Soku AIIntent-based targeting focuses on reaching users who are actively showing purchase intent — researching products, comparing options, reading reviews, visiting pricing pages, and performing other actions that signal an upcoming purchase decision. Rather than targeting users based on who they are or what they have done in the past, intent-based targeting identifies users based on what they are about to do.
This approach sits at the intersection of behavioral targeting and predictive audience targeting. It uses behavioral signals to infer current intent, then targets users who are in an active decision-making phase — the moment when advertising has the highest probability of influencing a purchase.
The term is also written as intention-based targeting, and the two are used interchangeably in practice. Neither is a formal platform term: Google calls its version in-market audiences, Meta describes purchase intent categories, and LinkedIn markets buyer intent. "Intent-based targeting" is the vendor-neutral name for the whole class.
How intent-based targeting works
Intent signal identification is the foundation. Strong intent signals include category-specific search queries ("best project management software for teams"), comparison content consumption (reading "Asana vs. Monday.com" articles), pricing page visits, demo request page views, and product review engagement. Weak intent signals include general category browsing and social media engagement with category content.
Signal aggregation combines multiple weak signals into strong intent indicators. A single visit to a product category page is weak evidence of intent. But a user who searched for the product category, read two comparison articles, visited three vendor websites, and watched a product demo video in the past week demonstrates strong purchase intent.
In-market audience construction groups users by intent strength and category. Google's In-Market Audiences and Meta's purchase intent categories are pre-built intent-based segments that advertisers can target directly. Custom intent audiences can be built using search query data, website visit patterns, and content engagement signals.
Real-time intent scoring assigns each user a dynamic intent score that fluctuates as new behavioral signals are observed. Users with rising intent scores receive higher bid priority, while users whose intent signals have gone cold are deprioritized.
Why intent-based targeting matters
Timing is everything in advertising. Reaching a user during active research is far more effective than reaching them before they are in-market or after they have already purchased. Intent-based targeting aligns ad exposure with the decision window, maximizing the probability of influence.
Higher conversion rates result from reaching users who are already predisposed to purchase. Intent-based campaigns typically achieve 2–5x higher conversion rates than awareness-focused campaigns because the audience is further along the purchase journey.
Efficient spend allocation concentrates budget on users most likely to convert in the near term. This reduces waste on users who are not ready to buy and improves ROAS) and CPA) metrics. Platforms like Soku AI help advertisers identify and target intent signals across multiple ad platforms simultaneously, ensuring no high-intent user is missed.
Competitive advantage comes from reaching prospects during the consideration phase. Users researching a category are comparing options — an advertiser who appears during this comparison phase has the opportunity to shape the decision.
Where the signals actually come from
Intent is inferred, never observed directly, and the inference quality depends entirely on the source. Ranked roughly by reliability:
- Explicit search queries. The strongest signal, because the user typed their intent. This is why search advertising has always been the highest-converting channel — the targeting is the query.
- First-party behaviour on your own properties. Pricing page visits, demo requests, repeat sessions on a comparison page. Reliable, fully owned, and unaffected by third-party signal loss — but limited to people who already found you.
- Platform-owned in-market segments. Google's in-market audiences and Meta's purchase-intent categories, built from behaviour across the platform's own surfaces. Broad reach, no visibility into how the segment was constructed.
- Third-party intent data. Aggregated cross-site browsing, common in B2B. The weakest of the four and the most affected by cookie deprecation.
The practical implication: as third-party signals degrade, the centre of gravity moves toward the first two. Teams that treat intent-based targeting as something you buy from a vendor are working from the weakest end of that list.
Intent-based targeting under automated bidding
The category has quietly changed shape. On automated bidding surfaces — Performance Max, Advantage+, AI Max for Search — you increasingly do not select an intent audience at all. You supply conversion signal and the platform's model infers intent itself.
That shifts the work in two ways:
- From audience selection to signal quality. If the model is inferring intent from your conversion events, the completeness and accuracy of those events is your real targeting control. This is why server-side tracking has become a bidding input rather than a reporting improvement.
- From targeting to exclusion. With keyword-level and audience-level control reduced, negative keywords, brand exclusions and audience suppression become the sharper instruments. You steer more by saying what not to reach.
Intent-based targeting as a manual discipline still applies fully on search campaigns with keyword control and on first-party audience building. On automated surfaces it has largely become an emergent property of the signal you feed.
Challenges and considerations
Intent signal accuracy varies by source. Search queries are the strongest intent signals because they reflect explicit user actions. Inferred intent from browsing behavior is weaker and more prone to false positives — a journalist researching an article about CRM software shows similar behavioral signals to an actual CRM buyer.
Short intent windows mean timing is critical. For many purchase categories, the active research phase lasts only days. By the time behavioral data is collected, processed, and used for targeting, the user may have already made their decision. Real-time or near-real-time signal processing is essential.
Privacy-driven signal loss is reducing the availability of cross-site intent data. As third-party cookies disappear and tracking restrictions increase, intent signals that rely on cross-site browsing data become less reliable. Contextual targeting and first-party data strategies are becoming important complements.
Competition for intent audiences drives up costs. Every advertiser in a category wants to reach users showing purchase intent, which concentrates bidding pressure on these high-value audiences. Balancing intent-based targeting with upper-funnel prospecting ensures a sustainable customer acquisition strategy.
Over-indexing on bottom-funnel can starve future demand. If all budget targets users who are already in-market, the advertiser becomes dependent on demand generated by competitors and organic discovery. A healthy targeting strategy combines intent-based targeting with awareness and consideration campaigns that create future intent.
