Share of Voice in 2026: Why the Metric My Client Called BS Still Wins

About six years ago, I had a client who refused to believe in share of voice. He called it a vanity metric. He said it was the kind of thing agencies invented to justify bigger budgets. He wanted clicks, conversions, ROAS, and nothing else on the dashboard.

I spent the better part of a year proving him wrong. We ran a full marketing mix model. We pulled comparative spend data from his category. We built scenarios showing what would happen if we held him at his current weight versus pushed him slightly above his share of market in paid presence. We tested. We measured. We adjusted.

By the end of that engagement, his campaign efficiency had improved by roughly 300%. I will not pretend the entire lift came from share of voice. It did not. We rebuilt creative, fixed audience strategy, killed wasted spend in three channels, and rewrote his landing pages. But the strategic posture, deciding to spend at a level above his current share of market in the category, was the structural shift that made the rest of the work pay off.

Without that, we would have been optimizing inside a budget too small to matter.

I still think about that client. Not because he was difficult, though he was. Because his skepticism was reasonable. Share of voice is one of the most misunderstood concepts in marketing, and most of the people who fight against it are doing so because they have only seen it used badly. Let me try to fix that.

What share of voice actually is

Share of voice is the percentage of total category communication your brand controls in a given period. If your category collectively spends 10 million dollars on advertising in a quarter and your brand spends 800 thousand of that, your share of voice is 8%. That is the simple version. The useful version is comparing it to share of market, which is your share of category sales over the same period.

The relationship between those two numbers is where the real planning happens. Brands that consistently spend a larger share of category voice than their share of category sales tend to grow. Brands that spend less tend to shrink. The gap has a name: excess share of voice, or ESOV.

This is not a hunch. Les Binet and Peter Field analyzed effectiveness data from the Institute of Practitioners in Advertising databank and found that an ESOV of 10 percentage points produces about 0.5% to 0.7% of annual market share growth, depending on category. Nielsen's analysis of 123 brands landed on roughly the same number. The B2B Institute confirmed the rule applies in B2B markets, where 10 points of ESOV drove around 0.6% annual growth. Four decades of research, multiple databanks, hundreds of brands, and the relationship keeps holding. The original articulation goes back further still, to John Philip Jones in a 1990 Harvard Business Review piece called "Ad Spending: Maintaining Market Share."

If you take only one thing from this entire post, take this: under-investment is a near-guaranteed path to decline. The studies consistently show that brands with negative ESOV lost market share in roughly 80% of cases analyzed. Quiet brands shrink. They do not hold steady. They shrink.

Why this matters more in 2026, not less

The standard counterargument is that the rule was developed in a TV-and-print world and digital fragmentation has broken it. The opposite is true. Fragmentation made share of voice harder to measure, but it made the underlying mechanism, mental availability, more important.

Mental availability is the Ehrenberg-Bass concept that explains why share of voice works at all. When a category buyer enters a buying moment, they consider a small set of brands that come to mind quickly. The brands that come to mind quickly are the ones that have built the most associations through consistent presence over time. Share of voice is the input. Mental availability is the asset it builds. Sales is what happens when the asset gets activated.

Three things have changed the playing field in 2026, and none of them weaken the argument.

First, attention is more fragmented than it has ever been, which means the cost of reaching a buyer once is higher and the cost of reaching them ten times is significantly higher. Brands that are not actively defending share of category presence are losing ground to whoever is.

Second, generative AI flooded every channel with content. My teenagers, looking at a professionally produced campaign for one of my clients, asked me if it was AI-generated. The actual ad. From a real brand. With a real production budget. That is the world we are in. When everything looks fake, the brands that show up consistently and credibly are the ones that get believed.

Third, and this is the new piece almost nobody is tracking yet, a meaningful share of buyer research now happens inside AI assistants. ChannelEngine's 2026 marketplace report found 58% of consumers have used AI tools to research products. Adobe Analytics measured AI-driven referral traffic to US retail sites up 4,700% year over year. When a buyer asks ChatGPT or Perplexity or Claude for the best vendor in your category, the brands that appear in that response are the new shortlist. Not appearing means not existing.

