Marketo lead scoring: how to build a scoring model in Adobe Marketo Engage
Marketo has no scoring screen. You build the model from score fields, smart campaigns, and tokens. Here is how, following Adobe's own documentation.
Marketo lead scoring is a model you build in Adobe Marketo Engage out of score fields and smart campaigns. Each campaign listens for a behavior or a field value, then runs the Change Score flow action to add, subtract, or set points on a person’s score. Most teams keep three fields: Behavior Score, Demographic Score, and the combined Person Score.
There is no single scoring screen in Marketo Engage. Adobe’s glossary defines a scoring model as “a framework that assigns point values to person records based on demographic attributes and behavioral actions,” and the framework is yours to assemble. This guide follows Adobe’s own documentation, linked where it is used. Where the docs are thin, check your account.
Marketo lead scoring at a glance
| Question | Answer |
|---|---|
| Where is it? | Marketing Activities, in smart campaigns you build. Score fields are created in Admin, then Field Management |
| Which plans? | The Adobe pages used here do not tie Change Score to an edition. Account Score needs Target Account Management. Check your account |
| What does it score? | People (leads and contacts). With Target Account Management, named accounts get an Account Score |
| How points work | The Change Score flow action takes +5 to add, -5 to subtract, and =5 or =-5 to set an exact value |
| Decay | Not a setting. You build it as a campaign that subtracts points after inactivity |
| AI or predictive | Not in the Marketo Engage docs used here. Adobe documents predictive scoring in Real-Time CDP B2B Edition |
| How reps see it | Score fields synced to the CRM, plus stars and flames in Marketo Sales Insight |
Before you open Marketo: plan the model
Adobe’s guide to building person scoring models starts on paper, with sales in the room.
- Fit factors. Three to six attributes your won deals share, such as industry, company size, job title, and department. Your ideal customer profile is the source, and our five-step framework for building an ICP scoring model covers this side.
- Engagement signals. Adobe’s examples are pricing page visits, events, webinars, white paper downloads, and demo request forms.
- Negative signals. Adobe lists no interaction in 90 days, careers page visits, and unsubscribes.
- The handoff threshold. Adobe’s example is 50 points, with the advice that it “should be large enough so that a person needs to complete multiple interactions with your brand to meet it.”
Two cautions. Marketo has an Opens Email trigger, but Litmus reports that over 50% of email opens come from devices with Apple Mail Privacy Protection, so score clicks and forms instead. And a demographic rule can only fire on a field that is filled in: in Validity’s 2025 survey, 76% of CRM users said less than half of their data is accurate and complete.
How to set up lead scoring in Marketo, step by step
Step 1: Create the score fields
Person Score is the standard field. For the other two, follow Adobe’s steps to create a custom field: go to Admin, then Field Management, click New Custom Field, choose the Score type, and name it. Adobe’s program library expects the names Behavior Score (API name BehaviorScore) and Demographic Score (API name DemographicScore).
Adobe warns to “choose carefully, as it cannot be renamed after saving.” If you want the field in your CRM, Adobe’s tip is to create it there first, so it appears in Marketo through the sync.
Step 2: Create or import the scoring programs
Adobe’s new instance checklist recommends a tokenized person scoring program that breaks out behavior and demographic scores, and says to import one from the Marketo Program Library to start. The two library programs are OP-Scoring-Behavior and OP-Scoring-Demographic.
The behavior program ships with campaigns for email clicks, forms, key page visits, webinars, and events, plus a Score Decay folder. The demographic program covers annual revenue, industry, job title, number of employees, source, and generic email domains.
Step 3: Put the point values in Score tokens
Open the program’s My Tokens tab and create one token per rule. Adobe’s page on My Tokens lists a Score token type with the instruction “use this token in the change score flow step.” A token looks like {{my.Webinar - Attended}} and holds a value such as +20.
You change a value in one place and every campaign that uses it follows. Tokens are inherited from parent folders and can be overridden lower down, so a regional copy of the program can carry its own values.
Step 4: Build the behavior campaigns
Adobe’s simple scoring quick win shows the pattern. Create a smart campaign, then:
- On the Smart List tab, add a trigger such as Fills Out Form, Visits Web Page, or Clicks Link in Email, and constrain it to the form or page that matters.
- On the Flow tab, add Change Score. Choose Behavior Score and enter the token.
- Add a second Change Score step for Person Score with the same token. Adobe’s guide says to update both identically so Person Score stays aligned.
- On the Schedule tab, activate the campaign.
