B2B Demand Generation: A 7-Step Strategy That Fills Pipeline (Not Forms)

The $94K demand gen budget that produced 11 customers

A SaaS founder forwarded us his last quarter’s marketing report in January. $94,000 spent across LinkedIn, Google, and a “demand gen agency” running webinars and gated reports. The deck had 41 slides. MQLs were up 38%. SQLs were up 22%. Pipeline was, in his exact words, “weirdly flat.”

When we pulled the data, it took about twenty minutes to find the problem. The agency was running a textbook B2B demand generation strategy — top-of-funnel content, gated assets, retargeting, paid social — but every optimization decision pointed at the wrong number. They were optimizing for form fills. The form fills were students, job-seekers, and competitors downloading the report. Out of 1,200 “leads,” 11 became customers. CAC was $8,545.

This is what most B2B demand generation looks like in 2026. Not bad teams. Not lazy teams. Just teams running 2018’s playbook against 2026’s buyer.

This piece is the playbook we actually run for our SaaS clients — the channel mix, the metrics we ignore, the ones we obsess over, and the parts of the conventional wisdom we’ve thrown out. If you’re a Series A or B SaaS company, around $1M to $10M ARR, with a 5- to 90-day sales cycle, this is for you.

What B2B demand generation actually means (and what it doesn’t)

B2B demand generation is the work of making your ideal customer aware that they have a problem, that solutions exist, that you’re a credible option, and that now is the time to act. Lead generation is what happens after the demand exists — capturing intent and routing it to sales.

Most “demand gen” teams are doing lead gen with a fancier name. They’re harvesting people already in-market. That works until your category is saturated and your CAC starts climbing 30% a year.

A real B2B demand generation strategy does three things:

It creates demand from people who don’t yet know they need you. It captures the demand that’s already searching. And it accelerates demand that’s stuck — accounts that know about you but haven’t moved.

If your strategy is only doing the second one, you’re playing a finite game. Here’s how to play all three.

The seven-step B2B demand generation strategy we run for SaaS

Step 1 — Define the buyer, not the persona

We’re done with persona docs. They’re useful for content teams and useless for paid media. The level of abstraction is wrong.

What we build instead: an ICP filter that any campaign target list, any audience segment, any account list has to pass. Six fields, no more.

• Industry (specific, e.g. “B2B SaaS, vertical = HR tech”)
• Company size (employee count + ARR band)
• Tech stack signal (using competitor X, or using Salesforce, or using a specific data warehouse)
• Buying committee role (the four titles that actually attend deal calls)
• Trigger event (recent funding, new hire in a key role, M&A, expansion to new geo)
• Geography

If you can’t filter your target list to under 5,000 accounts using these six fields, the filter isn’t sharp enough. Most B2B SaaS companies should be running demand gen against a Total Addressable Account list of 1,500 to 4,000 — not against “marketing operations professionals interested in B2B.”

Step 2 — Pick the three channels you’ll actually win on

Most demand generation strategies fail because the team is “running everything.” Five channels at half-effort beats running every channel.

Here’s the channel matrix we use to pick:

Stage	Best for	Worst for
LinkedIn Ads	Awareness + ABM. Targeting precision is unmatched	High-intent capture. Expensive per click
Google Ads	Capturing existing demand. Branded + non-branded search	Creating new demand. Display is mostly waste
Meta Ads	Self-serve SaaS, PLG products, founders / SMB buyers	Enterprise. The targeting collapsed in 2022
Reddit Ads	Technical buyers, devs, founders. Cheap CPMs, high trust	Anything that needs polish. Reddit hates ads
YouTube	Brand association, repeat exposure to ICP	Direct response. Don’t measure on last-click
Webinars	Mid-funnel, account-warming, sales handoff	Top-of-funnel. Most webinar registrants don’t show
Content / SEO	Long-term compounding moat	Anything you need this quarter

For most Series A B2B SaaS companies, the winning combination for SaaS demand generation is LinkedIn (creating demand) + Google Search (capturing demand) + content/SEO (compounding). Add a fourth channel only when you’ve got the first three actually working. “Working” means CAC inside payback target and pipeline volume that justifies a sales rep.

We’ve watched founders try to run LinkedIn, Meta, Google, Reddit, and TikTok simultaneously on a $40K/month budget. Every channel gets starved. Every channel underperforms. Every channel gets blamed.

Step 3 — Build the content engine before turning on the budget

This is the step everyone skips. The temptation is to launch ads first because budget feels like progress. Then a quarter later you realize the ads point at landing pages with no proof, no narrative, and no answer to “why now.”

