You've got the slide in front of you already. One number says 23% Share of Voice, the room goes quiet for a second, and then someone asks the only question that matters: 23% of what, exactly? That's where teams realise the headline figure is less important than the method behind it, because the method decides whether the number is useful or just neat-looking.
In the UK, that question gets sharper fast. Total UK advertising expenditure reached £42.6 billion in 2024, with online display at £16.7 billion and search at £16.9 billion WARC, so any single-channel headline share figure is already incomplete. A competitor can lead search, another can own broadcast, and a third can dominate regional coverage without that showing up in one blended percentage.
For founders, marketers and comms leads, the practical move is to treat share of voice analysis like a newsroom review, not a dashboard flourish. If you're benchmarking a sector and need a place to start, find early-stage advertising investors in the UK can help you think about how visibility plays differently across audiences and funding stages.
Why a Single Share of Voice Number Is Misleading UK Teams
A brand can lead Google results, lose broadcast coverage and barely register in trade press during the same reporting period. A single share of voice figure compresses those different realities into a number that looks precise but may answer no useful question.
Before accepting the result, check three things: which channels were counted, which competitors were included, and whether the period contained a launch, crisis or unusually quiet news cycle. A number built on unclear boundaries is not necessarily wrong. It is difficult to defend.
Practical rule: change the market definition and you change the share of voice result.
The UK media mix makes this problem harder. Online formats accounted for approximately four in every five pounds spent in the 2024 market, while television remained significant at £5.27 billion and out-of-home reached £1.39 billion. Those figures describe spend, not attention or trust. They also cannot show whether a brand is visible in paid search while losing earned coverage, or appearing frequently without influencing the conversation.
Carlos Alba Media approaches the measurement through channel-by-channel visibility, covering broadcast, online, print and social coverage rather than presenting one blended score. That structure reflects how UK media behaves and helps keep comparisons consistent across channels. It still measures visibility, not automatic attention, credibility or representation in AI-generated answers. Those outcomes need separate checks.
For the wider framework, see the agency's explanation of owned, earned and paid media. Teams benchmarking a sector can also find early-stage advertising investors in the UK and consider how visibility differs between audiences and funding stages.
Defining Share of Voice Across Media, Search and Social
Share of voice is not one metric. It's a family of measures that answer different questions, and the right one depends on whether you're checking campaign performance, competitor movement or brand health.
Media share of voice
Media share of voice measures how much of the category's coverage your brand earns across press, broadcast, online news and trade titles. In practice, that means comparing your share of relevant mentions, impressions or advertising value with the competitor set you've fixed for the analysis.
A Scottish tourism brand might care most about regional and national coverage in travel press, local broadcast and destination features. A regulated business will care more about where its message appears, whether the outlet is authoritative, and whether the story is framed as a product announcement, a compliance issue or a reputation moment.
Search share of voice
Search share of voice is about visibility in organic and paid results for the queries that matter. It's not just ranking position, it's whether your brand appears when a buyer searches for the category, the problem or the solution.
For a hospitality operator, this could mean visibility around destination, booking and comparison terms. For a Scottish tech SME, it could mean branded and non-branded query presence against rivals who are spending more heavily or publishing more aggressively.
Social share of voice
Social share of voice measures how much of the brand-relevant conversation happens around you on platforms where people talk, share and react. The useful version weights that conversation by audience and relevance, rather than treating every mention as equal.
For a tourism campaign, a single creator post with strong reach may matter more than dozens of low-engagement mentions. For a regulated sector, the signal may be smaller in volume but more important in terms of trust, scrutiny and message control.
The point is simple. Share of voice analysis becomes useful only when the definition matches the decision. If the business question is “who is winning the press fight?”, media SOV is the right frame. If the question is “who appears in the buying journey?”, search and social need to sit alongside it.
Calculating Share of Voice the Way Editors Actually Read It
Raw share of voice is the cleanest formula in the whole discipline. Take your brand mentions, divide by total category mentions, then multiply by 100. If a category has 200 relevant mentions and your brand appears in 50, your raw SOV is 25%. That's easy enough for a junior analyst to calculate, which is exactly why it gets overused.
Why weighting changes the answer
A newsroom would never treat every mention as equal, and neither should you. A national broadcaster's coverage can carry more reach and more credibility than a stream of low-authority duplicates, and the balance matters in a fragmented UK market. Ofcom evidence cited in Parliament shows 96% of UK adults consume news in some form, with BBC output reaching 73% of adults, ITV 42% and Sky 24% UK Parliament evidence.
That's why a weighted model is better for board reporting. Give higher weight to nationally influential outlets and lower weight to small aggregators, duplicated wire copy or weakly relevant posts. Keep the weight rules fixed from period to period, or the comparison becomes cosmetic.
A small amount of highly visible coverage can outrank a much larger pile of weak mentions.
Message share of voice
The number that often matters most is message SOV. That's the percentage of category coverage carrying your priority message, compared with the percentage carrying a competitor's message. If your launch story keeps appearing, but the key proof point never shows up, the media coverage is not doing commercial work.
A practical scoring sheet only needs a few fields, outlet, platform, prominence, audience estimate, key message and sentiment. From there, you can see whether you're winning on raw volume, on quality-weighted visibility, or on the actual message that should move enquiries.
Building a Reproducible Measurement Pipeline
A repeatable pipeline starts before the first scrape or clip lands in a spreadsheet. Fix the observation window, choose three to five genuine competitors, and hold both steady long enough to spot movement that means something. For most UK teams, 30 or 90 days is enough to show direction without letting one big news cycle distort the picture.