How to build share of voice across channels

The mistake I see most often, across the hundred-plus engagements I have worked on, is treating share of voice as a paid media question. It is not. It is a presence question, and presence is built across every place a category buyer might encounter your brand.

Paid media is still the foundation. If you are running brand-building campaigns at a level proportionally above your share of category sales, you are buying ESOV. The Binet and Field rule of thumb is that the optimal split between brand-building and activation is roughly 60/40 in favor of brand. Most of the brands I work with are running closer to 80/20 in favor of activation, because activation is easier to measure and easier to defend in a budget meeting. That short-termism is exactly what creates the share of voice gap their competitors are exploiting.

Search has become the most accurate proxy for traditional share of voice in most categories. Mark Ritson and others have argued that share of search, the percentage of branded category searches that include your brand name, tracks closely with share of market and predicts share movement. It is also free to measure. If you are not pulling Google Trends data for your brand against your top three competitors quarterly, you are leaving the cheapest competitive intelligence in marketing on the table.

Earned media and PR carry disproportionate weight in 2026 because they feed two systems at once. They build human-credibility share of voice, real journalists writing about real brands, and they feed the AI citation graph. Recent analysis suggests earned media generates around 89% of AI-cited links. A single Forbes feature is worth more than ten owned blog posts in terms of how AI systems weight your authority in the category. This is also why I am so skeptical of brands that have outsourced PR entirely to automated pitch tools. The pitches do not land, the relationships do not get built, and the citation footprint stays small.

Social presence still matters, but the math has flipped. Polished branded content is now the suspicious version. Customer-generated content, employee voices, and nano-influencer partnerships are doing the work that paid celebrity endorsements used to do. I had a healthcare client last year where a single recommendation from a small account with about twelve thousand followers drove more qualified consults than a six-figure media buy. That is not an outlier. That is the trust economy at work, and it shows up in share of voice as a multiplier on every dollar spent.

Owned content, your blog, your podcast, your newsletter, is where you compound. Recent research suggests it takes roughly 250 substantial documents to meaningfully shift how a large language model perceives a brand in a category. That is a lot of writing. Most of my clients give up at 30. The ones who get to 100 start showing up in AI responses. The ones who get to 250 start dominating them. There is no shortcut, and there is no AI tool that produces 250 substantial documents in a weekend that will actually move the needle. Thin content does not count. The threshold is substantial, expert-attributed, original.

AI visibility is the new frontier and the one most brands are completely unprepared for. Spotlight's analysis of 2.4 million AI responses found that brand share of voice varies wildly by platform. You might be at 40% share in ChatGPT and 12% in Perplexity for the same category prompt. Each platform pulls from different sources, weights authority differently, and rewards different content structures. Tracking AI share of voice across at least ChatGPT, Perplexity, and Google AI Overviews should be on every CMO's quarterly review by now. It is on almost none of them.

What I tell skeptical clients in 2026

I think about that early client every time a new one tells me they only care about ROAS. I tell them what I told him. ROAS is a measurement of how efficiently you are harvesting demand that already exists. Share of voice is what determines how much demand exists for you in the first place. You can optimize a small pond into oblivion, or you can make the pond bigger.

The 300% efficiency lift we got eventually was not a clever campaign. It was a structural decision to invest enough to be heard, and then to make sure what we said was worth hearing. Six years later, with attention more fragmented, content more synthetic, and buyer research increasingly mediated by AI, that decision matters more, not less.

Share of voice is not a vanity metric. It is the bet you make on whether your brand will exist in three years. The brands that figure that out in 2026 will own their categories in 2029. The ones that do not will spend the next decade wondering where their growth went.

Wondering where your brand sits on this chart?

Most of the brands I work with have never measured their share of voice against their actual share of market. They are flying blind, optimizing the wrong metrics, and watching slow erosion they cannot explain.

If you want to know where you sit, I offer a free 30-minute diagnostic call. We will look at your category, your spend, and your competitive position, and you will leave with a clear sense of whether share of voice is the lever you should be pulling, or whether the constraint is somewhere else entirely.

Linda Orr, PhD is the founder of Orr Consulting and a fractional CMO with 100+ client engagements across DTC, healthcare, telehealth, and B2B services. She is rated in the top 1% on Upwork with an 86% client retention rate and a 100% five-star review record.

Book a free diagnostic call →