Demand generation vs lead generation, the metrics each uses and when gating makes sense

Lead generation optimizes for contacts, and demand generation optimizes for pipeline and buying intent. One counts the names you captured. The other counts qualified opportunities from buyers who already want to talk to you. The gap shows in the numbers: Refine Labs puts the benchmark for declared intent form fill to qualified opportunity at 30 to 40%, and 78% of business-to-business customers buy from the vendor that responds first.

What is the difference between demand generation and lead generation?

Lead generation optimizes for contacts. Demand generation optimizes for pipeline and buying intent. One counts the names you captured. The other counts the qualified opportunities that come from buyers who already want to talk to you.

ZoomInfo describes lead generation as the motion that "captures contact information from in-market buyers showing purchase intent", and demand generation as the work that "creates market awareness and interest before buyers are ready to purchase". For the two halves of demand generation, see demand creation vs demand capture.

Three figures show why the distinction matters for a pipeline number:

• Buyers research alone.
• Speed decides deals. Refine Labs cited the figure that 78% of B2B customers purchase from the vendor that responds first, alongside a Chili Piper and Navattic study where the average vendor response time is two days.
• Declared intent converts. Refine Labs' framework set the benchmark for declared intent conversion form fill to qualified opportunity at 30 to 40%.

The scorecards differ in the same way. ZoomInfo lists these success metrics for each motion:

MQL is a marketing qualified lead and SQL is a sales qualified lead.

A worked check against the benchmark

Take your demo and contact-sales form fills for one quarter and compare them to the qualified opportunities they produced. The numbers below are illustrative.

Declared intent form fills (quarter):        200
Benchmark low  (30%):  200 x 0.30 =           60 qualified opportunities
Benchmark high (40%):  200 x 0.40 =           80 qualified opportunities
Actual qualified opportunities:               24
Actual rate:           24 / 200   =          12%
Gap to benchmark low:  60 - 24    =           36 opportunities

In our reading, a rate that far under the range points to two causes worth checking first: slow follow-up, which the response-time figures above put at the center, or forms that collect contacts from people who never meant to buy. The first is a handoff problem. The second is a lead generation program counting names that demand never produced.

What each one optimizes for

Lead generation is judged on a count. In New Breed's framing, the traditional marketer "is measured on the number of leads they bring in" (New Breed). Two working numbers follow from that: lead volume for the period, and cost per lead (CPL), which is program spend divided by leads captured.

Demand generation is judged on money. The same guide says the demand generation marketer "is measured on their total contribution to revenue" (New Breed). The leads that team hands to sales have to fit the company well enough to close.

Why the two goals pull apart

Alice de Courcy, former Cognism chief marketing officer, described the split as: "convert the 1% already in-market, but focus on generating demand in the 99% who aren’t" (Cognism).

In our reading, that split explains the tension. A team paid on lead volume has a reason to put forms in front of the 99%, because every download counts toward the number. A team paid on revenue spends on the 99% so they know the brand before they start buying, and keeps the path short for the 1% who are buying now.

A worked example with our own numbers

The rates below are illustrative, not benchmarks.

Program A (gated ebook)
  Leads: 800          CPL: 20,000 / 800 = $25
  Lead to opportunity: 2%   -> 16 opportunities
  Win rate: 20%             -> 3.2 deals

Program B (ungated content, demo requests captured)
  Leads: 100          CPL: 20,000 / 100 = $200
  Lead to opportunity: 25%  -> 25 opportunities
  Win rate: 24%             -> 6 deals

Cost per deal:  A = 20,000 / 3.2 = $6,250
                B = 20,000 / 6   = $3,333

On a CPL dashboard, Program A looks eight times cheaper. On cost per deal, Program B costs about half as much. The gap in real programs can be wider. Liam Bartholomew of Alt21 reported: "Our close rate from content leads was around 0.2% compared to nearly 20% from direct inbound enquiries" (Cognism).

The practical step is to report both numbers side by side and let cost per deal settle budget questions. Count a marketing qualified lead (MQL) as a step in the funnel, not the result. The demand generation strategy playbook covers how to set those targets.

