How to run B2B demand generation when buyers research privately: judge marketing by real sales opportunities, follow buying signals and earn trust before the first call.
Demand generation is the marketing work that makes companies want to buy from you and turns that interest into sales conversations. In many business-to-business (B2B) firms it still runs on a playbook from a decade ago: publish a guide, put it behind a form, score the downloads and pass the names to sales. The reports look healthy. Sales, meanwhile, says the leads rarely answer the phone, and the cost of winning each customer keeps creeping up.
The playbook that works now does three things differently. It judges marketing by qualified sales opportunities rather than by the number of leads, it concentrates budget on companies that show signs of buying, and it builds the kind of credibility that makes buyers contact you before anyone else. A small team can run all three without expensive software.
Below you will find what changed in how B2B buyers choose suppliers, then the five parts of the playbook: targets, buying signals, content, the conversations you cannot track, and how to split your budget. It ends with five things to do this month.
Three changes broke the old model, and none of them is going away. First, much of B2B research happens where you cannot see it: direct messages on LinkedIn, WhatsApp and Slack groups, podcasts, and calls between peers. Marketers call this dark social, because it leaves no trace in your analytics.
Second, search has changed. Google shows AI Overviews, short AI-written summaries above many results, and buyers also put their questions to assistants such as ChatGPT. A buyer can learn the basics of your category, and even draw up a shortlist, without visiting your website, so the early stages of their research never reach your reports.
Third, tracking is thinner. Safari and Firefox block third-party cookies by default. Chrome still allows them, since Google dropped its plan to remove them, but European consent rules mean many visitors are never tracked, and in the European Economic Area Google requires Consent Mode v2 for ads measurement. So attribution reports, which assign each deal to the marketing that produced it, show only part of the picture, and tend to credit whatever happened just before someone filled in a form.
A marketing qualified lead (MQL) is a contact that marketing judges ready to hand to sales, usually because they downloaded something, attended a webinar or reached a score in your marketing software. For years it was the standard target for B2B marketing teams. The trouble is that a download is cheap to get and says little about intent to buy, so a team rewarded for MQLs will find ways to produce more of them, whether or not anyone buys.
Replace that target with qualified sales opportunities: conversations in which sales has confirmed a real need, a rough budget, the people who decide and a timeline. Write the definition down together with your sales lead, then report three numbers every month: opportunities created, cost per opportunity, and the share of opportunities that become customers.
An illustration shows why this matters. Say you spend €6,000 a month on a gated e-book campaign and collect 150 MQLs at €40 each; sales accepts 12, and three become real opportunities, so each opportunity costs €2,000. A second campaign, LinkedIn ads offering a short diagnostic call, brings in just 25 leads at €240 each, but eight of them become opportunities, at €750 each.
Judged on MQLs, the first campaign wins easily. Judged on opportunities, the second costs less than half as much per real sales conversation. Expect lead volume to fall when you make the switch, and warn the rest of the leadership team before the first monthly report shows it.
At any given moment, most companies in your market are not buying. Spending equally on all of them wastes money, so concentrate budget and sales time on the accounts showing signs of an active search, which marketers call intent data. Start with the signals you already own, because they cost nothing and are specific to your business:
Third-party intent data is the paid version. Providers track which companies are reading about particular topics across large networks of business websites and flag those whose reading suddenly rises. It can help you rank a list, but it shows interest rather than a purchase, and coverage of smaller European markets and languages varies, so ask for a trial on a sample of your own target accounts and check the results against what your sales team knows.
Then let the signals change where effort goes. LinkedIn lets you upload a list of companies and show ads only to people who work there, so one monthly list of high-signal accounts can steer both your ad spend and your sales team's calls. Expect a higher cost per lead, and judge the result on cost per opportunity.
Generic articles now cost almost nothing to produce, so there are far more of them than anyone can read, and buyers skim past them. What stays scarce is material only you could have written, because it comes from your own data, customers and judgement. That is what earns trust, and it gives AI search tools something specific to quote.
In practice, four kinds of content earn that trust. Original research, such as a survey of your customers or an analysis of anonymised data you hold, gives buyers and trade journalists something new. A clear point of view signed by a named person shows how you think, detailed case studies with the customer's own numbers show what you deliver, and plain answers to the questions your sales team hears every week catch buyers at the moment they compare options.
Do not put all of it behind a form. Gating everything collects details from people who are not ready and hides your best material from those who are, including AI search tools, which cannot read behind a form. A workable rule: keep articles, case studies and research summaries open, and ask for contact details only when you offer something with one-to-one value, such as an assessment or a consultation.
Then distribute it with as much effort as you put into making it. One research piece can feed a month of LinkedIn posts from your leadership team, a newsletter, a follow-up email for sales and a talk at an industry event.
If buyers make up their minds in private conversations, you need to be part of those conversations, and that is earned, not bought. Find the two or three places where your buyers actually talk, such as a sector association, a LinkedIn group, a regional business network or a podcast they all follow. Contribute there regularly, with answers rather than pitches.
People trust people more than logos. A founder or senior expert who posts specific, useful observations on LinkedIn every week builds a familiarity that a company page rarely matches. Pick one or two people, agree the topics they own, and give them editing help and a schedule they can keep.
Because most of this influence never appears in analytics, measure it directly. Add an open-text field to your contact and demo forms asking 'How did you hear about us?', and read the answers every month: you will often find colleagues, podcasts and LinkedIn posts that your reports credit to a search for your brand name. Over the quarters, also watch searches for your brand name and direct visits to your site; a steady rise is a sign that awareness is growing.
Demand capture means being found by people who are already looking: search ads on the terms buyers use when they are ready, review and comparison sites, and retargeting. Demand creation means reaching the much larger group who are not looking yet, through content, LinkedIn and community work. Capture pays back quickly but has a ceiling, because you cannot buy more searches than people make; creation takes months to show in sales, but it is what raises that ceiling.
As an illustration, say you have €10,000 a month. You might put €5,000 into search ads and retargeting, and €5,000 into LinkedIn ads aimed at your target accounts, original content and your leadership team's posts. Review the split each quarter: if search results are flat while more of your 'How did you hear about us?' answers mention LinkedIn or a podcast, move money towards creation.
Judge creation work on a longer clock. Give it at least two quarters, and in the meantime watch the early signs: more target accounts engaging, more searches for your name, more prospects who mention your content on a first call. None of this replaces good campaigns, but it changes what they are for.
The best demand generation strategy is not a campaign; it is becoming the company buyers already want to call when they are ready.