Share of voice, share of mind.

In 1975, a market researcher named James Peckham noticed something consistent across hundreds of consumer product categories: brands that spent proportionally more on advertising than their current market share tended to grow. Brands that spent less tended to shrink.

He called it the “Share of Voice / Share of Market” relationship. And fifty years on, it remains one of the most empirically robust frameworks in marketing. It is also one of the most consistently ignored when budgets are being set.

The core idea is simple. Your share of voice (SOV) is your brand’s advertising spend as a proportion of total category spend. Your share of market (SOM) is your brand’s revenue as a proportion of total category revenue. When SOV exceeds SOM, brands grow. When SOV falls below SOM, brands shrink.

The SOV/SOM relationship is not a coincidence. It follows from the mechanics of how brands acquire and retain customers.

Byron Sharp’s work on mental availability explains the transmission mechanism. Category buyers, at any given moment, are in different stages of consideration. Some are ready to buy now. Most are not. They are living their lives, half-forming opinions about brands based on whatever advertising they encounter. The brand that shows up most frequently and consistently builds the strongest memory structures — the instinctive associations that fire when a purchase decision eventually arrives.

When a brand’s share of voice matches its share of market, it is spending just enough to maintain its current position. It is holding the ground it has. When SOV exceeds SOM, it is building into the future — accumulating memory structures that will convert to purchases over the months and years ahead. When SOV falls below SOM, it is drawing down the reservoir.

“The long-term market share of a brand is determined by the proportion of category attention it commands, not by the cleverness of its creative or the precision of its targeting.”

— LES BINET & PETER FIELD, EFFECTIVENESS IN CONTEXT

Les Binet and Peter Field quantified the SOV/SOM relationship using the IPA Databank — the largest database of marketing effectiveness evidence in existence. Their analysis found a consistent, measurable relationship between Excess Share of Voice (ESOV) and market share change.

For every 10 percentage points of ESOV a brand sustains, it can expect approximately 0.5 percentage points of market share growth per year. The relationship is not perfectly linear, but it is remarkably consistent across categories and geographies.

The implication is that market share is not random. It is, to a significant degree, predictable from historical SOV. Brands that consistently hold positive ESOV compound their market position over time. Brands that consistently run below parity — spending less than their current market share would imply — reliably erode.

The IPA’s database shows the relationship visually as well as statistically. Plot share of voice against share of market for any mature category and you will find the same pattern: brands above the SOV = SOM diagonal tend to be growing brands. Brands below it tend to be in structural decline.

The brands sitting exactly on the diagonal are in equilibrium: holding their position, not growing, not declining. Many brands are happy here — category leaders in mature markets often optimise for this position deliberately. The point is to know where you are on this chart, and to choose your position intentionally rather than by accident.

The most uncomfortable implication of the SOV/SOM framework is what it suggests about brands that “go dark” — reduce or eliminate advertising during downturns, pivots, or cost-cutting exercises.

Going dark is appealing on a short-term P&L basis. It cuts costs immediately. It rarely cuts sales in the same period — the brand is living off the memory structures it has already built. But those structures decay. And once they are gone, rebuilding them is far more expensive than maintaining them.

The IPA’s research on brands that went dark during the 2008–09 financial crisis and the COVID-19 period reached the same conclusion. The brands that maintained their SOV when competitors cut back emerged with disproportionately stronger market positions. Not because they were braver — because the mechanics of mental availability work the same way in downturns as in growth periods.

The practical application of the SOV/SOM framework is simpler than it sounds. You need two numbers: your current market share, and your target market share. The gap between them, expressed as excess SOV, gives you the budget direction.

Binet & Field’s research suggests a general heuristic: to grow market share by 1 percentage point per year, most brands need approximately 10 percentage points of positive ESOV. For a brand with 15% SOM wanting to reach 18%, that means running at 25%+ SOV until the target is hit — and then stepping back to parity to hold it.

Category leaders can hold SOV roughly equal to SOM. They are defending, not attacking. Challenger brands — those with SOM below their ambition — need to hold positive ESOV until the gap closes, then step back to parity. Small or niche brands, paradoxically, need the highest proportional ESOV because the category threshold effect is working against them: below a certain absolute level of advertising, brands effectively become invisible to most category buyers.

If the SOV/SOM relationship is this empirically consistent, why do most companies set budgets without it?

Because the framework requires knowing your competitors’ spend. And because it implies that the right budget is not a percentage of revenue or a fixed historical number — it is a function of what the category is doing around you.

When a competitor doubles their media spend, your effective SOV halves — even if you changed nothing. That is deeply uncomfortable for a finance function that treats marketing budgets as a cost line rather than a market position investment. The SOV/SOM framework forces a conversation that most companies are not structured to have.

The brands that do have that conversation tend to be the ones still growing a decade later.