B2B Lead Scoring with Adobe Marketo Engage and AEP: A Guide

Sales says marketing's leads are no good; marketing says sales never follows up. Both usually have a point, and the real problem sits between them: nobody agreed on what a sales-ready lead is or which data decides it. A lead scoring model fixes exactly that when it's designed with sales, built on reliable data and reviewed against business results. When it isn't, it becomes a number that keeps climbing and nobody looks at.

At WolfSellers we implement Adobe Marketo Engage and Adobe Experience Platform (AEP) for B2B companies across North America and Mexico. We cover what Marketo Engage is and when it makes sense in our Adobe Marketo Engage guide. This article comes one step before any campaign: how to design, implement and govern a lead and account scoring model that sales actually uses. Everything we say about Adobe features comes from Adobe's Experience League documentation, reviewed in October 2026.

---

What is B2B lead scoring

Lead scoring is a model that assigns points to each person in your database based on how closely they match your ideal customer and how much interest they've shown, so you can decide when they go to sales and in what order they're worked. In B2B it's paired with account scoring, which rates the whole company, because a buying committee makes the decision, not one person.

A healthy model keeps two dimensions apart, and Adobe Marketo Engage's own glossary separates them: the demographic score measures fit and the behavior score measures interest.

Two more pieces keep the model from inflating:

• Negative scoring subtracts points for signals that point away from a sale. Adobe's templates subtract for generic email domains, invalid names and visits to "undesirable" pages.
• Score decay subtracts points over time when there's no activity, so the number reflects today's interest. Adobe's behavior template includes a Score Decay folder with a "Decrease Score - No Activity" campaign.

People or accounts? Both. In an enterprise deal, the IT director may have a low score while two people on their team download all your content: the person score decides who to contact, and the account score decides which accounts to prioritize.

How to implement lead scoring in Adobe Marketo Engage

Score fields, Change Score and smart campaigns

In Marketo Engage, scores live in a person's score fields, and you can create several. Adobe's scoring templates call for two custom fields: Demographic Score and Behavior Score.

Points change through the Change Score flow step inside a smart campaign: +5 adds, -5 subtracts, =5 sets an exact value and =-5 sets an exact negative value. The = option is useful for resets, for example when sales rejects someone who goes back into nurture.

Trigger campaigns add points the moment someone fills out a form or visits a page. Batch campaigns run against a group at a set time and can recur on a schedule: they're what applies score decay.

Adobe's program library includes two starter templates. OP-Scoring-Demographic scores job title, industry, annual revenue, number of employees and lead source. OP-Scoring-Behavior scores forms, webinars, trade shows, key pages, downloads and email clicks, and includes decay. Their values are program tokens, so you tune the model in one place.

Lifecycle: MQL, SAL and SQL in the Revenue Cycle Modeler

• MQL (Marketing Qualified Lead): in the words of Adobe's glossary, the person has exhibited the behavior and characteristics that meet your criteria to be passed to sales.
• SAL (Sales Accepted Lead): sales reviewed the MQL and agreed to follow up.
• SQL (Sales Qualified Lead): sales confirmed a real opportunity and logged it in the CRM.

The score takes someone to MQL; moving to SAL and SQL is a call made by a person in sales. Marketo Engage models the funnel with the Revenue Cycle Modeler: a new model starts with six stages, smart campaigns define the transitions between stages, and there are three stage types. Inventory stages track inflow, outflow and balance; gate stages split paths, such as named accounts; and SLA stages enforce a maximum response time. Adobe's example is a 72-hour SLA to contact someone who asked to be contacted, with infractions tracked.

Syncing with Salesforce or Microsoft Dynamics

A score is only worth what sales sees in the CRM:

Because leads and contacts sync both ways, scores can show up in the CRM; opportunities flow into Marketo Engage and are the raw material for calibration. With Marketo Sales Insight, reps see flames (how much a score has risen recently) and stars (score relative to the rest of the database) inside the CRM, recalculated nightly. Adobe warns that with too few scoring campaigns, those indicators aren't useful.

Account scoring with Target Account Management

Target Account Management (TAM) is an add-on to Marketo Engage for account-based marketing. Its Account Score sums person scores within the same account and can use up to five fields. If person scores are wrong, the account score just adds up the errors.

