Measuring B2B Lead Generation: The Data Guide

Contents

• Why counting your leads isn't enough
• Defining a qualified lead: MQL, SQL and the thresholds to set
• The 5 measurement KPIs you absolutely must track
• Calculating cost per lead and ROI by channel
• Multi-touch attribution when the sales cycle is long
• Building your measurement with GA4, Search Console and Ads
• The 6-metric dashboard for an executive
• Where to start based on your current lead volume
• Finally connect your leads to your closed deals with Lysible
• Frequently asked questions

Why counting your leads isn't enough

A lead that never becomes a customer costs money without ever bringing any in. The volume of contacts captured does not measure whether a channel pays off. Only the pass rate through to the sale does. It's the first number I look at when someone shows me a report that talks only about volume.

Take two channels. The first brings in 100 leads a month, the second only 20. Seen that way, the first wins. But if the first converts 2% of its leads into customers and the second 15%, the math flips: 2 customers versus 3. The channel that looked less effective actually brings in more. Judging a channel by its number of leads is like judging an angler by the number of casts rather than the number of fish brought in.

Lead volume: the metric that lies the most

Lead volume is the most deceptive metric in B2B marketing. It climbs easily without revenue moving a single euro. A good call-to-action, a free white paper, a short form: all of it pushes the counters up. None of it guarantees that these contacts have a budget, a real need or the power to decide.

The B2B buying journey explains this fragility. According to Gartner's work on the B2B buying journey, a buying group averages six to ten decision-makers. Each one gathers information on their own. A single form therefore represents only a fraction of a collective decision that is still far from being made.

Consultancy studies confirm this inertia. McKinsey observes that B2B buyers now use around ten channels during a single buying cycle, well before any direct sales conversation.

Connecting every lead to real revenue

A lead is only worth its probability of becoming a customer, multiplied by the average order value. That's the expected value you need to track, not the raw count. To get there, you have to connect three worlds that rarely talk to each other: the site (where the lead declares itself), the CRM (where the sale plays out) and the ad platforms (where you spend).

Many small businesses measure the top of the funnel in Google Analytics 4 and the bottom in a sales spreadsheet, without ever bridging the two. The result: nobody knows which channel generated the deals actually closed. To understand where your qualified traffic comes from before even talking conversion, reading your website traffic statistics gives you a first map of the sources to watch.

Defining a qualified lead: MQL, SQL and the thresholds to set

Before measuring anything, write down in black and white what a qualified lead means at your company. Without shared criteria, marketing and sales will count two different numbers for the same month. It's the number one cause of disagreement between teams.

What is lead generation, concretely?

Lead generation means turning an anonymous visitor into an identified contact who has left a way to be reached in exchange for information or an offer. In B2B, this contact is rarely ready to buy right away: they enter a sales pipeline where they'll be qualified, nurtured with content (nurturing, the process of maturing the prospect), then handed to sales once they're ripe enough.

This approach differs from demand generation, which aims to create a need and awareness upstream, without necessarily capturing a contact right away. One feeds the other: demand creates interest, lead generation turns it into an identified contact.

What is an MQL and an SQL in B2B?

An MQL (Marketing Qualified Lead, a contact marketing considers ripe enough to work) shows real interest, but not yet a buying intent. An SQL (Sales Qualified Lead, a contact sales agrees to take on) is a prospect whose need, budget and timeline are clear enough to justify a meeting.

The distinction isn't just about vocabulary. The reference definitions published by HubSpot on the MQL to SQL handoff stress one point: the MQL belongs to marketing, the SQL to sales. The move from one to the other is the moment of truth for the whole chain. If that handoff goes badly, you generate leads nobody handles.

I advise against complex automated scoring for a low-volume organization. A grid with a few criteria is enough, with the advantage of being readable by everyone.

Who decides the criteria: marketing or sales?

The criteria are decided together, marketing and sales side by side, otherwise the handoff jams in the very first month. Best practice: a short meeting where sales validates what it considers an acceptable SQL. Set SMART goals (specific, measurable, achievable, realistic and time-bound) for each handoff stage, so nobody argues over fuzzy thresholds. Without that written agreement, marketing will send contacts sales refuses, and each will blame the other. I've seen that tug-of-war block measurement for entire quarters, solely for lack of a shared three-line definition.