Is lead generation part of demand generation?

Yes. In our reading, lead generation is one mechanism inside demand generation, not a parallel program. ZoomInfo's guide puts it plainly: "Demand generation encompasses two distinct motions: demand creation and demand capture." Refine Labs adds a third. Evan Hughes, VP of Growth at Refine Labs, outlines three distinct stages: demand creation, demand capture and demand conversion.

Three stages, one system

• Creation. Refine Labs describes this stage as generating interest and brand awareness before a buyer is in-market.
• Capture. In the same model, once a buyer starts showing intent, demand capture channels help bring them into a conversation, often through search engines or review sites.
• Conversion. The final stage focuses on how quickly and efficiently you can convert interest into revenue.

Lead capture belongs to the second stage. The form fill, demo request or trial signup is the moment an in-market buyer hands over contact details. That is why we treat lead generation as a subset of capture, not as a rival to demand generation. ZoomInfo states the dependency directly: "you can't capture demand that doesn't exist, and creating demand without a capture mechanism wastes opportunity."

Where the budget goes

The stage model also shapes spend. Refine Labs allocates 60 to 70% of paid media budget toward demand creation, and "Only 30% is allocated toward demand capture, such as paid search." Hughes warns that teams over-invest in capture simply because it is trackable and tied to last-click attribution.

Here is that split applied to a sample quarter. The budget figure is illustrative.

Quarterly paid media budget:        $120,000
Creation at 60% to 70%:             $72,000 to $84,000
Capture at 30%:                     $36,000
Unassigned in the 60% case:         $12,000

Lead generation programs (paid search, review sites,
gated offers, demo forms) draw from the $36,000 capture line.

Read this way, a lead target is a target for one line of the capture budget. It is not a target for the whole program. If capture spend climbs past its share because it reports cleanly, the creation line shrinks. Over time, in our reading, fewer buyers then arrive at capture already familiar with the brand.

For how to split work and measurement between the first two stages, see our guide to demand creation vs demand capture.

Which metrics does each use?

Each approach gets judged on a different scoreboard. The table below maps programs to the numbers that move first (leading) and the numbers that settle the argument (outcome). Copy it into a planning doc and fill in your own targets.

ACV here means annual contract value. Pipeline value is the outcome metric both teams can share, because it multiplies deal size by the count of real opportunities and ignores how many forms were filled.

The flat CPL trap

Lead metrics can look healthy while the business number drops. One agency describes the pattern its prospects bring in: "Click-through rate is solid. CPL is holding. And yet the deals coming in aren't the right ones, or there aren't enough of them." Part of the reason is timing. Martal Group reports that 73% of B2B leads are not ready to purchase on first interaction, so a form fill often counts a person who will not become an opportunity this quarter.

Here is a worked example with two quarters at the same spend and the same CPL.

Pipeline value      = ACV x opportunities
CPL                 = spend / leads
Lead-to-opp rate    = opportunities / leads
Pipeline per dollar = pipeline value / spend

Quarter 1: spend $60,000, leads 400, CPL $150
           opportunities 20, lead-to-opp 5%
           ACV $30,000, pipeline $600,000, pipeline per dollar $10

Quarter 2: spend $60,000, leads 400, CPL $150
           opportunities 12, lead-to-opp 3%
           ACV $30,000, pipeline $360,000, pipeline per dollar $6

A dashboard that stops at CPL reports two identical quarters. In this example, the outcome columns show pipeline down by 40%. We think this is the case the table's warning column is built to catch: check pipeline per dollar beside CPL every month, and treat a steady CPL with a falling lead-to-opportunity rate as a quality problem, not a cost win.

Two practical steps follow from the example:

1. Report CPL next to pipeline per dollar on the same row, never on separate slides. Our demand gen dashboard template is laid out this way.
2. Compare your CPL with business-to-business cost per lead benchmarks only after the opportunity rate holds. Add pipeline velocity once you have two quarters of opportunity data, since it adds win rate and cycle length to the same view.

