Measuring Brand in B2B: Share of Voice, Branded Search, and Honest Proxies

The reason brand measurement frustrates B2B teams is that they keep reaching for the wrong instrument. Attribution models are built to connect a touch to a conversion inside a trackable window, and brand works precisely outside that window: in memory, in Slack recommendations, in the shortlist someone writes down before ever visiting your site. No dashboard will hand you a number called brand ROI, and vendors who claim otherwise are usually selling you a media mix model wearing a costume. The workable alternative is triangulation, several imperfect proxies that, moving together, tell a story no single metric could.

Triangulation means you stop asking did this brand campaign generate pipeline and start asking are more of the right people thinking of us, more often, more favorably, than they were two quarters ago. That question has observable consequences even when the causal chain is invisible. People who think of you search your name, visit your site directly, mention you in communities, reply to your outreach at higher rates, and say I have heard good things on first calls. Each of those leaves a trace you can measure, as long as you resist the urge to dress the traces up as attribution.

Branded search volume, the trend in how many people search your company or product name, is probably the single most useful brand proxy in B2B because it is close to pure intent to find you specifically. Pull it from your search console data and keyword tools, chart it monthly, and watch the trend rather than the absolute number. A steadily rising branded search line while your category terms stay flat is a strong sign brand activity is landing. The caveats: PR spikes and funding announcements inflate it temporarily, and rebrands or product renames break the series, so annotate those events on the chart.

Direct traffic and dark-referral traffic tell a similar story from a different angle: people typing your URL or clicking links shared in places analytics cannot see, like private Slack groups and DMs. Rising direct traffic alongside rising branded search is corroboration; one moving without the other deserves a closer look. Also watch the branded-to-unbranded ratio in your paid search account. When brand strengthens, you often find branded clicks getting cheaper and unbranded capture converting better, because the click is a re-encounter rather than a first meeting.

Share of voice asks what fraction of your category's conversation mentions you versus competitors. In B2B, the practical version is not global media monitoring but a defined arena: the podcasts, newsletters, communities, analyst notes, and social conversations where your actual buyers pay attention. Count mentions there, yours and competitors', on a consistent method, and track the trend. There is a long-observed relationship in marketing research between share of voice exceeding share of market and subsequent growth, and while the precise math varies by category, the direction is a useful planning instinct: to grow share, you typically need to be more present in the conversation than your current size would predict.

The honest caveats are real. Mention counting rewards volume over sentiment, so a pricing controversy counts as voice. Automated tools miss the private channels where much B2B discussion actually happens. And self-serving arena definitions, counting only the communities where you happen to be strong, turn the metric into a mirror. Define the arena once, include the venues where competitors beat you, and hold the definition stable so the trend means something.

Brand strength eventually shows up in sales mechanics, and those traces are worth mining. Win rate against known competitors is one: when your brand strengthens, you tend to lose fewer deals to we went with the vendor we had heard of. Time-to-close and discount depth often move too, because a trusted vendor gets less procurement friction and less price pressure. None of these are clean brand metrics, they are confounded by product, pricing, and sales skill, but tracked over quarters alongside the other proxies, their direction is informative.

The most underrated instrument is simply asking. A how did you hear about us field on demo forms, answered in free text rather than a dropdown, and a first-call habit of asking what prompted the buyer to reach out now, produce self-reported attribution that consistently surfaces the invisible channels: a colleague's recommendation, a podcast, a post someone half-remembers. It is biased toward the memorable and vague on timing, and it is still often the closest thing you will get to ground truth about how brand actually operates. Feed it into the same review as your other signals so the qualitative and quantitative pictures can argue with each other productively.