The 5 measurement KPIs you absolutely must track

Five metrics connected as a cascade are enough to steer B2B lead generation, from the first click through to the closed deal. Taken in isolation, none of them decides anything. As a cascade, they form a complete diagnosis of where money comes in and where it leaks out.

Here's the chain I recommend, each link feeding the next:

Conversion rate by funnel stage

Conversion rate is never measured globally, but stage by stage. It's the only way to pinpoint exactly where your prospects drop off. A B2B funnel has at least three steps: visitor to lead, lead to MQL, MQL to SQL then customer.

If 1,000 visitors produce 30 leads, of which 15 MQL, of which 5 SQL, of which 1 customer, you know where to act. A drop-off between MQL and SQL points to a qualification problem. A drop-off between SQL and customer points to a sales or pricing problem. The same logic applies page by page: if an offer page converts poorly, a structured landing page audit identifies the friction before you touch the traffic.

Cost per lead calculated by channel

Cost per lead is only calculated at the level of each source, never as an average. Dividing your total budget by your total number of leads gives you a number with no decision value whatsoever. An expensive but highly profitable channel disappears into the mass, and you end up cutting the wrong budget.

The summaries from CXL on conversion rates remind us that the same traffic can convert anywhere from one to three times as well depending on how clear the stated promise is.

MQL to SQL to customer rate

The double rate MQL to SQL then SQL to customer is the most telling metric in the whole chain. It measures the real quality of leads, not their quantity. A channel showing a 10% MQL to customer rate crushes a channel at 2%, even if it produces five times less volume. Quality beats volume. Every time.

Calculating cost per lead and ROI by channel

Cost per lead is calculated by dividing a channel's spend by the number of leads it generated over the same period. ROI, for its part, compares the revenue closed to that spend. The real difficulty isn't the formula: it's attributing every euro and every customer to the right channel.

How do you calculate the ROI of lead generation?

The formula: (revenue generated by the channel minus channel cost) divided by channel cost, multiplied by 100. A channel that costs €2,000 and brings in €10,000 in closed revenue shows a 400% ROI. You still need to know which revenue comes from which channel, which means connecting the CRM to your traffic sources.

Watch out for the timing trap. In B2B, revenue closed in May often comes from leads captured in February or March. Comparing May's spend to May's revenue distorts the whole calculation. You have to think in cohorts: connect the revenue to the leads that actually produced it, on their entry date into the pipeline.

Why think by channel and never as an average

Thinking in global averages systematically hides the channel that's most expensive to enter but most profitable to exit. SEO is a perfect illustration of this flaw. Its apparent cost per lead is often high in the first year, except that its medium-term profitability stays the best once the effort has paid off. Cutting SEO on the strength of a global average is the classic mistake.

These ranges vary widely depending on the industry and the order value. They serve as a benchmark, not an absolute truth. Analyses from Forrester on B2B acquisition confirm that the cheapest channels to enter are almost never the most profitable to exit. The benchmark on customer acquisition cost by channel details these gaps across several models, from e-commerce to B2B SaaS, and confirms that SEO remains the channel with the lowest CAC once amortized.

Multi-touch attribution when the sales cycle is long

When the sales cycle runs longer than a few weeks, the last-click model systematically underrates SEO and content. It gives all the credit to the final interaction, erasing what built the decision beforehand. In B2B, that's a measurement error with heavy budget consequences.

Why last click deceives in B2B

Last click gives 100% of the credit to the channel that preceded the conversion, erasing all earlier touchpoints. Take a prospect who discovers your brand through a blog article, comes back a month later via a branded search, then clicks on an ad before filling out the form. With last click, Google Ads scoops up all the credit. SEO, which started the relationship, stays invisible in your dashboard.

Data from Gartner on the complexity of the B2B buying journey shows that buyers spend most of their time doing independent research. That work precedes any sales contact.

The rule of 7 and its impact on measurement

The rule of 7 holds that a prospect is exposed to a brand an average of seven times before taking action. Whether or not you take that figure literally, the underlying idea holds: a single exposure rarely triggers a B2B decision. Your measurement must therefore value repetition, not just the final click.

On this point, I take the opposite view from the usual recommendation. Google pushes toward data-driven attribution, which splits the credit algorithmically. For an organization generating a few dozen leads a month, this model becomes unstable. It doesn't have enough data to learn, and its results vary from one month to the next with no readable logic. The Google Analytics documentation on attribution models acknowledges this itself by conditioning data-driven on sufficient volume.

