ESOV, explained for people who have to defend it
ESOV — excess share of voice — is your share of category advertising minus your share of category market. Positive, and you should expect to grow; negative, and you're renting decline. The exchange rate, from Binet and Field's analysis of the IPA Databank: roughly 10 points of ESOV buys about 0.5 points of market share growth per year, on average, with creative quality acting as a multiplier on the rate.
You might think it's not even that big of a concept. But this is what will be the dealbreaker in your next budget meeting: how to compute it for a niche B2B category where nobody publishes ad spend, what the multiplier means if you're outspent, and the three ways ESOV gets misused to justify bullshit.
WTF is ESOV?
Share of voice (SOV) is your slice of the category's advertising presence. Back in the original research that meant your share of category ad spend. Share of market (SOM) is your slice of category revenue. ESOV is the first minus the second.
The pattern Binet and Field documented across the IPA's effectiveness cases is close to law-like: brands whose voice runs ahead of their size tend to grow, and brands whose voice trails their size tend to shrink. Spend at your weight and you hold position. The market has an equilibrium, and advertising weight is one of the few levers that moves a brand off it.
Despite the ad land's proclivity for esoteric mysticism (us included), the mechanism isn't that mysterious. About 95% of your buyers are out-of-market at any moment (Dawes, Ehrenberg-Bass) — not comparing vendors, just slowly forming memories they'll use later. Your share of those memories tracks your share of voice, and future demand tracks the memories. Voice is the input you control today; share is the output that arrives on a delay. ESOV is the gauge between them. Simple, right?
How is ESOV calculated? A worked example
The formula is subtraction. The work is in the inputs. Here's a fictional B2B software niche, round numbers on purpose:
Read that last row the way it's written. Half a point per ten is an average across categories, years and campaigns. Category maturity moves it. Distribution moves it. Product moves it. Creative moves it most of all — which is the challenger's entire opening, and we'll get there. Plan with ESOV like weather, not physics: it tells you which way you're leaning, not exactly where you'll land.
The reverse reading matters just as much. If that same brand cut spend to $600k — SOV 6%, SOM 8%, ESOV −2 — the model expects share to erode. Slowly, invisibly, in next year's numbers, where nobody connects it to this year's "efficiency." Negative ESOV is buying decline at a discount and booking it as savings.
How to measure share of voice when nobody publishes ad spend
In FMCG you buy the data. In a B2B or DTC niche, no report exists — so you build a proxy index and track it consistently. Five proxies, cheapest first:
- Paid search coverage. Your impression share on the category's money terms (Google Ads reports it), plus a monthly manual check of who actually appears on the ten queries that matter.
- The LinkedIn Ad Library. Every competitor's live ads are public. Count them monthly — number of actives, formats, how long creatives run. Free competitive intelligence almost nobody collects.
- Event presence. Booths, speaking slots, sponsorships at the three events your buyers attend. A sponsorship prospectus is a price list for voice.
- Niche media share. Who sponsors the newsletters and podcasts your market reads and hears? Rate cards plus sponsor archives get you close.
- Sales-call anecdata. "We keep seeing them everywhere" from prospects is a share-of-voice reading with a delay. Log it.
Score each proxy quarterly, weight them however you like, and keep the method identical from quarter to quarter. The absolute number is soft; the direction is decision-grade. If your voice index trails your market share two quarters running, you're funding someone else's growth — that conclusion survives any reasonable weighting.
The multiplier
The 10:0.5 exchange rate assumes average work. Work isn't average by accident in either direction.
System1 and Peter Field priced the downside: dull advertising needs roughly 2.6x the media spend to produce the growth interesting advertising produces. And the upside compounds — genuinely interesting, feeling-something work drives 6.1x more share growth than dull-rational work. Same media invoice, wildly different yield per point of voice.
For a challenger this is the whole game, because you can't win the numerator. The leader can always buy more voice. What they usually can't do — through eleven approvals and a brand council — is make anyone feel anything. Famous work trades above the ESOV exchange rate; dull work trades below it. If your budget is the small one at the table, the multiplier isn't a nice-to-have. It's the only column you control outright.
And if ten points of national ESOV is fantasy money, shrink the denominator: define a market small enough that your voice is loud in it. That move has its own name and its own math — the Challenger CMO Playbook is built on it.
Don't misuse ESOV - it gets you in trouble
The spray. "We need awareness" with no named market is not an ESOV argument; it's a spend request wearing one. ESOV only means something after you've defined the denominator — which market, whose spend, measured how. No denominator, no strategy.
The wrong denominator. A niche player computing national SOV is measuring their voice in a room they don't sell to. Your category is the buyers you actually compete for — compute against that room, not the industry's. Choosing the denominator honestly is the strategic act; everything after is arithmetic.
The covenant read. Treating 0.5-per-10 as guaranteed output gets CMOs fired in eighteen months. It's an average with real variance, a direction with error bars. Present it that way — "the evidence says leaning above our weight buys growth; here's the range" — and it survives finance scrutiny. Present it as a promise and the first slow quarter takes the whole framework down with it, which is how good models get banned from boardrooms.
FAQ
What is ESOV? Excess share of voice: your share of category advertising minus your share of category market. Positive ESOV predicts share growth (~0.5 points per year per 10 points of ESOV, per Binet & Field's IPA analysis); negative ESOV predicts slow decline.
How is ESOV calculated? ESOV = SOV − SOM. Share of voice is your advertising spend (or a proxy index of presence) divided by the category total; share of market is your revenue divided by category revenue. In niches without published spend data, build a consistent proxy index — paid search coverage, ad libraries, event and media sponsorship share.
How much growth does ESOV buy? On average, about 0.5 market-share points per year for every 10 points of ESOV (Binet & Field, IPA Databank). Creative quality multiplies the rate in both directions: dull work needs ~2.6x the media to keep up, while genuinely interesting work grows share far faster per point.
Does ESOV work in B2B and niche categories? Yes — the relationship holds wherever advertising builds memory ahead of purchase, which includes long-cycle B2B. The catch is the denominator: define the market you actually compete for, and measure voice with consistent proxies rather than published spend.
based. is a strategy, media and creative agency that builds fame-first marketing for challenger B2B and DTC brands. If you don't know your own ESOV — or suspect it went negative the last time "efficiency" won a budget meeting — the B2B marketing effectiveness audit computes it as part of the workup.
ESOV says growth can be bought. The multiplier says it can also be earned. Challengers live on the multiplier.