Account Engagement (Pardot) gives every prospect two separate ratings: a score and a grade. Most teams I audit use one of them, misunderstand the other, and route leads on incomplete information because of it.
The distinction is simple once it clicks.
Score = interest. A number that goes up when a prospect does things — visits pages, clicks emails, submits forms. It answers: "How engaged is this person?"
Grade = fit. A letter (A+ through F) based on who the prospect is — industry, company size, job title, location. It answers: "Is this the kind of customer we actually want?"
You need both, because each one alone lies to you.
Picture two prospects.
Prospect A has a score of 210. They've read eleven blog posts, downloaded three guides, and opened every email you've sent. They're also a college student researching marketing automation for a class project.
Prospect B has a score of 15. One pricing-page visit, one email click. They're also the VP of Marketing at a 500-person company in your exact target industry.
If you route on score alone, sales calls the student. If you route on grade alone, sales cold-calls a VP who barely knows you exist. Score-times-fit is the whole game: high score + high grade is a sales-ready lead; high grade + low score is a nurture target; high score + low grade is a content consumer — politely ignore.
Scores change through completion actions, automation rules, and Account Engagement's default scoring model — points for page views, form submissions, email engagement, and so on. The defaults are a starting point, not a strategy: they reward volume of activity rather than quality of intent unless you tune them. I've written a full framework in How to Set Up Lead Scoring in Account Engagement, so I won't repeat it here.
The one thing to remember for this discussion: a score only tells you that someone is active. It says nothing about whether they're worth being active on.
Grading is where Account Engagement gets underused, because the mechanics are slightly odd and most setups never get past the defaults.
Every prospect starts at a D. Grades move up or down from there through profiles — sets of criteria you define that describe your ideal customer. The default profile has criteria like industry, company size, and location; each criterion a prospect matches moves the grade up by a third of a letter, and mismatches can pull it down.
To make grading useful:
1. Define your ideal customer honestly. Not aspirationally — look at your best current customers. What industries, sizes, and titles actually close and stick around?
2. Build profile criteria around data you actually collect. Grading criteria are only as good as the fields feeding them. If you don't ask for industry or company size on any form, no automation can grade on it. This is where progressive profiling on your forms earns its keep.
3. Use multiple profiles if you have distinct segments. Selling to both healthcare systems and SaaS startups? One "ideal customer" profile will average them into uselessness. Assign different profiles by segment.
4. Let automation rules do the matching. Match criteria automatically ("industry contains healthcare → matches Industry criterion"). Manual grading doesn't survive contact with reality.
Here's the practical payoff. Route leads on the combination:
Set this up once — grade thresholds and score thresholds working together in your assignment automation — and the perennial "marketing sends us junk leads" complaint mostly disappears, because "junk" now has a definition and a filter.
Don't obsess over making the numbers perfect. Score and grade are decision tools, not scientific instruments. The goal is a shared language between marketing and sales: "B+ and 100 points means we call within a day." If both teams believe the ratings enough to act on them, the system works. Precision without trust is just decoration.
If your instance currently has scoring nobody believes and grading nobody set up, that's a very normal starting point — and a fixable one. Scoring and grading architecture is core to our Account Engagement work. Let's talk.
Founder of Thompson Technology. Salesforce and Account Engagement consultant for B2B companies.