At low volume, a simplified multi-touch model (linear, for example, which shares the credit equally across all touchpoints) is more reliable and, above all, understandable. A model you don't understand is a model you won't use.

Building your measurement with GA4, Search Console and Ads

A trio of entirely free tools, Google Analytics 4, Search Console and Google Ads, is enough to seriously measure your lead generation before any paid software purchase. Each one covers an angle the other two can't see.

Configuring lead conversions in GA4

GA4 measures what happens on your site. Each completed form becomes a conversion event that you declare once and for all. The steps are documented in the Google Analytics help center on conversion events. You create an event triggered by the form submission, then mark it as a key conversion.

To set these triggers without touching your site's code, going through Google Tag Manager is the cleanest route. If you're starting out, the step-by-step Google Tag Manager guide for beginners explains how to install the container and create your first triggers without a developer.

If you reject the Google ecosystem for privacy reasons, an alternative like Matomo offers comparable conversion tracking, with the data hosted on your own servers.

Cross-referencing Search Console queries with leads

Search Console tells you which Google searches bring in the visitors who then become leads, information GA4 alone doesn't give you. By cross-referencing the pages that convert in GA4 with the queries feeding them in Search Console, you identify the keywords that truly generate contacts, not just traffic.

several months The length of a complex B2B buying cycle is often counted in months, not days. This means you have to connect each lead to its entry date to calculate a fair ROI, rather than comparing spend and revenue for the same month. Source: Gartner, B2B buying journey

This cross-referencing between Search Console and GA4 is done by hand in about thirty minutes a month, or continuously if you automate the tracking of these sources. The data is free. It's the time to cross-reference it that costs.

Tracking the real cost per form in Google Ads

Google Ads gives you the real cost paid for each lead coming from advertising, provided you import conversions from GA4 or track them natively. That's what makes your advertising cost per lead exact rather than estimated. For anything beyond basic management, bringing in a Google Ads expert is justified above a certain budget threshold, but the starting measurement stays within your reach.

The 6-metric dashboard for an executive

A useful dashboard fits in six lines at most, each translated into a business decision, and gets re-read every week in under five minutes. Beyond six metrics, attention scatters and nobody looks at anything anymore.

Which metrics to show and which to leave out

Show what triggers a decision, leave out what merely describes. Page views, time spent on the site, number of sessions: these are descriptive metrics, useful for diagnosis, useless on a steering dashboard. What must stay permanently visible is the cascade that leads to the euro.

These targets are starting benchmarks, not universal norms: your industry and your average order value shift them. The key is to set your own, ideally as SMART goals, and not move them for a quarter.

How often should you re-read your numbers?

Re-read volumes and costs every week, conversion rates and ROI every month. Never more often. Conversion rates move slowly in B2B. Looking at them every day generates noise and needless panic. This is exactly the calculation we wanted to make readable in Lysible, because no executive redoes this cross-referencing by hand every week without eventually giving up.

Where to start based on your current lead volume

Your starting point depends entirely on your lead volume. At 5 a month, you make the counting reliable. At 500, you industrialize attribution. A single method for every stage doesn't hold up.

Below 10 leads a month, forget attribution models and automated scoring. Simply set up conversion tracking in GA4, note the source of each closed deal by hand in a spreadsheet, and check that marketing and sales talk about the same definitions. At this stage, your real problem isn't fine measurement, it's volume: reading how to increase your website traffic through SEO and referrals is more urgent than polishing an attribution model.

Between 20 and 100 leads a month, structure the full cascade: the five KPIs, cost per lead by channel, a simplified multi-touch model. This is the stage where a six-line dashboard really changes your budget trade-offs, because you have enough data to see the gaps between channels.

Beyond a few hundred monthly leads, data-driven attribution becomes usable, a CRM like HubSpot is justified for scoring, and automatic integration between site, CRM and ad platforms becomes the real productivity lever. At this volume, the time spent cross-referencing data by hand becomes your main hidden cost.

Finally connect your leads to your closed deals with Lysible

At this stage, you know what to measure: the cascade from click to customer, cost per lead by channel, a multi-touch model suited to your volume. The real obstacle left isn't the data, it's already in GA4, Search Console, Ads and your CRM. It's the time to pull it together each week without making mistakes. Many executives abandon tracking after two months, not for lack of willpower, but because manual cross-referencing is exhausting. Lysible centralizes these sources so your lead cascade stays readable without spending your evenings on it.