Before a company buys: the B2B buying journey and the dark funnel

When a demo request comes in through your form, the company has often already done most of the journey. It has identified its problem, looked for solutions, listened to peers and sometimes ruled out suppliers. You arrive at the end of a story that started without you.

Understanding that journey serves one purpose: knowing when to write, and what to say, so you arrive while the choices are still open.

What are the stages of a B2B purchase before first contact?

Timings change from one purchase to the next, but the order is almost always the same.

1. A tension appears. Recurring delays, a tool that can’t keep up, an overloaded team. Nobody is looking for a supplier yet: people just notice a problem.
2. The problem is recognised internally. It comes up in meetings, someone is put in charge of it. It becomes a priority, but the budget and criteria aren’t set yet.
3. The company looks for solutions. The person in charge reads, asks peers, attends a webinar, comments on a post. This is the stage most visible from the outside.
4. It compares a few suppliers. A shortlist forms, the criteria harden, sometimes a specification is written.
5. A group approves and signs. Finance, purchasing, senior management: new people join, who look at risk and cost more than at the solution.

Gartner sums up what is at stake in its research on the B2B buying journey: buyers spend only about 17% of their journey talking to potential suppliers, all suppliers combined (Gartner, The B2B Buying Journey, study published in 2019). The rest happens among themselves, without you.

What is the dark funnel, in practice?

The dark funnel is the whole part of the journey your tools don’t measure. Your CRM and website analytics only see what people do on your own channels: a visit, a download, a form. Everything else escapes them:

• a chat between colleagues by the coffee machine;
• one founder recommending a supplier to another;
• a question asked in a private group or a direct message;
• an article read without logging in;
• a podcast listened to in the car.

That is where the first preferences take shape. When the prospect finally reaches you, they already have an idea of what they want, and often of who they want it from.

What public traces does this journey leave?

Not everything is invisible. Some stages produce public, professional traces, especially on LinkedIn, where reactions to posts are visible to everyone. Communities such as Slack or Discord, on the other hand, remain largely closed.

Stage	What you can see	Where	What stays invisible
1. Tension	A post about a difficulty, an irritated comment	LinkedIn, public reviews	Internal discussions
2. Problem recognised	A hire, an appointment, a project announcement	Job postings, LinkedIn, press	The budget, the timeline
3. Research	Repeated reactions to content on the topic, a question in a comment, a webinar sign-up	LinkedIn, public forums	Anonymous reading, advice from peers
4. Comparison	Reactions on competitors’ posts, a public request for opinions	LinkedIn, open communities	The shortlist, the specification
5. Approval	Almost nothing		The committee’s trade-offs

You will never see the whole dark funnel. But you don’t need to: you only need to spot the traces of stages 2 and 3 from the right people. To know which ones to look for, see trigger events in sales prospecting and the list of our signal sources.

When should you write?

During stages 2 and 3. The problem is recognised, so your message doesn’t fall flat. The criteria aren’t fixed yet, so you can help shape them.

Too early, at stage 1, the topic isn’t a priority: at best, you’ll hear “not now”. Too late, at stage 4, the comparison has started. The specification may have been written with another supplier, and all that’s left is to match on price.

Two common situations:

• A start-up announces a funding round. A supplier spots it and writes two months later. In the meantime, budgets have been allocated and providers chosen. The message gets filed away.
• A new production director arrives at a plant. He reviews his suppliers in his first weeks. Whoever writes to him six months later is addressing someone who has already made his choices.

In both cases, the signal was good. It was the delay that made it useless. We go into timing in more detail in how to find the best time to contact a prospect.

What do you say to someone who hasn’t asked you for anything?

At stage 3, your prospect is trying to understand their problem, not to buy your product. So the message has to speak to the stage they are at. No introduction, no demo: an observation about their situation and a simple question.

Illustrative example. Signal: the operations manager of a 120-person manufacturer asks, in a comment on a sector post, how other plants handle delays from their subcontractors.

Opener that falls flat: “Hi Mark, we are a certified subcontractor with 40 years’ experience and modern machinery. Would you be available for a presentation of our capabilities?”

Opener that gets a reply: “When a distant subcontractor falls two weeks behind on a part, it’s often a whole line left waiting. Are you looking to secure your current suppliers, or to add one closer to home?”

The second version picks up his own question, in his words, and invites him to clarify his need. If he replies, the next step is natural: a twenty-minute conversation about his parts and lead times, with a clear reason. The meeting comes from the conversation, not from asking for a slot in the first message.

Should you rely on website visits and intent data?

This data has its uses, but it answers a different question. Website visits only show people who already know you, often late in their journey, and most remain anonymous. Intent data sold by third-party providers indicates that a company is reading a lot about a topic. It doesn’t say who, or why, or what to write to them.

A public signal from a named person, on the other hand, gives you all three: a name, a topic, a context. That is what lets you write an opener that doesn’t look like the others. The difference is explained in intent data vs buying signals.

This is where the MeetMagnet AI works: spotting, every day on LinkedIn, the posts and interactions that reveal stages 2 and 3, from people who match your target, then writing the first message based on what they are going through.