What Is a Lead? A Definition B2B Teams Actually Use

Ask five people in the same sales meeting what a "lead" is and you will get five confident, slightly different answers, and all five will act surprised that the other four disagreed. It is one of the most argued-about four-letter words in B2B, mostly because everyone assumes everyone else means the same thing by it, right up until a Monday pipeline review proves otherwise.

A lead is a contact who has given you a way to reach them and shown some reason to think they might buy. Not every name in your CRM is a lead, and not every lead is worth a call today. The word gets used loosely enough that sales and marketing routinely argue about the same list while both technically agree on the definition on the whiteboard, which is exactly the kind of polite standoff that quietly wastes a quarter.

This guide draws the actual lines: what separates a lead from a plain contact, how a lead becomes an MQL and then an SQL, the four lead types you will see in a real pipeline, how many you actually need before you start chasing more, and why the argument between sales and marketing keeps happening even at companies that wrote the definition down.

What Is a Lead in B2B Sales?

A lead is a person or company that has both a way to be contacted and a signal suggesting they might need what you sell. Both halves matter. An email address with no signal is a contact, not a lead. A signal with no way to reach the person, a company that fired a job posting matching your ICP but no named contact yet, is a target, not yet a lead.

A lead is not the same as a contact

Every lead is a contact, but most contacts are not leads. A contact is anyone in your CRM: a past customer, a conference badge scan, a colleague who signed up to see the product once. A lead adds a reason on top: they visited a pricing page, replied to an outbound message, or their company started hiring for a role your product supports. The reason is what makes the record worth a rep's time, and a CRM with no way to separate the two just trains reps to ignore the field entirely.

Where a lead sits before it becomes an MQL or SQL

HubSpot's own lifecycle definitions put a lead ahead of both qualification stages. A marketing qualified lead, MQL, is "a contact who has shown interest in a product or service but isn't ready to buy yet". A sales qualified lead, SQL, has been "vetted and deemed ready for a direct conversation with sales." A plain lead sits before both: interesting enough to log, not yet qualified either way. Our guide to lead scoring covers exactly how that qualification step gets decided rather than guessed at.

The Types of Leads You Will Actually See

Most pipelines run four practical categories, and the table below is the plain-language version rather than the jargon. The evidence column is what should decide which bucket a record sits in, never a gut feel.

Cold, warm and hot are a temperature, not a definition

"Warm lead" describes how much friction is left before a conversation, not what qualifies the record. A cold lead sourced from a precise buying signal, a company that just posted a role your product replaces, can be a better use of an SDR's morning than a warm lead who opened three emails and clicked nothing. Temperature is a scheduling hint. Fit is the actual filter.

Inbound and outbound leads are qualified the same way, sourced differently

An inbound lead arrives because the reader found you, through a search, a referral, or content. An outbound lead arrives because you found them first, through a cold email, a call, or a buying signal you noticed before they went looking. Neither is automatically higher quality; our guide to inbound versus outbound sales walks through when each motion actually wins.

Put those four types on a whiteboard and they look tidy. In a real CRM they blur into each other constantly, a warm lead goes cold the moment the champion changes jobs, a cold lead turns red-hot the day a competitor's contract expires. The table is a starting point for the argument, not a replacement for looking at the actual account.

How a Lead Gets Qualified

Qualification is the step that turns a lead into an MQL or an SQL, and it should run on a written rule, not a rep's instinct. The two inputs that decide it are fit, does this account match who you sell to, and intent, is there a real reason to think they need it now. Our guide to lead qualification takes that decision apart in full, including the two dimensions that only appear once someone has replied.

The signals that separate a lead from noise

Firmographic fit (company size, industry, tech stack), role fit (does this person actually buy or use the product), and behavioural signal (a pricing page visit, a reply, a job posting, a competitor mention) combine into a score. A high firmographic fit with zero behavioural signal is a target list, not a qualified lead yet. A strong behavioural signal on a bad firmographic fit is noise dressed up as urgency. Both have to clear a bar.

Why sales and marketing disagree about the same lead

The disagreement almost never comes from a missing definition. It comes from an unwritten one, and an unwritten definition is infinitely flexible, which is exactly why it survives every attempt to resolve it in a meeting. Marketing counts a scoring threshold crossed. Sales counts a conversation that went somewhere. Both are defensible and both can be true of the same record at the same time while the two teams argue past each other like two people describing the same elephant from opposite ends. The fix, as SDR teams that run this well describe it, is writing the qualifying bar down jointly and reviewing rejected leads together on a schedule, not relitigating the definition every time a deal stalls.

How Many Leads Do You Actually Need?

"We need more leads" is the sales-team equivalent of "we need more coffee": always true, rarely the actual bottleneck, and a surprisingly effective way to avoid a harder conversation about quality. The honest answer works backward from quota, not forward from however many leads marketing can generate.

Start with the revenue target, divide by average deal size to get the number of deals needed, divide that by your close rate to get the number of qualified opportunities needed, and divide again by the rate at which a lead becomes an opportunity. The number that falls out the other end is the real lead target, and it is almost always smaller and less flattering than the number a team defaults to chasing on a slide.

Why more leads is not the same as more revenue

Salesforce's own 2026 State of Sales Report found that sales reps spend 60% of their time on tasks other than selling, which means every additional lead handed to a rep competes with a fixed, already-scarce pool of selling hours. Doubling the raw lead count without raising the qualification bar just moves that scarcity earlier in the funnel: reps spend more of their selling time triaging instead of closing, and the extra leads do not translate into extra revenue.

A smaller number of well-qualified leads outperforms a bigger pile

A team that halves its lead volume but raises the qualifying bar usually sees close rate rise enough to offset the smaller number, because reps spend the freed-up time on accounts more likely to close rather than working through a longer queue at the same average quality. The volume question and the qualification question are the same question asked from two directions, which is exactly why a written scoring model matters more than a bigger top of funnel.

Where Lead Definitions Break Down

Three failure modes account for almost every broken lead process, and none of them require better software to fix.

Counting activity instead of evidence

A newsletter open is activity. A pricing page visit followed by a reply to a follow-up is evidence. A scoring model that weighs both the same way manufactures MQLs that go nowhere, and a sales team that gets handed enough of those stops trusting the label entirely, at which point the whole qualification step becomes theatre.

Letting the definition live only in someone's head

A lead definition that exists as a shared understanding rather than a written document survives exactly as long as the person who remembers it stays in the role. The fix costs one working session: write the criteria down, put real thresholds on them, and revisit the document on a schedule rather than after a disagreement.

Treating every source the same regardless of evidence quality

A lead sourced from a demo request and a lead sourced from a purchased list are not the same kind of evidence, even if both technically satisfy "gave us an email address." A definition that does not account for how a lead arrived ends up scoring a cold, purchased contact the same as someone who read three articles and asked a specific question, which quietly rewards volume over relevance across the whole pipeline.

How Leadalf Defines a Lead

Leadalf treats a lead as a signal plus a person, not a name plus a guess. It watches LinkedIn, Reddit, X, the Meta Ads Library and job postings for buying-intent activity, then scores every account that fires a signal against the ICP you define, so a record only becomes a lead once there is a real, dated reason attached to it. That is the same distinction the guide to lead scoring covers in depth: a lead earns the label through evidence, not through sitting in a database long enough.

Frequently Asked Questions