← Back to the catalog
B3SALES & REVENUE

Lead scoring and intent detection

Watches activity across the systems where leads show up — website visits, content downloads, email opens, support inquiries, product trial signups — and produces a continuously updated score per lead: how likely they are to be a fit, how engaged they are, what they appear to be interested in. Surfaces high-scoring leads to sales in priority order and flags accounts where engagement patterns suggest active buying intent. Replaces the gut-feel triage of marketing-qualified leads with something measurable.

WHERE THIS FITS
BUSINESS SHAPES
B2B servicesProduct company
VOLUME THRESHOLD
Below 100 new leads per month a month, the payback rarely earns the build. Patterns this shape reliably pay back at 1,000+.
FITS BEST
Funnel-driven B2B with existing won/lost data.
PAYBACK · 4-7 moBUILD · MediumVALUE · $60k-$200kWHEN · >500 leads/quarter
FAILURE MODE TO DESIGN AROUND
Training data biased toward leads the team historically engaged with. Build a shadow-review of dropped leads.
REQUIREMENTS · 5 REQUIRED, 1 OPTIONAL

Requirements describe capabilities the pattern needs in your environment, not the vendors you must buy. Any system that fills a requirement satisfies it — that’s what makes the catalog portable across the long tail of SMB tooling.

  1. lead_record_store
    REQUIREDREAD + WRITErequest

    Where lead records live. The pattern writes scores back to these records so they're visible to sales reps in their normal working surface.

    DATA SHAPE
    Lead records with identity, source, current status, history of stage changes, and a scoring field the pattern writes to.
    COMMONLY FILLED BY
    • CRM with lead records
    • marketing automation system with lead lifecycle
    • internal lead database
  2. engagement_event_stream
    REQUIREDREADevent

    Stream of behavioral events: page visits, content downloads, email opens, product activity. The raw material the score is computed from.

    DATA SHAPE
    Per-event records with lead/visitor identifier, event type, timestamp, and event-specific properties. Volume can be high.
    COMMONLY FILLED BY
    • analytics platform with event tracking
    • marketing automation activity stream
    • product usage event log
    • combined event bus aggregating multiple sources
  3. firmographic_data_source
    REQUIREDREADrequest

    Company-level attributes that determine fit: industry, size, geography, tech stack. Distinct from engagement; engagement is what they do, firmographics is what they are.

    DATA SHAPE
    Per-company records: industry codes, employee count, revenue band, geography, technologies in use.
    COMMONLY FILLED BY
    • data enrichment service appended to CRM records
    • imported firmographic data from a list provider
    • internal company profiles
  4. score_destination
    REQUIREDWRITEevent

    Where the score becomes visible to sales reps. Inside their existing working surface, not a separate dashboard they have to remember to open.

    DATA SHAPE
    Per-lead score with subscores (fit, engagement, intent) and recent-change indicator. Updated whenever the underlying signals change materially.
    COMMONLY FILLED BY
    • field on the lead record in the CRM
    • lead queue sorted by score in the sales tool
    • Slack alerts for high-score changes
  5. score_definition_inputs
    REQUIREDREADcorpus

    What 'good' means for this client. The pattern doesn't invent scoring criteria; it operationalizes ones the client provides.

    DATA SHAPE
    Ideal customer profile description, key engagement signals that matter (with weights or relative importance), examples of converted vs. unconverted leads from history.
    COMMONLY FILLED BY
    • ICP document maintained by marketing leadership
    • scoring configuration in a small admin UI
    • historical conversion data the pattern learns from
  6. intent_signal_external
    OPTIONALREADbatch

    External buying-intent signals beyond what the client's own systems see. Optional but powerful.

    DATA SHAPE
    Per-account signals like research activity on third-party sites, hiring patterns suggesting expansion, technology adoption.
    IF MISSING
    Pattern still produces scores based on first-party signals. External intent typically adds about 15-20% lift in catching deals earlier; not transformative.
    COMMONLY FILLED BY
    • third-party intent data feed
    • manual addition of competitor or industry trigger lists
    • news monitoring system surfacing relevant company events
RUNTIME FLOW · 7 STEPS
  1. 01
    Continuously ingest engagement events as they happen
    engagement_event_stream
  2. 02
    For each affected lead, fetch firmographic data and any external intent signals
    firmographic_data_sourceintent_signal_external
  3. 03
    Compute fit subscore from firmographics matched against ICP definition
    score_definition_inputs
  4. 04
    Compute engagement subscore from recent activity weighted by recency and signal strength
    score_definition_inputs
  5. 05
    Compute intent subscore from buying-pattern signals: research-heavy visits, multiple stakeholders engaging, trial usage
  6. 06
    Combine into overall score; write back to lead record
    lead_record_store
  7. 07
    If a lead crosses a threshold or jumps significantly, surface to the score destination with a recent-change indicator
    score_destination
    DECISION Only surface on threshold crossings or large changes; constant updates create alert fatigue.
EMISSIONS · 3

Structured outputs this pattern produces. Other patterns and client systems can subscribe to them, which is how the catalog composes over time.

  • qualified_lead_pipeline

    Stream of leads crossing into qualified status, with reason codes.

    CONSUMED BY
    • sales rep alerting
    • marketing attribution dashboards
    • pipeline forecasting
  • scoring_calibration_signal

    Per-lead score vs. eventual conversion outcome, used to tune the model over time.

    CONSUMED BY
    • scoring model refinement
    • monthly RevOps review
  • icp_drift_signal

    Patterns where the ICP description doesn't match converted-customer reality, surfaced as candidates for ICP refinement.

    CONSUMED BY
    • marketing leadership
    • annual strategy reviews