison period) ÷ comparison period × 100.
Hypothetical example: 500 clicks in Period A and 425 in Period B means a change of (425 − 500) ÷ 500 × 100 = −15%. State the underlying counts as well as the percentage. A change from 4 to 2 is −50%, but the absolute difference is only two events.
If the comparison-period value is zero, the percentage-change formula is undefined. Report the counts and explain that a percentage comparison cannot be calculated in the usual way.
Step 4: Annotate the timeline
Record known changes near the chart: opening hours, temporary closure, website migration, campaign launch, price change, new branch, analytics tag change, public holiday, or major service disruption. These are possible explanations to investigate, not automatic causes.
Step 5: Check whether the change is broad or concentrated
Split Search Console data by page, query, device, and country where relevant. For profile reporting, examine each location separately before relying on a group total. One branch can mask another branch's decline.
Reporting template
- Period: [dates] vs [dates]
- Metric/source: [exact metric and platform]
- Counts: [current] vs [comparison]
- Change: [absolute difference] and [percentage, if defined]
- Known context: [closures, campaigns, tracking changes]
- Interpretation: [what the data supports]
- Limitations: [missing data, seasonality, small counts]
Sources
Next step: Standardise your reporting windows and annotate unusual business events before presenting month-on-month conclusions.