What Adobe Experience Platform adds to lead scoring

Real-Time CDP B2B Edition and predictive scoring

Real-Time CDP B2B Edition combines data from multiple sources into person and account profiles, with B2B XDM schemas that closely match CRM objects and a dedicated Marketo Engage connector and destination. We explain it in our Adobe Real-Time CDP guide.

Its predictive lead and account scoring service learns from real conversions —a lead conversion, or an opportunity moving to a stage such as Contract— and produces person and account scores. Here's what the documentation, last updated June 18, 2026, says:

The model learns from what your CRM records: if opportunities aren't logged consistently, it learns noise. It also doesn't replace your rules-based model. We recommend running both in parallel for a quarter and checking which one better anticipates opportunities before letting the predictive score shape your MQL definition.

Buying groups in Adobe Journey Optimizer B2B Edition

Adobe Journey Optimizer B2B Edition is built on AEP, including Real-Time CDP B2B, and scores the buying group: the people at an account involved in a purchase decision, each with a role. Adobe talks about marketing-qualified buying groups: the group is what gets qualified.

With the 30-day window, activities expire and the score goes down on its own, which the documentation calls score decay. Activity weights are set through your own weighting models, and only one can be active at a time. We describe how we implement it in our AJO B2B Edition article.

AI features for scoring: what exists today

Scoring Studio doesn't train a model on your conversions: the assistant proposes dimensions and signals with points based on Marketo Engage fields and activities, and you decide what gets published. Confirm with Adobe which of these your contract includes.

How to design a lead scoring model sales will actually use

1. Run a workshop with sales before touching the tool

Bring in sales leadership, two or three reps and marketing operations. Review recent won and lost deals together: which roles were involved, what they did before the first call and which signals turned out to be misleading. You walk away with an ideal customer profile, buying signals and disqualifying signals.

2. Define the stages and the service-level agreement (SLA)

The SLA runs both ways: marketing delivers leads that meet the definition, and sales works them within an agreed time and records the outcome. An illustrative example:

Rejection reasons are the most useful input for recalibration: "no budget" adjusts the threshold; "wrong person" adjusts the fit score.

3. A scoring table: an illustrative example

These values aren't a recommendation for your company: they're a starting point for the workshop.

Adobe's templates start at a similar scale: +30 for a contact form, +20 for attending a webinar, +2 for an email click, −10 for inactivity and −2 for a generic email domain.

4. Thresholds: score fit and engagement separately

With a single combined score, a very active student and a barely active director can tie. Define the MQL with two thresholds:

5. Calibrate against won and lost opportunities

1. Export opportunities from the last 12 to 24 months, with their contacts, from the CRM.
2. Rebuild what score those people had before their first conversation with sales.
3. Compare conversion by score band: if the top band doesn't convert better than the middle one, the model isn't discriminating.
4. Before launch, run the model in shadow mode for a few weeks: it scores, but leads are still routed as they are today.

After launch, review it every quarter and whenever your offering or target market changes.

Data and governance: what makes a score trustworthy

• Duplicates. Marketo Engage dedupes new people as they come in, but not against the Salesforce or Microsoft Dynamics sync or manual entries. A duplicate splits the score across two records, and merging them can't be undone.
• Standardized fields. If job title and industry arrive as free text, no rule reads them well: use picklists in forms and in the CRM.
• Identity. In AEP, a person's identity and their link to the account drive everything else; we cover it in data governance in Adobe Experience Platform.

Consent and Mexico's LFPDPPP

If you market to people in Mexico or run marketing operations from there, review your scoring design against Mexico's federal personal data protection law (Ley Federal de Protección de Datos Personales en Posesión de los Particulares, or LFPDPPP). It defines personal data as any information about an identified or identifiable person (Article 2); in B2B that includes name, email, job title and what the person does on your site. What the current text says (last amended in the Official Gazette on November 14, 2025):

• Article 7: as a general rule, implied consent is valid when you've made the privacy notice available and the person didn't object, unless another provision requires express consent. Financial or asset data does require express consent, and sensitive data requires express, written consent (Article 8).
• Articles 11 and 15: the privacy notice must list the purposes of processing, and using data for a different purpose requires new consent. If you score prospects, that purpose belongs in the notice.
• Article 17: if you didn't collect the data directly, for example when enriching with third-party data, you must disclose the change in the privacy notice.
• Article 26, section II: a person can object to automated processing that, without human intervention, evaluates personal aspects or predicts their behavior and produces unwanted legal effects or significantly affects their interests. In the design we recommend, the score prioritizes and a person in sales decides.

