The New Share of Voice: Why Custom Themes Matter

Andrew Wyatt

Chief Product Officer

  • Executive Reporting
  • Insights

The new share of voice is not a bigger clip count. It is share of voice inside the themes your strategy actually cares about. Delve (delve.news) lets communications teams define custom themes, tag coverage against them, and read competitive share of voice for that slice in the same dashboard.

Overall SoV still matters as a category barometer. Theme-scoped SoV answers the question heads of comms hear every quarter: are we winning the conversations we set out to own?

For the classic definition and calculation, see what is share of voice in PR. For presence versus prominence inside articles, see share of voice vs share of mentions. This page is about the next cut: SoV by theme.

What you need to know — TL;DR

Overall SoV is a relative volume measure

What does classic share of voice actually tell you?

How much of the category conversation mentions you, relative to a competitive set. It does not tell you whether you are winning on AI safety, pricing fairness, or whichever pillars sit in your OKRs.

The new SoV is thematic

What changes when you scope SoV to a theme?

The denominator. Instead of all category coverage, it becomes coverage matching a strategic theme. A brand can look average overall and still lead, or lag, in the one theme leadership funded the team to move.

Custom themes make it operational

How do you run this without a spreadsheet?

Define the theme once with a title and a description of what should and should not qualify, let matching coverage be tagged continuously, then filter the dashboard to that theme. Share of voice recalculates against the filtered set.

Themes are not key messages

What is the difference?

Themes are categories of coverage: AI identity, customer stories, regulatory risk. Key messages are the specific claims or framing you want coverage to carry. Use themes to slice the landscape, message pull-through to see whether your framing appeared inside it. See what narrative tracking is.

Methodology still decides whether the number is honest

What makes theme SoV misleading?

It inherits every SoV methodology choice: competitive set, qualifying coverage, and the publication universe behind the denominator. A vague theme definition produces a confident percentage about nothing.


Why classic share of voice stopped being enough

Share of voice earned its place in PR reporting for good reasons. Executives understand proportions. Competitive sets make the number feel fair. Trajectory over quarters is easy to chart.

The failure mode is just as familiar. A team reports healthy overall SoV while the coverage that matters strategically belongs to someone else. You can win the month on volume because commodity product news spiked, and still lose the quarter on the theme your CEO asked about in the last board meeting.

That gap is why teams often end up segmenting SoV by topic manually. Export clips, code themes in a sheet, rebuild the competitive math, and hope the coding stays consistent next month. The method is right. The operations are slow, brittle, and hard to defend when leadership asks for the same cut next week.

The new share of voice is that method productized: define the themes once, keep tagging continuous, read SoV where strategy lives.


What share of voice by theme actually measures

Theme-scoped share of voice asks one question:

Of the qualifying coverage in this competitive set that is about this theme, what share is ours?

It is still a relative volume metric. It is not quality, not sentiment, not message pull-through, and not business ROI. What changes is the denominator.

Use classic SoV for the wide aperture. Use theme SoV for the strategic apertures. Report both when the board needs context, rather than one number pretending to answer both questions.

One limit worth stating plainly, because it is the same one that governs every SoV number: this is a measure of presence in published coverage. It sits at the output stage of AMEC's evaluation framework. It tells you where the conversation is, not what anyone took from it. How to measure PR ROI in 2026 covers where that boundary falls.


Custom themes: the operating layer

Delve analyzes tracked coverage against themes. Teams start with a built-in catalog of 500+ predefined themes covering common corporate and industry categories, and custom themes extend it with definitions your team writes. The predefined catalog stays in place; custom themes are additive.

  1. Create the theme in project settings with a title and a description of up to 500 characters
  2. Say what qualifies and what does not. The description is where precision comes from, and the exclusions do as much work as the inclusions. "Coverage about model safety commitments, red-teaming, or frontier-safety evaluations. Exclude general AI hype or product-launch stories that never discuss safety" is a usable definition. "AI safety" is not. Keep the theme about the type of conversation, not about whether your brand is named. Brand presence is a different filter, and folding it into the theme definition turns a map of the conversation into a map of your own coverage
  3. Apply themes across the project so brand, competitor, and industry trackers share one taxonomy
  4. Filter to the theme on the dashboard and read competitive share of voice for that set. Filters update dashboard metrics in real time, share of voice included

That is the claim in plain language: custom theme, filter, theme-scoped SoV in one place. It is not a promise that volume equals reputation, and it does not replace reading the coverage.