When gating content makes sense

Gating is a trade. The buyer gives a name and an email, and you give them something they could not get for free. The rule we use follows from that: gate only when the exchange is fair to the buyer and sales will act on the contact within the week. If either half fails, take the form down.

The decision rule

Material that teaches the problem stays open. ZoomInfo describes demand generation as relying on ungated content that "creates category demand through thought leadership without friction". A reader learning why their forecast keeps slipping is not ready to talk to a rep. A form at that moment loses the reader and adds a name to the database that no one will call.

Assets for buyers already evaluating can carry a form. The same ZoomInfo comparison lists lead generation content as gated content: "whitepapers, webinars, product demos". Martal Group names gating all content too early among the common mistakes teams make.

Run each asset through three questions before it goes live:

Gate test (all three must be "yes")
1. Is the asset specific to the buyer's situation
   (a calculator, an audit, a benchmark against their peers, a demo),
   not a general explanation of the problem?
2. Would a buyer reasonably expect a follow-up after asking for it?
3. Has sales agreed, in writing, who calls these contacts and how fast?

Any "no": publish ungated and measure engagement, not form fills.

How common assets sort

Bought contacts are not created intent

Some programs look like gating but work differently. Content syndication pays a publisher to put your asset in front of its audience and pass back the names of people who downloaded it. Licensed analyst reprints often work the same way.

In our reading, that figure describes contacts acquired, not demand created. The downloader chose the analyst's report, not your product. These programs can have a place, but we think they belong in the budget as a paid list with its own cost per opportunity, measured separately from organic form fills. Our content syndication guide covers how those programs are priced.

Whatever you gate, route the contacts into a defined follow-up. A name sitting untouched in the database means the buyer gave up their email and got nothing back. The lead nurturing playbook sets out sequences for contacts who are not yet ready for sales.

A worked example with the same budget measured two ways

For example, take one quarterly budget of $60,000 and run it against two targets. Every rate below is an illustrative assumption from our own method, chosen to show the arithmetic. Swap in your own figures before you defend either plan.

Version A: a lead target

In this example, the team gates an ebook and buys downloads at a $100 cost per lead. That yields 600 leads in this example, which looks strong on a CPL report. These are content downloaders, not buyers asking for a call, so in this example we assume 3% become opportunities. That gives 18 opportunities in this example.

Version B: a pipeline target

In this example, the same $60,000 funds ungated content and paid distribution, with one form: a demo or pricing request. Fewer people fill it out. That gives 42 opportunities in this example.

For this worked example, assume an average opportunity value of $40,000 for both.

Version A (lead target)
  Leads            = 60,000 / 100      = 600
  Opportunities    = 600 x 0.03        = 18
  Cost per opp     = 60,000 / 18       = 3,333
  Pipeline         = 18 x 40,000       = 720,000

Version B (pipeline target)
  Hand raisers     = 60,000 / 500      = 120
  Opportunities    = 120 x 0.35        = 42
  Cost per opp     = 60,000 / 42       = 1,429
  Pipeline         = 42 x 40,000       = 1,680,000

Version B with slow follow-up (assume half the hand raisers buy elsewhere)
  Opportunities    = 120 x 0.175       = 21
  Cost per opp     = 60,000 / 21       = 2,857
  Pipeline         = 21 x 40,000       = 840,000

Version A reports five times the lead volume and less than half the pipeline. On a lead dashboard it wins. On a pipeline dashboard it loses.

How response time erodes Version B

Version B only works if sales answers fast. The two-day average response time cited earlier comes from a study that also found 2.3 emails are sent per inbound request. Each extra email and each day of delay gives a buyer who already raised a hand time to book with someone else.

In the third column we assume half the hand raisers go to a faster vendor. Conversion halves, cost per opportunity doubles, and pipeline lands close to Version A, at a much higher cost per form. In our reading, this is why a pipeline target fails when routing is slow: the demand was there, and the handoff lost it.

