Your CEO asks a simple question: "Are we winning the conversation, or is our competitor?" You have mention counts and a sentiment chart. What you don't have is one number that compares you with the rest of your market.
That number is share of voice (SOV). This page works like a calculator: every variant of the formula, a worked example you can check line by line, a spreadsheet setup, and the inputs that quietly break the result.
The basic formula
If your category gets 1,000 mentions in a month and 250 are about you, your share of voice is 25%. Before that number means anything, fix three inputs:
- The competitor set. Three to five brands your buyers actually compare you with. Write the list down and don't change it mid-year.
- The period. A calendar month is the most common window. Every brand uses the same dates.
- The sources. Every brand is measured across the same channels, languages and markets.
The term "share of voice" covers three different measurements. Keep them apart, because they answer different questions.
| Type | What it counts | Question it answers |
|---|---|---|
| Earned media SOV | Articles in news sites, blogs and online media | How visible are we in the press compared with competitors? |
| Social SOV | Posts, comments and shares on social networks and messengers | How much are people talking about us compared with competitors? |
| Paid SOV | Ad spend or ad impressions in a channel | How much of the advertising in our category is ours? |
This calculator covers earned and social SOV, which both come from monitoring data. Paid SOV comes from ad platforms and works differently. If you are still deciding how news and social tracking fit together, read social listening vs media monitoring first.
Weighted variants
Raw SOV treats a post from an account with 40 followers the same as a front-page story. Weighted variants fix that. Each one swaps the input you count.
Reach-weighted SOV
This shows how many people could have seen each brand, not how often it was named. Reach estimates differ between tools, so compare reach-weighted SOV only within one tool and one method.
Sentiment-weighted SOV
Volume is not always good news. A product recall can double your share of voice in a week. Three formulas separate attention from approval:
Definitions of net sentiment SOV vary between agencies and tools. Pick one, write it in your report's footnote, and keep it.
Tier-1 SOV
First agree a short list of outlets that matter most: the national business press, the trade titles your buyers read, the channels your regulators follow. Tier-1 SOV is often the number leadership cares about most. It is also small, so one story can move it a lot.
Worked example: three brands
The numbers below are made up for illustration. They are not real brands or real data. Imagine three delivery apps in one market, measured over one calendar month across news and social. Here is where Brand A ends up on each variant.
Same brand, same month, four very different stories. Here is the math behind each one.
- Raw SOV. Brand A has 420 mentions, Brand B 980 and Brand C 600: 2,000 in total. Brand A: 420 ÷ 2,000 × 100 = 21%. Brand B: 980 ÷ 2,000 × 100 = 49%. Brand C: 30%. On volume, Brand B leads by a wide margin.
- Reach-weighted SOV. Estimated reach is 6.3M for A, 9.8M for B and 4.9M for C: 21.0M in total. Brand A: 6.3 ÷ 21.0 × 100 = 30%. Brand B: 46.7%. Brand C: 23.3%. Brand A's mentions sat in bigger outlets and larger accounts, so it gains 9 points. Brand C loses ground.
- Positive and negative SOV. Positive mentions: 168 (A), 196 (B), 180 (C), 544 in total. Negative mentions: 42, 294 and 60, 396 in total. Brand A: 168 ÷ 544 × 100 = 30.9% positive and 42 ÷ 396 × 100 = 10.6% negative.
- Net sentiment SOV. Subtract negative from positive. Brand A: 30.9 − 10.6 = +20.3. Brand B: 36.0 − 74.2 = −38.2. Brand C: 33.1 − 15.2 = +17.9.
- Tier-1 SOV. In the agreed tier-1 list there are 100 mentions: 38 for A, 40 for B, 22 for C. Brand A: 38 ÷ 100 × 100 = 38%. It is almost level with Brand B, with less than half of B's raw volume.
The full sentiment table makes the flip easy to see:
| Brand | Positive | Negative | Positive SOV | Negative SOV | Net sentiment SOV |
|---|---|---|---|---|---|
| Brand A | 168 | 42 | 30.9% | 10.6% | +20.3 |
| Brand B | 196 | 294 | 36.0% | 74.2% | −38.2 |
| Brand C | 180 | 60 | 33.1% | 15.2% | +17.9 |
| Total | 544 | 396 | 100% | 100% |
Illustration only: Brand A, B and C, one month. Neutral mentions are left out of this table.
In this example, Brand B's lead in volume came mostly from complaints, say a week of delivery delays. It owns almost three quarters of all negative conversation in the category. On raw SOV, Brand A looks like a distant third. The weighted views show it reaches more people per mention, carries the least negative share, and nearly matches the leader in tier-1 press.
