RRF = ?
Interactive Tool

WHAT IS
RECIPROCAL RANK

RRF is the algorithm that determines which sources AI systems cite. Play with the k constant, adjust rankings, and see why consistency across multiple retrieval systems beats single-system dominance.

RRF(d) = 1/(60 + rBM25) + 1/(60 + rVector) + 1/(60 + rCitation)

Drag the slider below to change k and watch scores recalculate

60
1 (Rank matters most)60 (Standard)200 (Consistency matters most)
DOCUMENTS
AYour Optimized Page
BCompetitor #1 (Backlinks Only)
CCompetitor #2 (SEO Only)
DCompetitor #3 (Social Signals)
FUSED RANKINGS

RRF SCORES

RANKPAGEBM25 SCOREVECTOR SCORECITATION SCORERRF TOTAL
1Your Optimized Page0.015380.015870.015630.04688
2Competitor #3 (Social Signals)0.014710.016130.014290.04512
3Competitor #1 (Backlinks Only)0.016390.013890.013330.04362
4Competitor #2 (SEO Only)0.016130.014710.012500.04333
THE RRF LESSON

Notice how the winner is not always the page that ranks #1 in any single system. With k=60, a page that ranks consistently across all three systems can outscore a page that dominates just one. This is why modern SEO requires optimizing for keyword search, semantic search, AND citation graphs simultaneously. Single-signal dominance is obsolete.

VISUAL COMPARISON

WHY CONSISTENCY WINS

1. Your Optimized Page0.0469
BM25
#5
VECTOR
#3
CITATION
#4
TOTAL
#1
2. Competitor #3 (Social Signals)0.0451
BM25
#8
VECTOR
#2
CITATION
#10
TOTAL
#2
3. Competitor #1 (Backlinks Only)0.0436
BM25
#1
VECTOR
#12
CITATION
#15
TOTAL
#3
4. Competitor #2 (SEO Only)0.0433
BM25
#2
VECTOR
#8
CITATION
#20
TOTAL
#4
FAQ

COMMON RRF QUESTIONS

What is Reciprocal Rank Fusion?

Reciprocal Rank Fusion (RRF) is a rank aggregation algorithm that combines results from multiple retrieval systems into a single ranked list. The formula is RRF(d) = sum(1/(k + r_i(d))) where k is a constant (typically 60) and r_i(d) is the rank of document d in each system. It rewards consistent top-10 performance across multiple systems over single-system dominance.

What does the k constant do in RRF?

The k constant (typically 60) in the RRF formula controls how much rank position matters. With k=60, the difference between rank 1 and rank 10 is very small, meaning a document that ranks consistently well across multiple systems scores higher than a document that dominates just one system.

Why does RRF matter for SEO?

AI search systems use RRF to combine results from multiple retrieval pipelines: keyword search, semantic search, citation graphs, and entity lookup. A page that ranks well across ALL systems achieves a higher RRF score than a page that dominates just one. This is why modern SEO requires optimizing for multiple signals simultaneously.