SfSkills Domains Docs GitHub

Home / admin / mcae-lead-scoring-and-grading

mcae-lead-scoring-and-grading

Use this skill when configuring MCAE (Account Engagement / Pardot) lead scoring models, grading profiles, score decay rules, or automation rules that fire on score/grade thresholds. Covers: scoring point values per activity, score decay for inactivity, Profiles for fit grading, combined MQL definitions (Score + Grade), and automation rules vs completion actions. NOT for defining the MQL model before build — use admin/lead-scoring-requirements. NOT for Engagement Studio nurture programs — use admin/lead-nurture-journey-design.

Open the skill folder on GitHub

MCAE Lead Scoring and Grading

Use this skill to configure the MCAE (Account Engagement / Pardot) behavioral scoring model and fit-based grading system. It activates when a practitioner needs to set up or tune scoring rules, score decay, Profiles, automation rules tied to score/grade thresholds, or define the combined MQL trigger for sales handoff.

Before Starting

Gather this context before working on anything in this domain:

Core Concepts

1. Scoring — Behavioral Engagement Dimension

Scoring measures how much a prospect has engaged with your marketing assets. It is a cumulative integer starting at 0. Points are added when a prospect performs a trackable activity and can be removed (deducted) for negative signals such as unsubscribing or opting out.

Default activity point values (configurable in MCAE > Admin > Scoring):

ActivityDefault Points
Form fill (submission)+50
Form fill (error)+5
Email link click+5
Email open+1
Page view (tracked)+1
Custom redirect click+5
File download+5
Webinar registration+10
Unsubscribe-50
Hard bounce-75

Points are configurable globally under Admin > Scoring. Per-asset overrides are also possible (e.g., a high-intent pricing page can award more points than a generic blog visit). Custom activities and page view scoring require the MCAE tracking code (JavaScript snippet) on the web property.

Score is only meaningful in context of the prospect's activity recency — which is why decay rules exist.

2. Score Decay — Time-Based Score Reduction

Score decay automatically reduces a prospect's score over time to ensure that old engagement does not keep a prospect at MQL indefinitely. Decay rules are configured in Admin > Scoring > Decay Scoring.

Key behaviors:

3. Grading — Fit Dimension via Profiles

Grading is an independent dimension from scoring. It measures how well a prospect matches your Ideal Customer Profile (ICP). Grade is a letter (A++ through F) assigned by matching the prospect's field values against a Profile.

A Profile contains criteria — each criterion specifies a field value match and a positive or negative grade point adjustment. MCAE converts the cumulative grade points into a letter grade. Criteria examples:

Each prospect is matched against exactly one Profile (the matching profile is selected based on the prospect's field values; only one profile can be the "default"). Profile criteria are evaluated whenever the prospect's fields update or the prospect is synced.

The letter grade map (approximate): 100 = A++, 85 = A+, 70 = A, 55 = B+, 40 = B, 25 = C, etc. Exact thresholds are fixed by the platform.

4. Automation Rules vs. Completion Actions

These are two distinct mechanisms for triggering score/grade-based actions. Confusing them is a common source of missed leads or duplicate actions.

Automation Rules (Admin > Automation > Automation Rules):

Completion Actions (on forms, emails, files, custom redirects):

Common Patterns

Pattern 1: Combined MQL Definition (Score + Grade Gate)

When to use: Most B2B orgs should use both dimensions to qualify leads. Score alone misses low-fit contacts who happen to be very active; grade alone misses high-fit contacts who haven't engaged yet.

How it works:

  1. Define the MQL threshold collaboratively with Sales: typically "Score >= 100 AND Grade >= B."
  2. Build one Automation Rule with two criteria: Prospect Score >= 100 AND Prospect Grade is B or better.
  3. Set the action to: Assign to [Sales Queue or User] + Notify assigned user + Add to "MQL" list.
  4. Set the rule to "repeat" only after the prospect's score drops below threshold (use score reset on recycle to prevent infinite reassignments).

Why not Score alone: Without a grade gate, marketing-active but low-fit prospects (competitors, students, wrong geography) flood the Sales queue with noise.

Pattern 2: Score Decay to Reflect Recency

When to use: Any org where prospects accumulate score from old campaigns and remain MQL-eligible indefinitely even after going dark.

How it works:

  1. Go to Admin > Scoring > Decay Scoring.
  2. Add a rule: Reduce score by 10 points after 30 days of inactivity.
  3. Optionally add a second rule: Reduce score by 20 points after 60 days of inactivity (stacks with first rule).
  4. Audit the MQL threshold — if decay is in place, the threshold should reflect expected score ranges for recently active prospects.

Why not rely on list management alone: Manual list cleanup is error-prone and creates data quality debt. Decay is automated and keeps scores reflective of actual recency without admin intervention.

Pattern 3: Profile Criteria for Multi-Segment Grading

When to use: Orgs with multiple ICP segments (e.g., SMB vs. Enterprise, Healthcare vs. Technology) where fit criteria differ by segment.

How it works:

  1. Create a separate Profile for each ICP segment (e.g., "Enterprise Technology" and "SMB Healthcare").
  2. Add matching criteria to each profile. Use the default Profile as a catch-all for prospects that don't match any segment-specific profile.
  3. The profile with the most matching criteria for a given prospect is automatically selected.
  4. Grade thresholds in the MQL automation rule apply across all profiles — the letter grade is normalized regardless of which profile was used.