Build it in a spreadsheet
You don't need special software to run the math. Export mention counts per brand from your monitoring tool, then set up one sheet.
- One row per brand. Put brand names in column A (rows 2–4) and raw mentions in column B. Add columns for reach, positive, negative and tier-1 mentions.
- Raw SOV. In the next free column, enter
=B2/SUM(B$2:B$4)and format it as a percentage. Copy it down. The$keeps the total fixed while the row changes. - Every other share. Repeat the same pattern for reach, positive, negative and tier-1 columns. Each share column should add up to 100%.
- Net sentiment. Subtract the negative SOV cell from the positive SOV cell for each brand. Show it as points, not as a percentage.
- Trend. Keep one tab per month with identical layout. A summary tab can then pull each brand's share across months into a single chart.
Inputs that break the result
Most SOV disputes are not about the formula. They are about what went into it.
Which number to report
Leadership wants to know three things: are we ahead, is it getting better, and why.
- Lead with one number and its trend. "Our share of voice was 30% in October, up from 24% in September." Then show the chart.
- Explain every big move. Name the two or three stories behind any shift of more than a few points, and link to them.
- Keep the method in a footnote. Competitor list, sources, languages, dates, and how you handled duplicates. When someone challenges the number, the footnote answers them.
- End with an action. "Brand B's delivery complaints are an opening for our reliability message" is more useful than a percentage.
On cadence: monthly is the standard rhythm. Go weekly for high-volume brands or during a launch, quarterly for niche brands and board reports, and daily only during a crisis. Our crisis monitoring playbook covers that workflow.
How Scano does the math for you
Scano is an AI media monitoring and social listening platform. It tracks news sites, social networks, messengers, blogs, forums and reviews across 1M+ sources in 20+ languages.
You set up each brand as a theme. The theme comparison and share of voice reports, two of Scano's 13 analytics reports, put your brand and competitors side by side over the same period, so you don't rebuild the spreadsheet every month.
It also covers several of the inputs above:
- Language coverage. Scano is strongest where English-first tools are thin: Türkiye, Central Asia, the Caucasus and the Gulf. It covers Telegram channels and VK as well as Instagram, TikTok and YouTube.
- Sentiment in more than English. Dedicated sentiment models for English, Turkish, Arabic, Russian, Kazakh, Persian, Hebrew and French support positive and negative SOV in those languages.
- Mentions without text. OCR reads brand names in images and video frames, and speech-to-text transcribes what people say in videos and stories. See image and video monitoring.
- Explaining the why. Smart tags classify mentions by topic. The AI Analyst answers questions such as "why did Brand B's negative mentions jump?" in plain language, with links to the source posts.
- Spotting sudden shifts. Spike signals alert you by email or Telegram when volume or negativity jumps.
Reports export to Excel, Word and PDF, and can be scheduled on your own template, so leadership gets the same format every month. Every plan includes unlimited users.
Conclusion
Share of voice is one of the simplest numbers in communications, and one of the most misused. The formula takes one line. The work is in the inputs: equal queries, clean data, full language coverage and matching time windows.
Report raw SOV with at least one weighted view, explain every big move, and keep your method in a footnote. Do that every month, and SOV becomes a number leadership trusts.
Book a Scano demo and we'll set up a trial workspace on your own brand and competitors, so you can see your share of voice across news, social and messengers in your markets.
FAQs
What is a good share of voice?
There is no universal benchmark. It depends on how many competitors you have and how big they are. A useful rule of thumb is to compare your SOV with your market share: if your share of voice is higher, you are getting more attention than your size suggests.
How do you calculate share of voice?
Divide your brand's mentions by the total mentions of your brand and all competitors in the same period, then multiply by 100. For example, 420 mentions out of 2,000 in total gives a share of voice of 21%.
What is the difference between share of voice and share of market?
Share of voice measures your part of the conversation or advertising in a category. Share of market measures your part of sales. Many teams track both, because a gap between them often shows where attention is not turning into revenue, or the other way around.
Should share of voice include negative mentions?
Raw SOV includes all mentions, positive and negative. That is why you should also report positive, negative or net sentiment SOV. Otherwise a crisis can look like a gain.
Which tools measure share of voice?
Most media monitoring and social listening platforms offer one. Compare them on coverage in your markets and languages, sentiment quality, and how easily you can export and schedule the report. Our guide to media monitoring covers what to check.