The complaint I hear most often from commercial directors: “We need more leads.”

The second most common: “The leads we get are terrible.”

These sound like opposite problems. They’re usually the same one. The company doesn’t have a clear definition of who they’re trying to reach. The data on incoming leads is too thin to qualify them properly. The handoff between marketing and sales is built on hope rather than information.

More leads won’t fix this. Better data will.

Start with the Ideal Customer Profile. Actually start with it.

Every B2B company claims to have an ICP. Most have a vague description of their target market dressed up as one. “Mid-sized manufacturers in Europe with 200+ employees” is a market segment, not an Ideal Customer Profile.

A real ICP is built from your existing customer base, not from aspiration. Which of your current customers are the ones you want more of? Not the biggest logos, not the ones that closed fastest — the ones that renew reliably, expand over time, and generate the most value. What did they have in common? Not just industry and headcount, but what was their situation when they bought? What problem were they solving? The patterns that emerge become your ICP, which becomes the filter for every lead generation, scoring, and qualification decision.

Why most lead data is useless at the point of entry

A lead comes in. You have a name, a company, an email address, maybe a job title. The sales development team is supposed to qualify this person. With what? They don’t know the company’s revenue, employee count, or industry segment. They don’t know if this person is a decision maker or an intern. They don’t know if the company matches the ICP at all.

So they call everyone and waste time on leads that were never going to convert, or cherry-pick based on gut feeling and miss good opportunities that didn’t look obvious. Either way, everything downstream suffers.

Data enrichment closes the gap

Data enrichment adds information to a lead record automatically, at or near the point of entry, using external sources. Company revenue, employee count, industry classification, technology stack, organizational structure. By the time a rep looks at a new lead, they already know whether the company fits the ICP and where the contact sits in the organization. The qualification conversation changes from “tell me about your company” to “I see you’re dealing with X, and companies in your situation typically experience Y.”

The specific tool (Clay, Clearbit, ZoomInfo, Apollo, 6sense) matters less than the principle: lead data should be enriched automatically before it reaches a human.

Lead scoring becomes meaningful only with good data

Most lead scoring setups are either too simple to be useful or too complex to be trusted — points for form fills and email opens, with leads above a threshold routed to sales. The problem is behavioral signals don’t tell you about fit. A person can visit your site ten times and still work at a completely wrong company. Effective scoring combines two dimensions: fit (how well the lead matches the ICP) and behavior (engagement and intent). Neither works alone.

quadrantChart
    title Lead scoring, fit vs behavior
    x-axis Low behavior --> High behavior
    y-axis Low fit --> High fit
    quadrant-1 Priority inbound
    quadrant-2 Outbound target
    quadrant-3 Skip
    quadrant-4 Nurture or skip

Both scores need to decay — a lead active three months ago shouldn’t carry the same score as one that engaged yesterday.

The handoff between marketing and sales

Even with a defined ICP, enriched data, and a scoring model, the handoff from marketing to sales is where most leads fall through. Marketing qualifies by threshold; sales qualifies by confirmed opportunity. Those definitions don’t match, and nobody bridges the gap.

Map the journey through graduated stages instead of a binary hand-off. At minimum, know the prospect’s situation and their pain before routing to sales — if you can’t articulate these, the lead isn’t ready.

The closed loop that most companies never close

In a well-designed system, data flows both ways — leads from marketing to sales, outcomes back. Most companies never close this loop. Marketing measures volume, sales measures revenue, and nobody connects the two to ask which leads from last quarter became retained, expanding customers. When you close it, the ICP sharpens with every cycle and campaign spend shifts toward channels that attract the right accounts.

flowchart LR
    ICP[ICP definition] --> GEN["Lead generation<br/>& enrichment"]
    GEN --> SC["Fit + behavioral<br/>scoring"]
    SC --> MQL["Marketing<br/>qualified"]
    MQL --> SQL["Sales<br/>qualified"]
    SQL --> REV["Revenue &<br/>retention"]
    REV -->|"What converted?<br/>What retained?<br/>What expanded?"| ICP

This requires the CRM to trace a contact from first touch through opportunity, close, onboarding, and renewal. If leads and customers live in disconnected records, the loop can’t close.

What this looks like in practice

A company I worked with came back from trade shows with hundreds of badge scans, uploaded them as leads, assigned a generic score, and routed them all to sales. Conversion to qualified opportunities: under 3%.

We changed three things: enriched every badge scan on import, segmented by fit score (high-fit to inside sales within 48 hours, medium-fit to nurture, low-fit to a thank-you email), and tracked which leads eventually converted to revenue to inform next year’s trade show decisions. Conversion went from under 3% to over 10%.

The lead problem most companies think they have is a data problem. Not enough information at capture, no systematic enrichment, scoring that doesn’t reflect fit, and a handoff that loses context at every step.

Fix the data and the rest follows — the ICP becomes a real filter, scoring reflects actual fit rather than activity, the handoff carries context, and the feedback loop gets smarter over time. Not more leads. Better decisions about the ones you already have.