Decision Guidance

SituationRecommended ApproachReason
Need to route a prospect to Sales when they hit a score+grade thresholdAutomation Rule with score AND grade criteriaAutomation rules evaluate continuously; completion actions fire only on a single event
Need to send an immediate follow-up email after a form fillCompletion Action on the formReal-time, event-scoped; no need to evaluate the whole database
Prospects are staying MQL-eligible from old campaignsScore Decay rules (30-day and 60-day)Automated recency signaling without manual cleanup
Multiple ICP segments with different fit criteriaMultiple Profiles with segment-specific criteriaOne Profile per segment; platform selects best match per prospect
Sales complains about low-quality leads in queueAdd or tighten grade gate in MQL automation ruleGrade filters out high-engagement, low-fit contacts
Admin wants to manually promote a prospect to MQLManual score/grade override in MCAE prospect recordOverride blocks automatic re-scoring; document the override reason
Custom web activity (e.g., pricing page) should score higherPer-asset score override on the specific page/formGlobal score rules still apply to all other assets

Step-by-step instructions for an AI agent or practitioner working on this task:

  1. Confirm prerequisites. Verify that the MCAE Business Unit is connected to the Salesforce CRM and that the Score and Grade fields are mapped in the connector field mapping. Without this, score/grade changes never reach Lead/Contact records.
  2. Document the MQL definition. Before touching any configuration, capture the agreed score threshold and grade threshold in writing (e.g., "Score >= 100 AND Grade >= B"). Get explicit sign-off from Sales. Record this in the work template.
  3. Configure global scoring rules. Go to MCAE Admin > Scoring. Review and adjust the default point values for each activity type to reflect the org's priorities. Document the final point table. Add per-asset score overrides for high-intent assets (e.g., pricing page, demo request form).
  4. Set up score decay. Go to Admin > Scoring > Decay Scoring. Configure at minimum one decay rule (e.g., -10 points after 30 days of inactivity). Add a second rule for 60 days if the sales cycle is long. Verify that the MQL threshold is still achievable with decay in place.
  5. Build Profiles for grading. Go to Admin > Profiles. Create a profile for each ICP segment. Add positive criteria for strong fit signals (job title, industry, company size) and negative criteria for disqualifying signals (competitors, wrong geography, student roles). Set one Profile as the default catch-all.
  6. Create the MQL Automation Rule. Go to Automation > Automation Rules. Create a rule with criteria: Score >= [threshold] AND Grade >= [letter]. Set the action: Assign to [Sales queue], notify [Sales user or group], add to [MQL Salesforce Campaign or list]. Set the rule to "repeat" and decide the reset logic (e.g., reset if prospect score drops below threshold).
  7. Test and validate. Create a test prospect. Simulate activities (form fill, page view, email click) and confirm score increments correctly. Verify decay fires on schedule. Check that the automation rule fires when threshold is crossed and that Score and Grade appear on the synced Lead/Contact record in CRM.

Review Checklist

Run through these before marking work in this area complete:

Salesforce-Specific Gotchas

Non-obvious platform behaviors that cause real production problems:

  1. Manual override blocks automatic re-scoring — When an admin manually overrides a prospect's score or grade in the MCAE UI, MCAE marks that record as manually overridden and stops updating it automatically. Future activities do not change the score. This is a silent block — there is no UI indicator visible to other admins. Teams discover the problem when a re-engaged prospect never moves through scoring. Fix: document the override policy and review override records regularly.
  2. Automation Rules are retroactive by default — When you activate a new Automation Rule, MCAE immediately evaluates all existing prospects against the rule criteria, not just new prospects going forward. If your rule assigns leads to Sales, activating it can flood the Sales queue with thousands of existing prospects at once. Fix: before activating a threshold-based rule on an existing database, filter by a "date created after X" criterion or run the rule in a lower-impact test segment first.
  3. Score decay does not trigger Automation Rules — Score decay silently reduces scores without firing completion actions or re-evaluating automation rules. If a prospect was above your MQL threshold and decay drops them below it, no rule fires to notify Sales or remove the prospect from the MQL list. You must build a separate recycling rule to handle score-drop scenarios, or rely on periodic CRM-side reporting.
  4. Prospect score and grade sync is one-way (MCAE → CRM) by default — Score and Grade flow from MCAE to the Lead/Contact fields in Salesforce. If a CRM workflow or flow overwrites the Score field on the Lead record, that change is not reflected back in MCAE. This creates divergence between MCAE score and CRM-visible score. Fix: protect the Score and Grade fields on Lead/Contact with field-level security or validation rules to prevent non-MCAE writes.
  5. Profile matching is based on the best-fit profile, not an explicit assignment — MCAE automatically selects the profile for each prospect based on how many criteria match. If a prospect matches criteria from multiple profiles equally, the behavior is non-deterministic. Avoid overlapping criteria across profiles. Always have one clearly designated default profile as a catch-all.

Output Artifacts

ArtifactDescription
Scoring point tableDocumented list of activity types and their configured point values (filled in the work template)
Score decay rulesActive decay rules with period and point-reduction values
Profile definitionsOne profile per ICP segment listing all positive and negative criteria
MQL Automation RuleActive rule with score+grade criteria and assignment/notification actions
CRM field mapping confirmationEvidence that Score and Grade fields are mapped in the MCAE connector
MQL definition documentSigned-off record of the score threshold, grade threshold, and reset logic

Dependencies

Package files