PQLs vs MQLs vs SQLs: How to Build a Sane Lead Scoring Model

The Decay of the Scoring Waterfall

Most lead scoring models are relics of 2014. They rely on "Marketing Qualified Leads" (MQLs) triggered by arbitrary whitepaper downloads or webinar attendance. In 2024, an MQL is often just a noise generator for the BDR team.

The traditional waterfall—MQL to SQL to Close—is fundamentally broken because it measures interest rather than intent. If a Director of Engineering at a Series C startup downloads your "Guide to K8s Security," they aren't necessarily looking for a vendor. They might just be writing their own internal documentation.

A sane model treats MQLs, PQLs, and SQLs as distinct data points with different shelf lives, not a linear progression.

PQLs: The Gold Standard for PLG

Production Qualified Leads (PQLs) only exist if you have a self-serve tier or a free trial. They are the most accurate predictor of revenue because they reflect usage, not promises.

Stop scoring PQLs based on "logging in." Logging in is a baseline, not a signal. A real PQL is triggered when a user hits a high-gravity feature.

• SaaS Example: For a project management tool, a PQL isn't "created 5 tasks." It’s "invited 3 team members and integrated Slack."
• DevTools Example: A PQL for a monitoring tool is "set up 2 custom alerts and passed 1GB of data."

If you’re a VP of Growth at a Series B, your PQL definition should be audited quarterly. If your PQL-to-Close rate is under 15%, your threshold is too low. You are sending "tourists" to your sales team and burning your AEs out.

MQLs: The Survival Guide for Enterprise Sales

If you don't have a product-led motion, you’re stuck with MQLs. The mistake most RevOps teams make is weighting "firmographics" too heavily. Yes, they need to be the right company size, but a "clean" lead from a Fortune 500 company is still a waste of time if they have zero intent.

To make MQLs sane, move to a "decaying" point system.

• Attended Demo: +50 points.
• Downloaded Case Study: +10 points.
• Inactivity for 14 days: -40 points.

An MQL should represent a window of opportunity, not a permanent status. If the Director of RevOps at a prospect account stops engaging, that lead should expire. Keeping them in the "MQL" bucket for six months ruins your pipeline velocity metrics.

The SQL: Where the Rubber Meets the Quota

A Sales Qualified Lead (SQL) is a lead that has been vetted by a human—usually a BDR or SDR—and has a confirmed "Need, Authority, and Timeline."

The friction point here is the handoff. Sales leaders often complain that MQLs are "garbage," while Marketing claims Sales isn't following up. The fix is a strict Service Level Agreement (SLA) on the SQL definition.

An SQL is not just someone who agreed to a meeting. It is someone who fits the ICP and has a documented pain point that your software solves. If the BDR can’t fill out three specific custom fields in the CRM—Current Tool, Primary Pain, and Expected Launch Date—it shouldn’t be an SQL.

Why "Discovery" is the Missing Metric

The biggest gap in the PQL/MQL/SQL framework is the lack of objective, unfiltered feedback from people who aren't in your funnel yet. You are scoring people based on how they interact with your assets, which is a biased data set.

Sophisticated operators are now looking for "Market Qualified" signals before a lead even hits their site. This involves understanding how your category is being researched in the wild. For example, using a platform like BuyerSignal allows you to pay verified professionals for structured research conversations. This gives you a baseline of what a "Ready to Buy" profile actually looks like before you waste points on a lead who just likes your blog.

Designing the Score: The 3-Tier Audit

Don't build your model in a vacuum. Sit down with a Senior AE and a Growth Marketer. Review the last 50 closed-won deals and trace them back to their first touchpoint.

1. The Identity Score (40%): Does the person have the power to buy? (Title, Company Size, Revenue).
2. The Intent Score (40%): Have they shown they are looking for a solution? (Pricing page visits, BuyerSignal research sessions, competitor comparison searches).
3. The Engagement Score (20%): Are they interacting with your specific brand? (Email opens, webinar attendance).

If a lead has a high Identity score but a low Intent score, they are a target for long-term nurturing, not a sales call. If they have high Intent but low Identity (e.g., a junior analyst), they are a "champion" to be enabled, not the person you pitch.

What Most People Get Wrong: The "False Positive"

Most scoring models ignore the "Negative Signal." If a prospect visits your "Careers" page three times in an hour, they aren't a high-intent buyer; they are a job seeker. Your model should include negative triggers that immediately disqualify or heavily penalize a lead score.

• Free email domain (Gmail/Yahoo): -100 points.
• Visits /careers: -50 points.
• Competitor IP address: Disqualify.

A sane model is as much about who you don't talk to as who you do. Sales time is your most expensive resource. Protecting it is the primary goal of lead scoring.

If your current scoring model feels like a guessing game, use BuyerSignal to connect with verified buyers and understand the real triggers that move a professional from "just looking" to "ready to buy." Use these insights to refine your PQL and MQL definitions based on actual market behavior.