Share of voice benchmarking · category-level

Know what good looks like in your category.

A share-of-voice number on its own means very little. Benchmarked against the leader, the median vendor and your own history, it becomes a target — and a board-ready story about whether you are gaining ground.

Category benchmark Sample · illustrative only
Your share against the leader, the median and your own last quarter.
How benchmarking works

Four reference points.

All computed from your own category, not a generic index.

01
The leader
Whoever actually holds the most category answers, which is often not the biggest company.
02
The median
The middle vendor in your category, which is the realistic near-term target for most challengers.
03
Your history
Your own share on the same prompt set over time, which is the only strictly fair comparison.
04
Per cluster
Benchmarks split by prompt intent, because a blended number hides where you are actually losing.
What you get

Six views that make the number usable.

A single percentage is a headline, not a plan.

Category distribution

How share is spread across every vendor named, not just the ones you listed.

Gap to leader

The distance to first place, per engine and per cluster, tracked over time.

Percentile placement

Where you sit in your category's distribution, which is more honest than a raw figure.

Cluster benchmarks

Comparison, pricing, integration and security prompts scored separately.

Historic baselines

Your own past quarters on the identical prompt set, so progress is not confounded by prompt changes.

Board-ready output

One page: where you are, where the leader is, and what moved. PDF or CSV, with answers linked.

Reading the numbers

What a share-of-voice figure typically means.

Ranges from our client cohort, not a universal standard.

Blended share
What it usually indicates
Sensible next move
Under 10%
Absent from most category answers
Fix retrievability and publish core facts
10–20%
Named occasionally, rarely first
Comparison content and segment positioning
20–35%
A recognised option, not the default
Attack the leader's strongest cluster
35–50%
Frequently the recommendation
Defend: monitor displacement weekly
Over 50%
Category default in most answers
Protect framing and stale-claim risk
Questions

About benchmarking.

No credible cross-industry standard exists yet, and any single number would be misleading. We benchmark inside your own category, which is the only fair comparison.

By actual share of your category’s answers — sometimes a smaller vendor with better published facts, not the market leader.

Quarterly for the board view, weekly for the operating view. Both come from the same runs.

It would, so historic baselines are always recomputed on the identical set and we flag any prompt change.

Yes — separate prompt sets per segment or market, each with its own distribution.

Benchmark your category.

Free first reading, with the leader and median named.