SERP clustering

Group keywords by actual Google top 10 overlap

How it works

Unlike semantic clusters (which compare the *meaning* of keywords), SERP clustering compares what Google actually ranks. For each keyword you provide (2 to 30), PivotSEO fetches the Google organic top 10, then counts, for every pair of keywords, how many URLs their two top 10s share. Two keywords that surface the same pages share the same intent in Google's eyes — the most reliable proof that a single piece of content can (or cannot) target both.

Decision thresholds

7-10 shared URLs → same page: Google treats both queries as equivalent. One content should target both (the highest-volume keyword becomes primary, the others become variants).
4-6 → same cluster: neighboring but distinct topics. Create separate pages, heavily interlinked (pillar ↔ spokes).
2-3 → adjacent clusters: different topics with an overlap zone. PivotSEO recommends cross-links between the two clusters.
0-1 → separate topics: no editorial link required.

Reading the results

Each cluster is presented as a pillar / spokes architecture (topic cluster): the pillar page is the keyword group with the highest cumulative volume, the spokes are the secondary pages of the same cluster. The linking rule is shown under each cluster: every spoke links to the pillar and vice versa. The full overlap matrix is expandable at the bottom (dark green = same page, light green = same cluster, yellow = adjacent). The save button turns a suggested cluster into a permanent PivotSEO cluster (visible in the Clusters tab).

Tip :Controlled costs: each keyword costs ~1 DataForSEO SERP request, but the top 10 is cached for 7 days. Re-running an analysis with the same keywords (or the SXO analysis on one of them) consumes nothing.

Warning :Results depend on the project's location (e.g. French Canada). The same keyword pair may overlap in Quebec but not in France.