How to Qualify B2B Leads (The Framework Used by High-Conversion Sales Teams)

Most B2B teams say their leads are 'qualified' when in reality, the only filter applied was an industry dropdown in a database. That is not qualification — that is a search query. Real lead qualification means evidence: about whether this company actually fits, actually has the problem you solve, and actually has a reason to act now. This article walks through the framework that separates contact dumps from genuine sales-qualified leads.

Why classic frameworks (BANT, MEDDIC) fall short for outbound

BANT (Budget, Authority, Need, Timing) and MEDDIC (Metrics, Economic buyer, Decision criteria, Decision process, Identify pain, Champion) are excellent — for inbound and account management. They both assume you are already in conversation with the prospect. For cold outbound, you need to qualify before you contact, on data you can verify externally. That requires a different model.

The 3-axis qualification model

Use three axes that can be measured from public data, before any conversation:

Axis 1 — Website quality (40% weight)

For any service that touches digital marketing, web design, SEO, conversion, or development, the prospect website is the primary qualification signal. Measurable on public data via Google PageSpeed, Lighthouse, Wappalyzer, and a URL scrape:

• PageSpeed performance score (mobile and desktop)
• Mobile usability and tap-target issues
• SSL/HTTPS configuration
• Title/meta/structured-data SEO basics
• Tech stack age and modernization signals (e.g., legacy WordPress vs modern Next.js)
• Visible UX/conversion issues (broken forms, dead links, slow images)

A site scoring 30 on PageSpeed mobile, no SSL, missing meta tags, and a broken contact form is a high-quality lead for an agency that fixes those things. A site scoring 95, modern stack, optimized funnel — low quality, because the prospect already solved the problem you sell.

Axis 2 — Business fit (35% weight)

Does this company match your ICP on size, vertical, geography, growth stage, and tech alignment? Measurable from:

• NAICS / industry signals
• Geographic match to your service area
• Employee count and revenue indicators (LinkedIn, Crunchbase, Google Business)
• Founding date and growth stage
• Tech stack alignment (you sell Shopify services → they run Shopify)
• Business activity signals (recent reviews, hiring, content publishing)

Axis 3 — Urgency (25% weight)

Will this prospect act in the next 30–90 days? Urgency signals are the hardest to detect from public data, but the highest-leverage. Look for:

• Active hiring (job posts for marketing/sales/dev roles)
• Visible ad spend (Google Ads, Meta library) — they are spending, the question is on what
• Recent website changes (relaunches, A/B test indicators)
• Funding events or major company news
• Negative urgency: declining reviews, scaling-down signals — sometimes a buying trigger
• Competitive pressure (new entrants in their market)

How to score on a 0–100 scale

Each axis produces a 0–100 sub-score. Combine with weighted average:

Tune the weights to your business. Web design agencies weight WebsiteQuality higher (50%+); B2B SaaS often weights BusinessFit higher (45%+). The exact percentages matter less than having a consistent score across all leads in a batch so you can sort by it.

The acceptance threshold

Set a minimum acceptance threshold and filter every lead below it before delivery to your sales team. Common thresholds:

A typical batch: 30% of researched candidates score below 55 and are dropped. Of the remaining 70%, the top 20–30% are manually reviewed for outbound. Your reps work the top of the score list first, not random rows of the spreadsheet.

Common qualification mistakes

1. Equating "in our database" with "qualified." Industry filter is not qualification.
2. Skipping urgency signals. The fit may be perfect but if there is no buying trigger, conversion lags by months.
3. Manual qualification at scale. Scoring 5,000 leads by hand in a spreadsheet kills your reps. Automate the data collection, manually review only the top decile.
4. Not refreshing scores. PageSpeed scores change; hiring stops; sites get rebuilt. Re-score every 60–90 days for sequence enrollment.
5. Optimizing volume over quality. 100 80-score leads outperforms 10,000 unscored leads — every time.

How Prometheus pre-qualifies before delivery

Every Prometheus lead passes through this exact framework before you receive it. We collect the public data, run the audit, compute the 3-axis score, drop low-fit leads, and ship the rest with the audit data attached. Your reps stop building lists and start working scored opportunities. See the qualified-leads page for the full breakdown, or read about exclusive vs. shared leads to understand why this only works on a private batch.