Two steps follow. Set a response-time target and report it next to pipeline, using the figures on speed to lead. Then work backward from your pipeline goal with the reverse funnel calculator, so the conversion rate you plan around matches the follow-up you can actually staff.

How to move a team off MQL targets

Teams that drop marketing-qualified-lead targets in one quarter usually lose the argument with finance before the new numbers mature. A staged change holds up better. Here is the sequence we use.

1. Run both reports side by side for one or two quarters

Keep the existing lead report exactly as it is. Next to it, build a pipeline view: qualified opportunities created, pipeline value, win rate and closed revenue, each split by source. Review both in the same meeting every month so nobody reads the new view as a replacement for the old one.

Expect the two to diverge. Refine Labs described one customer who saw a 20% drop in lead volume alongside a rise in qualified opportunities. The parallel period exists so leadership sees that pattern in your own data before any target changes.

2. Rewrite the service level agreement around qualified opportunities

Forrester lists misalignment across marketing, sales, and account-based marketing (ABM) among the problems demand leaders face. The service level agreement is where that gets fixed. Define a qualified opportunity in terms both teams sign, set a response time for hand-raisers, and record who accepted or rejected each one and why. Our marketing and sales SLA template has the fields.

3. Add self-reported attribution

Refine Labs tracks branded search volume, share of voice, and what buyers say when asked how they found a company. Self-reported answers surface the podcasts, communities and peer referrals that click-based models miss. The self-reported attribution playbook covers coding and reporting.

4. Retire MQL volume as a target, keep it as a diagnostic

MQL count stays on the dashboard. It stops appearing in goals and bonus plans.

If you score leads, keep the model for routing only. The lead scoring model template separates routing thresholds from goals.

5. Put brand, demand and expansion under one revenue number

Refine Labs describes a common failure: brand, demand and expansion running as three separate things, with different owners and different metrics and no one connecting them to the same revenue number. Give all three one shared target, such as new plus expansion revenue, and report each team's contribution against it. The sourced versus influenced method explains how to credit work that touches a deal without opening it.

Monthly transition review
- Leads this month / last month
- Qualified opportunities created / accepted / rejected (reason)
- Pipeline value created, by self-reported source
- Win rate, trailing quarter
- Revenue: new + expansion vs shared target
- Decision log: budget moved, from where, to where, why

Which one a small B2B team should focus on

Our view is that a small team should start from what it already has, not from a philosophy. Martal Group frames the rule plainly: use demand gen "when brand awareness is low or sales cycles are long," and prioritize lead gen "when you need a short-term pipeline and already have engaged audiences." We agree, and we add one thing: a quarterly check that tells you when the split has drifted.

The other direction is real too. Cognism reported that its pivot toward demand gen led to a 4x increase in inbound pipeline, with higher win rates and faster sales cycles. That result came from building demand first, not from capturing harder.

The coverage check that triggers a rebalance

Run this at the start of each quarter. It uses your own win rate, so the threshold fits your business.

required coverage = 1 / historical win rate
actual coverage   = open qualified pipeline / revenue target remaining

Worked example:
  win rate                  = 25%   -> required coverage = 4.0x
  revenue target remaining  = $400,000
  open qualified pipeline   = $1,200,000
  actual coverage           = 1,200,000 / 400,000 = 3.0x
  gap                       = 4.0x - 3.0x = 1.0x short ($400,000 of pipeline)

1. Coverage at or above required for two straight quarters: move spend back toward creation, for example by 10 points, because the next quarter's pipeline depends on demand you build now.

Cap the capture shift at one quarter. In our reading, a team that keeps moving money into capture to fix coverage ends up harvesting a shrinking pool and needs the shift again next quarter. If you need the rule two quarters running, the problem is awareness, and the fix sits in the creation row.

Track the ratio in the pipeline coverage calculator so the trigger is a number on a dashboard, not a debate in a planning meeting.

What this page does not cover

This page does not cover how to build a lead scoring model, how content syndication programs are priced, or how to code self-reported attribution answers in detail. The rates in the worked examples are illustrative and should be replaced with your own data. Nothing here is financial or legal advice.