Themes versus key messages

Turn every key message into a theme and theme SoV collapses into vanity tagging. Keep themes as the map of the conversation. Keep key messages as the evidence you showed up correctly inside it. See coverage quality benchmarks and AVE is dead.


How to run the new SoV in practice

Start from strategy, not from the theme catalog. Pick three to seven themes that map to current OKRs or board priorities. If a theme does not change a decision when it moves, it does not belong in the executive view.

Define competitors the way you define themes. Theme SoV is only as meaningful as the competitive set behind it. Keep the set stable across periods, because changing competitors midstream breaks the trend line. Same rule as classic SoV.

Consider a curated publication list as the barometer. Long-tail outlets can inflate denominators and dilute the strategic signal. Restricting the SoV view to a curated publication list makes the number reflect the outlets leadership cares about. Widen the aperture for discovery monitoring, tighten it for the board metric.

Read theme SoV next to quality and message metrics. A rising theme SoV built on off-message or low-relevance coverage is not a win. Pair it with message pull-through inside the same filter, sentiment trajectory for the same slice, coverage quality and readership where available, and share of mentions when presence is high but prominence is not. For board structure, see board PR reporting decision metrics and the board PR report template.

Treat theme definitions as living, versioned assets. When strategy shifts, update the definition deliberately and note the change in the report methodology. A silent rewrite makes a coding change look like performance.


What it looks like in a board report

Say a head of comms is responsible for three themes over the next six months: category leadership, trust and safety, and customer stories. Overall share of voice against competitors is roughly flat. Broken out by theme, the picture changes: category leadership is flat, trust and safety is down, customer stories is up.

The conversation with leadership stops being "we need more coverage" and becomes "we're losing ground on trust and safety, so that's where the effort goes." That is what share of voice by theme is for.

It doesn't prove revenue on its own. It shows where you stand in the coverage, and connecting that to results your organization already tracks is a separate step.


What Delve does, and what it does not

Delve does

  • Analyze tracked coverage against built-in and custom themes, defined by title and a description of what should and should not qualify
  • Recalculate dashboard metrics, share of voice included, against an active theme filter
  • Keep themes separate from key message analysis, so category and narrative stay distinguishable
  • Support the wider measurement stack: share of voice, share of mentions, coverage quality, message pull-through

Delve does not

  • Treat theme SoV as proof of business ROI
  • Replace judgment about which themes matter
  • Automatically rematch every historical article the moment you edit a theme definition; you select how far back you want to analyze
  • Equate themes with automatic story-thread clustering or frame detection

For platform fit against monitoring suites, see best PR measurement tools in 2026.


Frequently Asked Questions

What is the new share of voice?

Share of voice measured inside strategic themes rather than only across all category coverage. It answers whether you are winning the conversations your strategy named, not only whether you are generally visible.

How do custom themes change SoV?

Custom themes tag coverage against definitions your team writes, saying what should and should not qualify. Filter the dashboard to a theme and share of voice recalculates on that set. Same formula, strategic scope.

Is theme SoV better than overall SoV?

Neither replaces the other. Overall SoV is the wide barometer. Theme SoV is the OKR barometer. Reporting only one invites the wrong executive conclusion.

Are custom themes the same as key messages?

No. Themes categorize what coverage is about. Key messages are the specific claims or framing you want coverage to carry. Themes slice the landscape; messages show whether your framing appeared inside it.

Do I need Delve to run theme-scoped SoV?

No. Teams have done thematic SoV in spreadsheets for years. Delve keeps theme tagging and SoV in the same workflow, so the cut is continuous rather than a monthly rebuild.

What makes theme SoV misleading?

Unstable competitive sets, vague theme definitions, and unfiltered publication universes. A confident percentage over a badly defined theme is worse than no percentage. Fix the methodology before celebrating the number.