Before you spend a dollar on demand gen, you need:

• One pillar narrative — the 90-second version of why your category needs to exist and why your approach is the right one. Not a feature list. A point of view.
• Three “category education” pieces — long-form articles or videos that teach your buyer how to think about the problem. Not how to use your product.
• Three “proof” assets — case studies with real numbers, ideally one per ICP segment. Vague testimonials don’t work. “We cut sales cycle from 87 days to 41 days” works.
• One ROI calculator or audit tool — something that gives a personalized output. These convert at 4-6x the rate of gated PDFs.

Our blog post on building high-converting landing pages goes deeper on the page-level mechanics. The ROI calculators we host at onemetrik.com/tools/ are an example of what calculator-style assets look like in production.

If you skip this step, your CPL will look fine for the first month. Pipeline will be flat. You’ll think the channel is broken. The channel is fine. The asset shelf is empty.

Step 4 — Map the funnel to actual buyer behavior

The classic funnel — awareness → consideration → decision — is too clean for how B2B SaaS deals actually close. Real buyers do this:

They ignore you. Then they see you somewhere they trust. Then they ignore you. Then a peer mentions you. Then they Google you. Then they don’t reply. Then a trigger event happens at their company. Then they’re suddenly interested. Then they go dark for six weeks. Then they ask for a demo.

The implication: a demand generation strategy needs to keep the same accounts warm across multiple weeks of inactivity. We map this with three named tiers:

• Tier 1 — Awareness layer. LinkedIn and Reddit ads pointing at thought-leadership content and category narrative. Goal: be the third or fourth time the buyer sees the brand. CPM is the metric, not CPL. Don’t gate.
• Tier 2 — Consideration layer. LinkedIn retargeting + content syndication + non-branded Google Search. Goal: get the buyer to engage with proof — case studies, calculators, comparison content. Form fills allowed but not the goal.
• Tier 3 — Decision layer. Branded search, demo retargeting, account-specific outbound. Goal: book the call. CPL and CAC are the metrics here, not anywhere upstream.

Most “demand gen agencies” run all three tiers as if they’re the same campaign. That’s how you end up optimizing your awareness ads for cost-per-lead and wondering why your top of funnel collapsed.

Step 5 — Build the ABM layer for your top 200 accounts

If you’re selling above $20K ACV, a horizontal demand strategy isn’t enough. You need a deliberate, named-account layer running parallel to it.

The mistake most teams make: they call this “ABM” and treat it as a separate program with a separate agency, a separate budget, and zero coordination with the rest of marketing. Then they wonder why marketing-sourced pipeline doesn’t move.

What works: pick 200 accounts that match your ICP and represent at least 70% of your TAM revenue potential. Run a coordinated layer on top of demand gen for those accounts specifically.

The layer looks like this:

• LinkedIn ads filtered to those exact 200 accounts at the buying committee level (4-6 titles per account, ~1,000 individual contacts)
• Cold outbound from sales, sequenced and timed to ad exposure
• Content distribution direct to inboxes (not gated downloads — links to ungated assets they can read in 90 seconds)
• A named landing page or microsite for tier-1 accounts when sales cycles get serious

We’ve broken down the mechanics of running this in our SaaS ABM page and the 3 leaks costing B2B companies their pipeline post. The short version: ABM stops working when it’s a separate motion from demand gen. It starts working when it’s the targeting layer over the same demand gen channels.

Step 6 — Wire up the measurement (this is where most teams fail)

You cannot run a serious demand generation strategy on Google Analytics last-click attribution. We need to say this loudly because half of the SaaS audits we run have last-click as the primary attribution model.

What you need at minimum:

• First-touch and multi-touch attribution — so you can see which channels create demand vs. capture it
• Self-reported attribution on every demo form — a single field: “How did you hear about us?” with categories matching your channels. This is the most underrated data source in B2B marketing.
• Pipeline-by-source reporting, not lead-by-source. The unit of measurement is dollars in pipeline, not form fills.
• CAC payback by channel cohort, not blended CAC. Blended numbers hide which channel is actually working.

If your reporting still shows “MQLs from Channel X” as the primary KPI, your team is optimizing for the wrong outcome. Pipeline is the metric. Closed revenue is the metric. Everything else is a leading indicator at best, vanity at worst.

The MQL itself is not dead, but treating it as the goal is. We use MQLs as a diagnostic — a sudden change in MQL volume tells us a channel changed — but we never let a campaign be optimized against MQL count.

Step 7 — Set the budget so each channel can actually work

This is the math conversation most agencies avoid. Channels have minimum viable budgets, and running below those minimums is worse than not running them at all.