Then set how often it can fire. Under qualification rules, a trigger campaign defaults to Only once, so repeat actions do not score again. Click Edit Settings on the Schedule tab to switch to Every time or Once every a set number of days, weeks, or months. Once every day suits a pricing page: repeat interest counts, one busy afternoon does not.
Step 5: Build the demographic campaigns
Demographic rules read field values, so use the Lead is Created and Data Value Changes triggers for new and updated people. For people already in the database, run a one-time batch campaign with filters. In the flow, use choices to give different values different points, such as one token for a senior title and another for a manager. Adobe’s guide asks you to keep the choices mutually exclusive.
Step 6: Add negative scores and resets
The library programs subtract for a generic email domain, invalid names, and visits to undesirable web pages. For anyone who should never reach sales, such as employees, partners, or competitors, the =0 form of Change Score resets the score.
Step 7: Build score decay
Marketo has no decay setting. You build it:
- Create a batch smart campaign. In the Smart List, add inactivity filters. Adobe’s example is Not Visited Web Page in the last 30 days. Search for “Not” to find the others.
- Add a filter on Behavior Score with the at least operator, set to the size of the penalty. This stops the score from going below zero.
- In the Flow, add Change Score for Behavior Score and Person Score with a negative token.
- On the Schedule tab, click Schedule Recurrence and pick daily, weekly, or monthly, per Adobe’s steps for a recurring batch campaign.
- Set the qualification rules so a person can run through the flow on each cycle.
The zero floor comes from Adobe’s own scoring webinar, where the presenters advise to “never go below zero.”
Step 8: Set the handoff
Create a campaign that fires on the Score is Changed trigger and add filters for your thresholds. In the flow, use Send Alert with Send To set to Sales Owner, as in Adobe’s alert the sales rep walkthrough. The alert email needs the {{SP_Send_Alert_Info}} token.
Step 9: Test before you rely on it
Adobe’s test: submit a form as a test person, open Database, find the record, and confirm the score moved by the expected amount. Then pull 20 people who became customers and 20 who did not, and check that the two groups score differently.
An example Marketo scoring model
Demo data. A B2B software company sells to operations teams at companies with 200 to 1,000 employees. Values that match Adobe’s program library defaults are marked.
Demographic Score
| Rule | Token | Points |
|---|---|---|
| Job title: director, VP, or C-level in operations | Job Title - High | +15 (default) |
| Industry: core | Industry - High | +10 (default) |
| Number of employees: 200 to 1,000 | Number of Employees | +15 |
| Generic email domain | Generic Email Domain | -2 (default) |
| Competitor email domain | Competitor | =0 on all three fields |
Behavior Score
| Rule | Qualification rule | Points |
|---|---|---|
| Fills out contact or demo form | Every time | +30 (default) |
| Webinar attended | Every time | +20 (default) |
| Fills out content form | Every time | +15 (default) |
| Visits key web pages (pricing, comparison) | Once every day | +5 (default) |
| No activity in 30 days, Behavior Score at least 10 | Recurring monthly | -10 (default) |
Thresholds and plays
| Condition | What it means | Play |
|---|---|---|
| Demographic Score at least 30 and Behavior Score at least 40 | Right buyer, active now | Send Alert to the sales owner and sync to the CRM |
| Demographic Score at least 30, Behavior Score under 40 | Right buyer, quiet | Nurture, and let the threshold campaign catch the change |
| Demographic Score under 30, Behavior Score at least 40 | Active, poor fit | Automated reply or self-serve path |
Take an operations director at a 350-person company in a core industry. She earns 15 for title, 10 for industry, and 15 for size: a Demographic Score of 40. She attends a webinar (+20), views pricing on two different days (+10), and fills out a content form (+15): a Behavior Score of 45, and a Person Score of 85. She crosses both thresholds. A student who does the same three things has the same Behavior Score of 45, a Demographic Score of -2 from a generic email domain, and a Person Score of 43. On Person Score alone, with a threshold of 40, both would have gone to a rep.
A demo request should still route to sales whatever the score.
Putting the score to work
- Smart lists. Score fields support between, greater than, less than, at least, and at most, per Adobe’s field type glossary.
- Routing. The Sync Person to SFDC flow action pushes a person to Salesforce right away and uses Salesforce auto-assignment rules by default.
- Score fields in Salesforce. Marketo’s Salesforce setup has you create a Score field on both the lead and contact objects. For Behavior Score and Demographic Score, create matching fields in Salesforce and make them visible to the sync user.
