What is intent data?

Intent data providers track the content that company employees read and consume across a network of sites, then compare that activity against a historical baseline for each account. When a company's employees suddenly consume a cluster of articles about "sales automation software" or "CRM migration" in a compressed window, that pattern signals active research rather than casual browsing.

The mechanics differ by type. Third-party providers like Bombora run a data cooperative: 5,000+ B2B publishers share anonymized content consumption data, which Bombora's NLP models classify into a 21,600+ B2B topic taxonomy. A Company Surge Score from 0 to 100 reflects how much an account's research on a topic has spiked above its 12-week baseline; Bombora treats a score of 60+ as a statistically significant surge. Second-party providers like G2 share their own first-party data — who visited your category page, who read competitor reviews — under a commercial arrangement. First-party tools de-anonymize visits to your own site using IP resolution and identity graphs.

All three approaches feed into the same workflow: identify accounts in active research mode, score them by fit and signal strength, and route them to sales or activate them in ad targeting — ideally within 24–48 hours, before the signal decays and the account moves on.

First-party intent data is the most accurate type because you collect it directly. It covers every action a known or anonymous account takes on your own digital properties — pricing page visits, product demo requests, documentation views, email clicks, trial usage. First-party data is real-time, highly trustworthy, and has no marginal cost to collect. The constraint is reach: it only covers the roughly 5% of your total addressable market that has already found you.

Second-party intent data is someone else's first-party data shared with you under a commercial arrangement. The clearest example is review-site intent: G2 can tell you which companies in your category are actively reading competitor reviews or comparing features on its platform. Because these buyers are further down the funnel, second-party signals tend to be high quality and actionable — Dreamdata research found G2 intent influenced 12% of closed-won deals, and those deals were twice as large as deals that lacked a G2 signal.

Third-party intent data is aggregated across the open web by a co-op or publisher network. It covers the widest range of accounts — including ones that have never interacted with you — but it is also the noisiest, most commoditized, and often most stale. Data typically arrives in weekly batches, and every competitor with access to the same co-op sees the same surge signals at the same time. The key trade-off: breadth versus precision. Third-party data surfaces new in-market accounts; first-party data confirms and prioritizes them.

The core argument rests on a timing problem. Gartner research (2025) shows 70–80% of the B2B purchase journey happens before a buyer ever speaks to a vendor. Forrester's 2024 Buyers' Journey Survey found that 92% of B2B buyers enter a purchase process with at least one vendor already in mind, and 41% have a single preferred vendor selected before formal evaluation even begins. By the time a buyer fills out a demo form, the competitive race is largely over.

Intent data lets sellers engage earlier — during the research phase, when they can still shape the shortlist. The performance data from teams that use it is largely positive: companies using intent-based outreach programs report 93% higher conversion rates (LinkedIn research, cited by Landbase), and 96% of B2B marketers who use intent data report successfully achieving their stated goals. ZoomInfo case studies include Safety Services generating 200% more MQLs in its first month and Redwood Logistics cutting cost-per-click by 99% by concentrating ad spend only on in-market accounts.

The important caveat: 31% of sales leaders have called intent data "the most overrated technology in their stack" (widely reported, including by Warmly), and DemandScience's 2025 study of 750 senior marketing leaders found 87% of organizations report their marketing investments generate unreliable or inflated signals. The gap between expectation and results usually traces to poor ICP definition, ignoring signal decay, or acting on third-party intent alone without layering first-party confirmation.

Intent data is a subset of buying signals. Buying signals are any observable event that suggests readiness or urgency to buy — job changes, funding rounds, new hires in a specific function, technology installs, product-usage milestones, and research spikes. Intent data specifically captures the research-behavior layer: content consumption and search activity that indicates a topic is being evaluated.

A practical way to think about it: intent data tells you an account might be evaluating your category; a buying signal like a champion changing jobs tells you an account has just made a move that creates a specific buying need. Both are valuable, but they answer different questions. Intent data answers "who is in active research mode right now?" Event-based signals like funding or leadership changes answer "what just happened that should trigger outreach today?"

The strongest GTM programs treat intent data as the market-wide radar — surfacing which accounts to watch — and then layer real-world event signals to decide when to act and who to contact. This is the foundation of signal-based selling: a methodology that treats intent as one important input in a broader stack of triggers rather than a standalone automation trigger.

The most common use cases fall into three categories. The first is outbound prioritization: instead of calling down a static territory list, reps use intent scores to focus this week's effort on accounts actively researching their category. The second is account-based advertising: suppressing budget on low-intent accounts and concentrating spend on those showing a surge — a strategy Redwood Logistics used to cut cost-per-click by 99% and increase click-through rates by 310% using ZoomInfo intent.

The third — and often highest-value — use case is timing and personalization. When an account shows a surge on "revenue intelligence" and a rep reaches out referencing the problem the buyer is clearly researching, the outreach lands as helpful and timely rather than random. Message personalization tied to intent topics consistently lifts reply rates versus generic sequencing.

The operational requirement that separates successful programs from failed ones is speed. Intent signals decay fast — a surge that is two weeks old may reflect research that has already concluded and a vendor that has already been selected. Teams that extract the most value from intent data have a workflow that routes fresh signals to reps within 24–48 hours, with enriched account context and a ready-to-personalize message, not a dashboard row that gets checked monthly.

The dark funnel refers to the portion of the B2B buying journey that is invisible to traditional analytics — peer conversations, private Slack communities, LinkedIn DMs, Reddit threads, review sites, and increasingly, AI chatbot research. Gartner estimates 70–80% of the purchase journey now happens in these untracked channels before a buyer ever raises their hand to a vendor. When a buying committee member asks ChatGPT or Perplexity to compare CRM tools and then shares the response in a private Slack channel, that activity leaves no trace in any vendor's CRM or web analytics.

Intent data illuminates part of this dark funnel by tracking content consumption across publisher co-ops and review sites. A third-party provider like Bombora monitors what company employees read across 5,000+ B2B publisher sites — catching research that happens on industry blogs, analyst sites, and comparison platforms long before the buyer ever visits a vendor's website. Second-party signals from review sites like G2 add another layer, capturing accounts actively comparing products in a category.

The critical limitation: intent data captures the portion of the dark funnel that flows through trackable publisher networks. Research that happens in closed communities, over the phone, in AI chat sessions, or through peer referrals remains invisible even to the best intent data programs. This is why intent data is most powerful when combined with engagement signals, relationship intelligence, and event-based triggers rather than treated as a complete picture of buyer intent.

Komo monitors intent signals alongside the full range of buying signals — job changes, funding, hiring patterns, technographic installs — so a surging account does not just become a row in a report. When an account spikes on a relevant topic, Komo automatically researches the account and the right contacts, drafts a personalized outreach message grounded in the signal context, and queues it for human review.

The human-in-the-loop model matters here specifically because intent data can be noisy — a competitor checking a category, an analyst writing a report, or a student doing research can all produce a surge signal on the wrong account. Rather than firing automated sequences at every spike, Komo presents the signal with context — why this account, why now, what they appear to be researching — and keeps a human on every send that matters.

The result is that intent data stops being a dashboard metric and starts being a daily driver of booked meetings — which is the gap that most intent data investments fall into.