Then build the file the way a news desk would. Pull comparable mentions across broadcast, print, online and social. Remove syndicated repeats, obvious keyword junk and near-duplicate versions of the same story. If you skip that step, a wire pickup can inflate visibility without earning any extra attention or trust.
The workflow is simple, but the discipline sits in the consistency.
- Fix the window. Use the same date logic every cycle so month-on-month reporting stays comparable.
- Lock the competitor set. Swapping rivals mid-quarter changes the denominator and muddies the trend.
- Collect the sources. Bring in broadcast, print, online and social, and label source type clearly from the start.
- Deduplicate properly. One syndicated story should not appear as several separate wins.
- Classify the coverage. Record outlet, prominence, topic, sentiment and whether the key message appears.
Classification is where automated reporting often falls short. Visibility is only one layer. You still need a human read on whether the mention is central or passing, whether the brand is trusted or challenged, and whether it appears in the AI-surface summaries and search features that shape recall without delivering a clean click trail.
Carlos Alba Media fits this pipeline best in the judgement-heavy stages, especially the kind of PR effectiveness measurement that depends on source weighting and clean deduplication. That matters when a Scottish regional title should be logged differently from a national wire repeat, or when a trade piece needs context rather than blind sentiment tagging.
For the collection side, web data collection methods gives a useful primer on gathering and structuring the raw inputs.
Designing a Share of Voice Dashboard That Survives a Board Meeting
Most dashboards fail because they try to prove everything and explain nothing. A board-ready view needs only four charts: share by channel over time, weighted reach by competitor, message penetration versus competitor messages, and quarter-over-quarter movement with anomaly flags. Anything else should earn its place.
Keep the narrative, drop the clutter
A sentiment pie chart without context is usually decoration. So is total mention count by itself. Those views can tell you that something happened, but not whether it mattered, and not whether it changed the commercial picture.
What works better is a short narrative at the top of the dashboard, followed by the evidence underneath. If the headline says brand A lost share in broadcast but gained in search, the chart should show that shift plainly, with the competitor movement beside it.
For teams managing campaigns across multiple paid accounts, it also helps to keep the measurement view separate from ad operations. If you need a tool comparison before choosing infrastructure, find the right ad account management tool is a sensible starting point for the operational side.
If the dashboard can't survive a five-minute challenge from a sceptical CFO, it isn't a measurement tool, it's a decoration.
Tie the view back to outcomes
The best dashboard links visibility to business movement. That can mean enquiries, web traffic, reputation moments or inclusion in AI-generated answers. It should also hold the same competitor set and taxonomy for at least one quarter, so you can tell whether movement is real or just a measurement artefact.
For Carlos Alba Media, that same discipline appears in performance benchmarking, where the point is not to collect more charts but to make the comparison defensible. That's the mindset that survives a board meeting, and the one that keeps share of voice from turning into wallpaper.
Where Share of Voice Analysis Stops Working in 2026
The conventional model assumes people see a mention, click through, and then form an opinion. That no longer fits how a growing share of UK audiences encounter brands. Visibility still matters, but attention, trust and answer-surface representation matter more than raw mention count on their own.
Ofcom's 2025 research found that 51% of UK adults use social media for news, TikTok's news reach rose to 11% from 1% in 2020, Facebook reached 30% of adults for news, and TV and radio were trusted by 68% and 67% of UK adults respectively, compared with 44% for social media Ofcom. That creates a blunt measurement problem, a brand can win the volume game on social and still lose the trust game where reputation is formed.
AI search makes the gap bigger. Ofcom's online research indicates that around 30% of UK searches displayed AI Overviews, 53% of UK adults said they often saw such summaries, and ChatGPT recorded 1.8 billion UK visits in the first eight months of 2025, up from 368 million in the same period of 2024 Advanced Television. In that environment, being cited, described accurately and positioned prominently can matter even when no click follows.
For a tourism or hospitality brand, the practical risk is obvious. You can lead on raw coverage, yet still be underrepresented in the answers people read before they book. The measure now has to include inclusion in AI answers, the accuracy of the description, competitor presence and source citations, or you're measuring only part of the discovery path.
Your 30-Day Share of Voice Plan and Common Pitfalls to Avoid
Start with the objective in week one. Lock the competitor set, the reporting window and the channel mix, then decide whether the core question is campaign impact, brand health or competitor monitoring. Week two is for sources and rules, especially deduplication and the classification fields that keep the analysis stable.
Week three is where the numbers get built. Calculate raw share of voice, then calculate the weighted version so leadership can see both reach and quality. Week four is for the board-facing view, with the four core charts, a short written narrative and a clear call on what changed and why.
The pitfalls are predictable, and they're fixable:
- Changing the competitor set mid-quarter. Keep rivals fixed until the next reporting cycle.
- Mixing paid, owned and earned media without labels. Separate them or the headline becomes meaningless.
- Counting syndicated duplicates multiple times. Deduplicate before you calculate.
- Treating automated sentiment as final truth in Scottish or sector-specific language. Use human review where nuance matters.
- Chasing one headline percentage. Directional improvement is more useful than a vanity figure.
Share of voice works when it behaves like journalism. It has to be reproducible, sceptical and precise about what the numbers mean.
Carlos Alba Media uses newsroom discipline to make share of voice analysis useful for UK brands that need more than a monitoring screenshot. If you want a measurement approach that separates visibility, attention, trust and AI-surface representation, visit Carlos Alba Media and see how the team applies that thinking to PR, search and digital performance.