Other jurisdictions have their own rules; this isn't legal advice, so validate it with counsel. More in our guide to data protection and the LFPDPPP.

Common lead scoring mistakes

1. Designing it without sales: if sales didn't help define the MQL, they won't trust it.
2. A single score that mixes fit and engagement.
3. Scores that only go up, with no negative scoring or decay.
4. Too many rules that nobody can explain.
5. MQLs without an SLA or rejection reasons, which leaves the model without feedback.
6. A model nobody reviews after launch.
7. Turning on predictive scoring without well-logged opportunities in the CRM.
8. Scoring accounts without scoring people well first.

What to measure to know the model works

To measure the journey across channels on the same AEP data, we use Adobe Customer Journey Analytics.

How we do it at WolfSellers

We start with an assessment of your Marketo Engage instance: score fields and campaigns, duplicates, CRM sync and how your sales team works today. We facilitate the workshop and deliver the MQL, SAL and SQL definitions, the SLA and the rejection reasons. Then we configure the model —fields, tokenized programs, decay, Revenue Cycle Modeler and the CRM view—, run it in shadow mode, launch it with a dashboard and review it every quarter. If you already have AEP, we add account profiles, predictive scoring in parallel and buying groups.

If you're still choosing a platform, start with our guide to choosing a marketing automation platform. Our Adobe Marketo Engage, Adobe Experience Platform, Adobe Real-Time CDP and Adobe Journey Optimizer services cover everything from model design to operations. If you'd like to review your case, contact us.

---

Frequently asked questions about B2B lead scoring

What is an MQL?

An MQL (Marketing Qualified Lead) is a person who has shown the behavior and characteristics that meet your criteria to be passed to sales, as Adobe Marketo Engage's glossary defines it. In a good model it's defined by a fit threshold and an engagement threshold agreed with sales, and it comes with a follow-up SLA.

What's the difference between a fit score and an engagement score?

The fit score measures how closely the person and their company match your ideal customer. The engagement score measures their current interest. Keep them in separate fields so a poor-fit, very active person doesn't tie with an ideal-fit, barely active one.

Does Marketo Engage have predictive lead scoring?

Marketo Engage's native scoring is rules-based, using smart campaigns and the Change Score flow step. Predictive scoring lives in Real-Time CDP B2B Edition, which uses Marketo Engage as its conversion source. Marketo Engage's Predictive Audiences estimates event registration and attendance likelihood; it isn't lead scoring. Account Profiling, formerly AccountAI, is no longer available to new users as of 2025.

How often should a lead scoring model be recalibrated?

We recommend a quarterly review with sales, using conversion by score band and rejection reasons, plus an extra review whenever your offering, target market or lead sources change.

What is score decay?

It's subtracting points when a person stops showing activity, so the score reflects current interest. In Marketo Engage it's done with recurring batch campaigns; in Journey Optimizer B2B Edition buying groups it's built in, because the engagement score only counts the last 30 days.

Do I need Adobe Experience Platform to do lead scoring?

No. A solid rules-based model can be built with Marketo Engage and your CRM. AEP makes sense when you need account profiles from multiple sources, predictive scoring or buying group qualification.

Does lead scoring require express consent under Mexico's LFPDPPP?

As a general rule, no: Article 7 treats implied consent as valid when you've made the privacy notice available and the person didn't object, unless another provision requires express consent. What you do need is for prospect scoring to be among the purposes listed in the notice. Validate it with your legal team.

What is a buying group and how is it scored?

It's the set of people at an account involved in a purchase decision, each with a role. In Adobe Journey Optimizer B2B Edition it's scored with the completeness score, which measures whether the required roles are filled, and the engagement score, which measures members' activity over the last 30 days.

Related services

If this topic is relevant to your business, these services from WolfSellers can help you implement it:

• Adobe Experience Platform
• eCommerce Marketing
• AI & Automation