B2B Lead Generation

What a Lead Is, and Who Gets to Decide

A lead is a person who has taken an action that gives you permission to contact them about a commercial matter. That is the whole definition. Score, grade, stage and temperature are interpretations your organisation adds, and the interpretation is where the arguments start.

Lead is the most abused word in business-to-business marketing. It is used interchangeably for a newsletter subscriber, a badge scan, a whitepaper download, an inbound enquiry with a stated budget, and a name bought from a data broker. Those five have almost nothing in common, and reporting them under one heading is how a marketing department presents a number sales does not believe.

What makes b2b lead generation measurable is three agreed handover points:

• Marketing-qualified lead (MQL) — meets an agreed profile and behaviour threshold, and marketing will pass it on.
• Sales-accepted lead (SAL) — a named rep has agreed to work it. Most companies skip this stage, and it is the only one that produces an honest rejection reason.
• Sales-qualified lead (SQL) — the rep has made contact and confirmed a real need, a plausible timeline, and someone who can authorise the spend.

Write the definitions down, have the sales leader sign them, and date the document. Definitions agreed verbally get relitigated every quarter.

Why the Definitions Carry More Weight in Business Buying

In consumer marketing a lead is usually the buyer. Business to business sales leads almost never are, because the buyer is a group. Gartner's survey of 632 B2B buyers, fielded in August and September 2024, described buying teams ranging from five to sixteen people across as many as four functions, and reported that around 74% of those groups showed unhealthy conflict — conflicting objectives, disagreement on the right course, or being overruled by an external decision-maker.

Two consequences follow. A form fill tells you a group exists behind it, not that the group is ready. And one account will produce several leads over months from different functions; a system that treats each as an unconnected record double-counts demand and fragments the follow-up.

Gartner also describes the journey as non-linear, with buyers looping across six buying jobs — problem identification, solution exploration, requirements building, supplier selection, validation and consensus creation — rather than advancing through them in order. A lead stage therefore describes what you know, not a position on a track. Ignore the familiar claim that buyers are some fixed percentage through their process before contacting a supplier: it is untraceable, and it contradicts the looping model.

Running the Arithmetic Backwards From the Revenue Number

Every credible lead target is derived, not chosen. The derivation runs downward from revenue:

1. Start with the new revenue the business needs from marketing-sourced pipeline, not total revenue.
2. Divide by average deal value to get closed-won deals required.
3. Divide by your historical opportunity-to-close rate to get opportunities required.
4. Divide successively by the SQL-to-opportunity, SAL-to-SQL and MQL-to-SAL rates to reach raw lead volume.
5. Phase the result against your actual cycle length, not the calendar. Leads created in the fourth quarter mostly close in the following year.

Do this once with real conversion rates from your own CRM and one of two things happens. Either the volume is achievable, in which case you have a defensible target and a cost per lead to budget against. Or the volume exceeds the number of qualified companies in your addressable market — the common outcome, and the genuinely useful one. That does not mean marketing needs a bigger budget. It means deal value has to rise, conversion rates have to improve, or the target is arithmetic fiction and should be renegotiated before anyone is measured against it.

Volume, Quality, and What a Gated Download Produces

Sales wins the volume-versus-quality argument, because sales holds the pipeline and pipeline is what the board tracks. A team that hits its lead target while pipeline stays flat has lost the argument regardless of the dashboard.

The pull towards volume is structural rather than dishonest. Lead counts arrive immediately; pipeline takes a quarter or more. Vendors and agencies are usually paid against cost per lead, which rewards cheap volume. Ad platforms optimise towards whichever conversion event you give them, so they find form fills among the cheapest available audiences. Reduce form friction and cost per lead falls while cost per qualified opportunity rises — two metrics moving in opposite directions from one change.

Gated content sits at the centre of this. A download is evidence of interest in a document, not of a project, a budget or authority. What gating produces is a named list with role, company and a stated interest attached, because a registration form is one of the few remaining contexts in which a business buyer volunteers that. What it costs is reach, and the people most likely to abandon a form are often the senior ones. Gate the artefacts a buyer needs when building requirements — the comparison framework, the specification template, the calculator — and leave the thinking that builds your reputation open.

Follow-Up Speed, Routing and the Handover

Response time is the cheapest available improvement to most lead generation programmes, and it consumes no media budget. Someone who has just submitted a form has the problem in mind now, is probably looking at two or three other suppliers in the same session, and will be in a meeting within the hour. The circulating response-time studies come from individual vendor CRM datasets and are not comparable with each other, so ignore the numbers and fix the process.

What breaks in practice:

• Routing that assigns the lead to a territory owner on leave, with no fallback.
• A form that triggers an automated email but no CRM task, so nothing happens until someone reads a report.
• Leads held for a weekly review meeting, turning a same-day response into a five-day one.
• Enrichment and scoring steps that queue a record for hours before a human sees it.
• A first touch that reads like a template, sent from an address nobody monitors for replies.

Put a service level in the handover document: who responds, within what time, on which channel, how many attempts, and what happens when they do not. Then report compliance monthly.

Measuring It Without Misleading Anyone

A lead count is the wrong number to put in front of a board, for a straightforward reason: it is an input the marketing department controls, and it can be raised at will by lowering the definition. Report outcomes and efficiency instead.

• Pipeline value created by source, tagged with the quarter it was created in.
• Sales-accepted rate and rejection reasons by source.
• Cost per sales-accepted lead and cost per opportunity, trended.
• Conversion rate at each defined stage, so a fall can be located rather than argued about.
• Closed-won revenue by cohort — leads created in a period, tracked forward.

Three traps are specific to long cycles. Attribution across quarters: a lead created in March and closed in November is credited to whatever the model favours, and no model handles this well, which is why cohort reporting belongs beside it. Small samples: with a few dozen opportunities a quarter, most movement in a conversion rate is noise, and treating it as a trend produces decisions that get reversed. Self-reported source: a how-did-you-hear-about-us field is a memory test, useful directionally and never as a system of record.

When This Is the Wrong Frame, and What Good Looks Like

Lead generation assumes a market large enough that finding individually interested people is an efficient way to find buyers. Where that fails, so does the model. If your addressable market is a few hundred companies, chasing volume is the wrong instrument — you already know who the buyers are, and the work is getting into those accounts rather than discovering them. If the product sells on a card without a conversation, the useful unit is a signup. And if sales capacity is already consumed by existing pipeline, more leads will age in a queue and damage your standing with the people who filled in the form.

Where it does apply, good looks unglamorous. Definitions agreed and signed. Conversion rates taken from your own data rather than a benchmark report. A small number of sources you understand well, rather than a portfolio you cannot evaluate. A rejection reason on every declined lead. A response time you can prove.

Frequently Asked Questions