Lead generation KPIs for B2B teams: 12 metrics, their formulas, and the data that skews them

What lead generation KPIs are

Lead generation KPIs are the few metrics a team picks to judge whether its lead generation works: how many leads it produces, what each one costs, how many are worth a sales conversation, how fast they are handled and how many become pipeline. A metric is anything you can count; a KPI is a metric you have decided to be judged on, with an owner and a target. Most teams track dozens of lead generation metrics and should report on about a dozen lead generation KPIs.

Every lead generation KPI is a fraction, and its denominator is a list of leads. That is the point most KPI lists skip. A cost per lead calculated on contacts nobody can reach, or a marketing qualified lead rate calculated on records whose scoring fields are empty, no longer measures what its name says. So this guide gives, for each of the 12 KPIs below, four things: the definition, the formula, a worked example, and the data field that breaks it.

The worked examples all use the same illustrative campaign, so you can see how one KPI feeds the next. These are illustration figures, not benchmarks: over one quarter, a B2B team spends $10,000 on lead generation and collects 200 leads.

The 12 lead generation KPIs at a glance

These are the B2B lead generation KPIs that matter once a team has more than one channel; a single-channel team can start with KPIs 1, 3, 4 and 8. Read the last column as a checklist. Before you discuss whether a KPI is good or bad, check that the field it depends on is clean. The definitions of those data metrics (duplication rate, bounce rate, completion rate, decay rate) are in our guide to data quality KPIs; this page uses them, it does not redefine them.

Volume KPIs

1. Leads generated

The number of new, distinct leads collected over the period: form fills, demo requests, event scans, webinar sign-ups, replies to outbound. Formula: count of new leads, after deduplication. Example: 200 leads in the quarter. What breaks it: duplicates. The same person who downloads two guides and registers for a webinar is one lead, not three. If your forms create a new record each time, the count is inflated and every KPI below inherits the error.

2. Leads by source

The share of leads each channel brings. Formula: leads from a channel / total leads. Example: 120 from paid campaigns (60%), 50 from content and webinars (25%), 30 from outbound (15%). What breaks it: the source field. When it is empty, or when an import overwrites it with "list upload", the channel that really produced the lead disappears from the report, and budget moves away from it.

Cost KPIs for lead generation

3. Cost per lead

The average spend to get one lead. Formula: spend / leads generated. Example: $10,000 / 200 = $50 per lead. What breaks it: unreachable contacts. A lead with a mistyped email still counts in the denominator, so cost per lead looks better than what you actually bought. Market data in our 2026 B2B marketing performance report puts the median B2B cost per lead in the low hundreds of dollars, with wide variation by sector; the sector-by-sector detail is in our guide to cost per lead.

4. Cost per reachable lead

The same spend, divided only by the leads you can actually contact: a verified email or a confirmed phone number. Formula: spend / reachable leads. Example: 150 of the 200 leads are reachable, so $10,000 / 150 = about $67. The gap between $50 and $67 is the share of the budget that bought nothing. What breaks it: counting an email as valid because it has an @ in it. "Reachable" has to mean verified.

This is the KPI most lists do not have, and the one our performance report recommends: the benchmark you should care about is your own cost per reachable, qualified lead.

5. Cost per MQL and per SQL

One KPI, two cuts: what you pay for each lead marketing considers qualified, and for each lead sales accepts. Formula: spend / MQLs, spend / SQLs. Example: 60 MQLs and 15 SQLs, so about $167 per MQL and about $667 per SQL. What breaks it: a scoring model run on empty fields. If company size is missing on a third of the records, the score is a guess and so is the cost per MQL.

Quality lead generation KPIs

6. MQL rate

The share of leads that meet your marketing qualification criteria. Formula: MQLs / leads. Example: 60 / 200 = 30%. What breaks it: the completion rate of the fields your scoring reads (job title, company size, industry). A lead with no company size can never reach the threshold, so a low MQL rate can mean empty fields as much as poor leads.

7. Fill rate on scoring fields

The share of leads where every field your scoring uses is filled. Formula: leads with all scoring fields filled / leads. Example: 130 of 200 have job title, company size and industry, so 65%. The other 70 leads were scored blind. We use the same measure in our guide to inbound lead management, where it tells you whether routing can be trusted. Track it next to the MQL rate: when both move together, the scoring is measuring your data more than your leads.

8. MQL-to-SQL conversion rate

The share of marketing qualified leads that sales accepts as sales qualified. Formula: SQLs / MQLs. Example: 15 / 60 = 25%. Our performance report puts the market median somewhere around the low teens in percentage terms, with top-quartile teams converting at roughly double. What breaks it: MQLs that sales cannot reach, or whose job title has changed. A rep who rejects a lead because the number rings at the wrong company is rejecting the data.

To see where your own leads drop between stages, enter your counts in the calculator below. It gives the pass rate of every step and points at the one that loses the most.

Speed KPIs

9. Lead response time

The time between a lead's creation and the first real contact attempt. Formula: median of (first contact time - creation time). Use the median, because one lead left over a weekend skews the average. Example: a median of 9 hours. Our performance report notes that research on lead response has long shown that contacting a new inbound lead within the first few minutes multiplies the odds of qualifying it compared with waiting hours. What breaks it: a first call that reaches a switchboard or a wrong number. The KPI says "contacted in 5 minutes" while nobody was reached.