- Sales Insight. In Marketo Sales Insight, stars and flames turn the score into a priority. Stars show how a person’s score compares with everyone else’s. Flames show how much it has risen lately. Both run from 0 to 3 and recalculate nightly, and by default they read the Lead Score field. Under Admin, Sales Insight, Lead Scoring Settings, you can point each at a different score field.
- Account Score. With Target Account Management, go to Admin, Target Account Management, and edit Scoring Fields to create an Account Score from a person score. It is the sum of that score across the account’s people. Lead-to-account matching attaches new leads to named accounts.
Predictive and AI scoring
The Marketo Engage product documentation used for this guide describes rules-based scoring only. Adobe does document predictive lead and account scoring, but as a service in Real-Time CDP B2B Edition, a separate Adobe product.
Per Adobe, Marketo is currently the required data source, as the source of person-level conversion events. The service uses tree-based machine learning to predict opportunity stage conversion, and needs at least 10 qualified conversion events per prediction goal, with six months of data plus the prediction window as a minimum and two years as the ideal. It produces a score from 0 to 100 that is a relative likelihood, not a probability, and scoring jobs run daily.
Ask your Adobe account team what your contract includes.
Common setup mistakes
| Mistake | What happens | Fix |
|---|---|---|
| Only Person Score, no partial scores | Active poor fits clear the same threshold as good fits | Add Behavior Score and Demographic Score and filter on both |
| Changing one field and not Person Score | The three numbers stop adding up | Two Change Score steps in every flow, same token |
| Point values typed into each flow | Every change means editing many campaigns | Score tokens on the program |
| Leaving the default qualification rule | Repeat actions never score again | Set Every time or Once every |
| No decay campaign | Last year’s activity outranks this week’s | Recurring batch campaign with inactivity filters |
| Decay with no floor | People sink below zero and never surface | Add an at least filter on the score |
For the modeling errors underneath these, see where teams get ICP scoring wrong.
Limits to know, and workarounds
| Limit | Workaround |
|---|---|
| No decay setting. Decay subtracts fixed points per run, not a percentage | Run it on a schedule that matches your sales cycle, with a floor |
| No cap on a score field in the docs used here | Use qualification rules to limit repeats, and at most filters before adding points |
| The field holds a number, not the reason | Keep the rules documented, and use interesting moments in Sales Insight to show reps the activity |
| Account Score is a sum, with up to five score fields | Big accounts score higher by headcount. Read it next to fit data |
Where BoostIdeal fits
This guide is for teams on Marketo Engage, and BoostIdeal does not connect to Marketo. It works with HubSpot, or a CSV you upload: you build the model and set the weights by hand, and each score, shown out of 10, writes back to HubSpot properties with its reasons. It is not the right fit if your data is in Marketo and you do not want to work from CSV exports, or if you want a vendor to hand over a number you never question.
The principles carry over to Marketo. Keep fit and engagement apart, let activity fade, show the reasons on the record, and make sure a person owns the weights.
Frequently asked questions
How does lead scoring work in Marketo?
Smart campaigns watch for a behavior or a field value and run the Change Score flow action, which adds, subtracts, or sets points on a score field. A set of these campaigns, kept in one program with the values in Score tokens, is the scoring model.
What is the difference between behavior score and demographic score in Marketo?
Behavior Score counts what a person does, such as form fills, page visits, and event attendance. Demographic Score counts who they are, such as job title, industry, and company size. Both are custom score fields you create, and Person Score is the combined total.
Does Marketo have score decay?
Not as a built-in setting. You build decay as a recurring batch smart campaign that uses inactivity filters, such as Not Visited Web Page in the last 30 days, and a Change Score step with a negative value.
Can a Marketo score be negative?
Yes. Change Score accepts negative values, and =-5 sets a score to an exact negative number. Adobe’s scoring webinar advises keeping scores at zero or above.
How do I sync Marketo lead score to Salesforce?
Marketo’s Salesforce setup has you create a Score field on the lead and contact objects. For extra score fields, create them in Salesforce and make them visible to the Marketo sync user.
Does Marketo have predictive lead scoring?
The Marketo Engage product docs describe rules-based scoring. Adobe documents predictive lead and account scoring as part of Real-Time CDP B2B Edition, which uses Marketo data, needs at least 10 qualified conversion events per goal, and produces a score from 0 to 100.
The takeaway
Marketo gives you the parts for scoring, which makes the model fully yours, including the upkeep. Start with Adobe’s two library programs, keep Behavior Score and Demographic Score apart, store every value in a token, build decay with a floor, and hand off on both scores together. Then review the thresholds with sales each quarter.