Share of voice: own more of the conversation
Whoever gets talked about most usually gets bought most — marketers have measured that link for decades. The conversation just moved: it now happens in search results, social feeds, and AI answers. Here's how to measure your share of it, and grow it.
What is share of voice?
Share of voice (SOV) is the percentage of your market's total visibility that belongs to your brand. The formula is simple: your brand's visibility ÷ total category visibility × 100. If your category generates 1,000 mentions this month and 180 name you, your share of voice is 18%.
Why anyone cares: share of voice runs ahead of share of market. The advertising researcher Les Binet's long-run finding — brands whose SOV exceeds their market share tend to grow, and the gap (eSOV) roughly predicts how fast. Originally that was an advertising rule. It has survived every channel shift since.
The channel shifts are the interesting part. "Voice" used to mean ad impressions. Then organic rankings and social mentions. The newest arena is AI answers — when ChatGPT lists three vendors for your category, being one of them is share of voice in its purest form: there are only three slots.
How to measure share of voice
Pick the arenas your buyers actually use, measure the same competitor set in each, and hold the method steady month to month. The common ones:
| Arena | What you count | Also called |
|---|---|---|
| Paid search & ads | Impression share vs. category spend | SOV (classic) |
| Organic search | Your visibility across category keywords vs. competitors' | Share of search / SEO share of voice |
| Social | Brand mentions vs. total category mentions | Social share of voice |
| AI answers | Prompts where an engine names or cites you vs. competitors | AI share of voice / share of model |
How to calculate share of voice
SOV (%) = (your brand's mentions ÷ total category mentions) × 100
Worked example: your category's tracked prompts and keywords produce 1,000 brand appearances this month across the arenas you measure. 180 of them are yours. 180 ÷ 1,000 × 100 = an 18% share of voice. Swap "mentions" for the unit of each arena — impression share for ads, visibility-weighted rankings for organic, prompt appearances for AI answers — and keep the competitor set identical everywhere.
One warning from experience: don't average the arenas into a single blended number. A strong social share can hide the fact that AI assistants never mention you — and those are different problems with different fixes.
AI share of voice: the newest arena, the fewest slots
A results page has ten spots and a feed scrolls forever, but an AI answer names two or three brands. Winner-take-most dynamics, measured one prompt at a time. That's AI share of voice (some teams call it share of model): across a fixed set of category prompts, the percentage where each engine names or cites your brand.
It behaves differently from the older arenas. It moves when the sources engines trust start mentioning you — coverage, comparisons, industry pages — not when you publish more yourself. Which means the lever is earned coverage, and the measurement has to trace citations back to the pages that produced them.
This is the exact loop our AI visibility tracking runs monthly — your share per engine, per prompt, with every citation traced to its source. Growing the number is the placement work behind our AI SEO service, with generative engine optimization doing the citation building and earned media supplying the coverage engines trust most.
How to increase share of voice
- 1
Baseline every arena against the same rivals
You can't grow a number you haven't measured. One competitor set, every arena, one month of honest data — the gaps will rank your priorities for you.
- 2
Concentrate, don't spread
Share is relative — small budgets win narrow arenas. Own every prompt and keyword in one niche before widening. A 40% share of something beats 2% of everything.
- 3
Feed the sources, not just your site
Rankings, social reach, and AI answers all lean on third-party validation. Coverage on the publications your market trusts raises your share in every arena at once — it's the only lever that pays four times.
- 4
Re-measure monthly, react quarterly
Monthly data catches competitor moves while they're cheap to answer. Quarterly is the right cadence for strategy changes — faster and you're chasing noise.
Frequently asked questions
What is a good share of voice?
Higher than your market share — that gap (excess share of voice) is what predicts growth. A challenger with 5% of the market and 15% of the voice is set up to gain; a leader with 40% of the market and 20% of the voice is set up to shrink. The absolute number matters less than which side of your market share it sits on.
What is the difference between share of voice and share of market?
Share of market counts money — your slice of category revenue. Share of voice counts attention — your slice of category visibility. Voice moves first: sustain visibility above your market share and revenue usually follows it up, which is why SOV works as an early indicator.
What is share of search?
A search-specific proxy for share of voice: your brand's search volume (or organic visibility) as a share of all brands in the category. It became popular because it correlates well with market share and is cheap to measure from public data. It misses the newest arena though — AI answers don't show up in search volume.
What is AI share of voice or share of model?
The percentage of category prompts where an AI engine names or cites your brand, measured across a fixed prompt set per engine. "Share of model" is the same idea named after the models themselves. It's the scarcest voice arena — an answer names only two or three brands — and the one growing fastest in buyer influence.
How often should we measure share of voice?
Monthly, against a fixed competitor set and method. Weekly data is mostly noise except during launches or crises; quarterly leaves you reacting to competitor moves a season late. The discipline that matters most is not the frequency — it's never changing the measurement method mid-stream.
What Our Clients Say
We reported social SOV to the board for years while ChatGPT recommended competitors in a category we invented. The per-arena breakdown ended that blind spot in one report.
Miriam L.
CMO, B2B software
Finally a team that treats AI answers as a voice arena with its own math instead of bolting a chatbot column onto a social listening deck.
Tom W.
Brand Strategist
The baseline was eye-opening and the monthly tracking is solid. I'd buy a self-serve dashboard tomorrow if they shipped one.
Aylin D.
Head of Marketing
From the blog
All articlesWhat's your share of AI answers?
The free baseline report measures your AI share of voice against the competitors you pick — per engine, per prompt, sources included.
Get my baseline — free