Marketing Intent Engine
Unify external and internal buying signals to time and personalize marketing
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 7
- Confidence
- 99%
Define the ideal customer profile, then build a context layer from external and internal signals that indicate fit, timing, and purchase interest. External signals can include funding, hiring, job changes, product launches, research activity, or firmographic changes. Internal signals can include pricing-page visits, high-value content consumption, and meaningful actions inside a freemium product. Normalize these events into useful fields, distinguish weak curiosity from strong intent, and combine signal strength with customer fit. Make the resulting context available to AI-assisted campaigns, personalization systems, and sales workflows. Continuously compare signals with downstream outcomes so the engine learns which events actually predict conversion. The engine is foundational: richer, more relevant context supports better timing, more credible personalization, and smaller audience segments without relying on generic mass messaging.
Origin
Extracted from Marketing Against The Grain as the foundational context layer powering the hosts' later AI prospecting and micro-audience strategies.
Core principles
- 01Context determines the quality of AI experiences
- 02Intent signals must be specific to the ideal customer profile
- 03External events reveal changing opportunity
- 04Internal behavior reveals direct product interest
- 05Better context enables smaller and more relevant audiences
How to run it
- 1
Define the customer profile
Specify the companies and people most likely to benefit from and buy the product.
Pro tip Include disqualifying attributes as well as positive fit criteria.
Watch out A vague ICP makes collected intent data noisy.
- 2
Map external signals
Identify observable events that suggest a target account is entering a relevant buying window.
Pro tip Consider hiring, funding, executive changes, launches, research behavior, and technology shifts.
Watch out An event is not useful merely because it is easy to scrape.
- 3
Map internal signals
Identify owned behavioral events that indicate interest in the product, category, or purchase process.
Pro tip Weight high-intent actions such as pricing views or meaningful product activation more heavily.
Watch out Do not equate every website visit with purchase intent.
- 4
Unify the context
Normalize account, person, timing, and behavioral fields so systems can interpret signals together.
Pro tip Preserve the source and time of every signal.
Watch out Poor identity resolution can attach activity to the wrong person or account.
- 5
Score fit and intent
Combine ICP fit, signal strength, recency, and relevant combinations into actionable priorities.
Pro tip Use simple, inspectable rules before introducing complex models.
Watch out Do not let opaque scores determine high-impact actions without review.
- 6
Activate the engine
Supply the context to campaigns, AI agents, and sales workflows so messaging and timing reflect the current situation.
Pro tip Expose the evidence behind a recommendation to human users.
Watch out Personalization can feel invasive when it reveals excessive surveillance.
- 7
Calibrate against outcomes
Compare signals and scores with qualified meetings, opportunities, and purchases, then revise the engine.
Pro tip Evaluate predictive value by segment rather than relying only on global averages.
Watch out Do not optimize against superficial engagement metrics alone.
In the wild
A software company detects that an ICP account has raised funding, opened relevant job roles, visited the pricing page, and activated an advanced freemium feature. The intent engine combines these external and internal signals and routes the account into a timely, context-aware campaign.
→ Marketing prioritizes an account with both organizational capacity and demonstrated product interest.
Common mistakes
Collecting context without relevance
Large quantities of available data do not help if the fields have no plausible relationship to fit, timing, or intent.
Using one signal in isolation
A job change or page view can be ambiguous; combinations often provide more reliable buying context.
Ignoring privacy boundaries
Intent systems must use data lawfully and avoid personalization that exposes inappropriate surveillance.
Is it for you?
Best for
It is best for B2B companies with identifiable customer profiles and access to behavioral, firmographic, or event data.
Not ideal for
It is not ideal for organizations that cannot collect or use the required data lawfully, accurately, and transparently.
From the transcript
“So your intent engine is really how you get your internal and external intent signals around your ideal customer profile.”
“You need to transform how you time and personalize your messaging.”
“And so that context layer, which is your intent engine, is going to be do or die for having an incredible AI powered marketing machine.”
From the episode
6 AI Marketing Strategies To 3x Conversion in 2025