Rough floors we use for B2B SaaS, monthly:

• LinkedIn Ads: $10K minimum to get statistical signal at the audience precision B2B requires. $20K+ to actually run a tiered demand program.
• Google Search: $5K minimum for a focused branded + non-branded search program. Performance Max is usually a trap for B2B SaaS — we wrote about why in does Performance Max cannibalize search.
• Reddit: $3K minimum, but only if your buyer is genuinely on Reddit. Don’t force this one.
• Meta: $5K minimum for B2B, but only for self-serve / PLG / SMB-targeted SaaS. Skip for enterprise.
• Content production: $5-10K/month for serious pillar content + distribution. Less than this and you’re publishing without distributing.

A common pattern we see: founders allocate $40K/month across six channels at $6-7K each. Every channel is below its viable floor. Nothing works. The right answer is usually: pick three channels, fund them properly, and add the fourth in six months.

If your total budget is below $25K/month, you should probably not be running paid demand gen across more than two paid channels. Pick the two that match your motion and double down.

What we cut from conventional demand generation strategies

Three things we routinely tell clients to stop doing.

Gated long-form whitepapers. They worked in 2017. In 2026, your buyer doesn’t fill out an 11-field form for a PDF. They Google the topic, read three articles, and assume your gated content is the same as the free content. Ungate the educational content. Gate the personalized output (assessments, calculators, audits, ROI models). Form fill rate goes up. Quality goes up. Everyone wins.

Webinars at the top of the funnel. A 45-minute commitment from a stranger is too much to ask. Webinars work in the middle of the funnel, with warm accounts, when the topic is narrow and the speaker is credible. Stop running monthly horizontal webinars to drive net-new leads. The math has been broken for two years.

SDR follow-up on cold MQLs. This is the most expensive lead leak in B2B. The flow is: a stranger downloads a checklist. An SDR calls them within 5 minutes. The stranger is annoyed. The SDR is demoralized. The MQL gets disqualified. Three weeks of CRM hygiene later, the lead is dead. Most “MQL-to-SQL conversion rate” problems are actually “we shouldn’t have called these people” problems.

What this looks like put together

For a typical Series B SaaS client at $5M ARR, $30K MRR target from new pipeline, $50K monthly marketing budget, our recommended B2B demand generation strategy looks roughly like this:

• $20K/month to LinkedIn — split 60% awareness against ICP audience, 30% retargeting, 10% account-list ABM layer
• $12K/month to Google Search — 70% non-branded search against high-intent keywords, 30% branded
• $8K/month to content production + distribution (in-house or freelance writers, our perspective on this is in our B2B SaaS content strategy breakdown)
• $5K/month to tools and attribution (decent attribution platform, intent data, enrichment)
• $5K/month reserved for testing channel #4 — usually Reddit or YouTube — once the core three are clearly working

Pipeline targets we’d expect at this spend level after the engine has been running 4-6 months: $400K to $700K in net-new sales pipeline per month, depending on ACV and sales cycle length. CAC payback inside 12 months for healthy SaaS unit economics, inside 18 months as a tolerable ceiling.

The first 90 days will look worse than this. Demand generation has a delay built in. If your team or your agency is showing you “great results” in week three, they’re either reporting on captured demand (which already existed) or measuring the wrong thing.

Frequently Asked Questions

The starting point if you don’t know where to begin

If your team is running some version of this and the numbers are flat, here’s the diagnostic order we’d run:

First, audit attribution. If last-click is still the primary model, fix that before changing anything else. You’re flying blind otherwise.

Second, count active channels. If it’s more than three for budgets under $30K/month, cut. It will hurt. Do it anyway.

Third, audit the asset shelf. If you don’t have one pillar narrative, three category-education pieces, three case studies with real numbers, and one personalized assessment tool, your demand gen problem is not a channel problem. It’s a content problem masquerading as a channel problem.

Fourth, look at the ICP filter. If your audiences in any platform have more than 50,000 people in them and you’re under $25K MRR target, the targeting is too loose.

Fifth, benchmark CAC payback by channel, not blended. Whichever channel has the worst payback is your biggest leak — and it’s usually not the channel you’d guess.

If you’ve been through that list and still can’t see what’s broken, that’s exactly the kind of problem we run free audits on. We pull your ad accounts, your attribution, your content shelf, and we tell you specifically where the leaks are. No deck, no fluff — usually a 45-minute conversation with the founder or marketing lead, screen-shared, with our actual recommendations on what to cut and what to scale. Grab a slot at cal.com/onemetrik/30min if it’s useful.

Demand gen isn’t broken. The 2018 playbook for it is. The teams that work out which parts to keep and which to cut are the ones whose pipeline will look very different in 12 months.