10. Lead velocity rate

The month-over-month growth of qualified leads, read as a leading indicator of future pipeline. Formula: (qualified leads this month - qualified leads last month) / qualified leads last month. Example: 22 MQLs this month against 18 last month: (22 - 18) / 18 = 22%. The definition is also in our glossary entry on lead velocity rate. What breaks it: duplicates, again. A cleanup that merges duplicates makes the rate drop for one month without anything changing in the market.

Outcome KPIs

11. Lead-to-opportunity rate

The share of leads that turn into a sales opportunity. Formula: opportunities / leads. Example: 10 opportunities from 200 leads = 5%. What breaks it: outdated job titles. Leads collected a year ago include people who have changed jobs, and they drag the rate down without saying why.

This is where lead generation hands over to sales. What happens next (pipeline generated, win rate, coverage, sales velocity) belongs to pipeline metrics, covered in our guide to sales pipeline stages.

12. Customer acquisition cost

What it costs, in sales and marketing combined, to win one new customer. Formula: sales and marketing cost / new customers. Example: $10,000 of lead generation plus $14,000 of sales time, for 3 new customers: $8,000 per customer. See also the glossary entry on customer acquisition cost. What breaks it: spend on records that were never reachable. It sits in the numerator and brings nothing to the denominator.

CLV:CAC, a reading of KPI 12. Divide the lifetime value of a customer by its acquisition cost. If a customer brings $24,000 over its life, the ratio is 3:1. It tells you whether the CAC is affordable, not whether lead generation works.

Outbound lead generation metrics

If part of your leads come from outbound prospecting, four metrics sit before the 12 KPIs above. They are not numbered because they measure the prospecting motion itself. Example, still illustrative: 400 people contacted in the quarter. Every reply, positive or not, also counts among the 30 outbound leads of KPI 2.

• Contact rate: people actually reached / people targeted. Broken by bounced emails and dead numbers.
• Positive reply rate: positive replies / people contacted. 14 / 400 = 3.5%.
• Meeting booked rate: meetings booked / positive replies. 9 / 14 = 64%.
• Meeting held rate: meetings that took place / meetings booked. 7 / 9 = 78%. No-shows count as lost.

How many people to contact to reach a given number of deals is a calculation of its own, worked through in our guide to B2B lead generation.

When a lead generation KPI looks wrong: the bad number and its likely cause

A lead generation KPI that moves the wrong way has two possible explanations: the leads changed, or the data did. Check the data first, because it is faster.

The lead generation KPI dashboard: 3 numbers for leadership, 12 for the team

Leadership does not need twelve lines of lead gen KPIs. Show three, every month: cost per reachable lead, MQL-to-SQL conversion rate and customer acquisition cost. Together they say what a usable lead costs, whether sales wants it, and whether the whole thing pays.

The team needs the twelve, with the data column next to each one. A practical layout: one row per KPI, this period, previous period, the variance, and the related data metric (fill rate, bounce rate, duplicate count) on the same line. When a KPI moves, the data metric next to it tells you in a second whether to open the campaign or the database.

Compare each KPI with your own previous quarter before you compare it with a market figure. A market median mixes teams whose data is clean with teams whose data is not.

4 mistakes that make lead generation metrics lie

1. Treating lead volume as success. 200 leads with 50 unreachable is 150 leads.
2. Measuring a rate on an unverified list. Every rate inherits the share of dead records in its denominator.
3. Scoring on empty fields. A low MQL rate then measures your forms.
4. Not segmenting by source. A blended cost per lead hides the channel that brings unreachable contacts.

Fixing the denominator with Derrick

Most of the data fields in the last column of the table can be fixed before your lead generation KPIs are calculated. Derrick works on your lead list in the web app, with nothing to install, one column per piece of missing information:

• Email Verification checks whether each email can receive mail, which turns "leads" into "reachable leads" for KPI 4. 1 credit per verified email, on the paid plans from the Mini plan.
• Enrich Leads takes each lead's LinkedIn profile URL and returns the current job title and company: part of the scoring fields behind KPIs 6 and 7, and the up-to-date titles behind KPI 11. 1 credit per profile, on the free and paid plans. It needs your LinkedIn account connected through the Derrick Chrome extension, on every surface, MCP and API included.
• Find Duplicates flags the same person entered twice, which keeps KPIs 1 and 10 honest. Free and unlimited, in the Google Sheets sidebar only.

A worked example on the 200 leads. Find Duplicates first, in the Google Sheets sidebar, at no cost. Then Enrich Leads on the 200 profiles: 200 credits, one per profile, so the free plan's 100 monthly credits cover half the list and the Mini plan (9 euros a month, 4,000 credits) covers all of it. Then Email Verification on the 200 emails: at most 200 more credits, a separate step on the paid plans. Two steps that use credits, 400 at most in total, well inside the Mini plan.

The web app is the front door. If your leads already live in a spreadsheet, the same features run from the Google Sheets sidebar. To check one lead from a chat, ask Claude, ChatGPT or any MCP client through the Derrick MCP. To clean every new lead as it enters your CRM, use the REST API. The API and the MCP are available from the Plus plan (47.50 euros a month).