I get asked to make introductions most weeks. I say yes to some and not to others, and the deciding factor is almost never how much I like the person asking.

It is whether I can predict what will happen next. If I introduce two people and it goes well, both of them think slightly better of me. If it goes badly, or worse, if one of them is vague and wastes the other's afternoon, I have spent something I cannot easily get back.

That is the thing founders miss about networks. They are not a collection of names. They are a set of small accounts of credibility held by other people, and they behave like any other asset. They can be built deliberately, they depreciate if neglected, and they can be spent badly.

01Why is reach the wrong measure?

Because the useful unit is not how many people you know. It is how many people could describe what you do accurately, without you in the room.

Test this. Pick ten people from your contacts. Ask yourself what each of them would say if a friend asked whether they knew anybody who does what you do. For most founders, the honest answer for seven of the ten is that they would not think of you at all, and for two more they would describe you as something you were four years ago.

That is the actual size of your network. Not the connection count. The number of people carrying a current and accurate description of you.

A hundred people who could place you correctly is a working network. A thousand who could not is a mailing list.

02What does a mapped network look like?

Take your next three objectives. Not aspirations, objectives: the specific things you need to happen in the next eighteen months. Enter a sector. Find a non-executive with experience of your size. Win two customers of a particular type. Prepare for a transaction.

Now, for each objective, write down the names of people who could materially move it. Not people who work in that world generally, but people who could open a specific door.

Two things happen when founders do this. The first is that most objectives have one or two names, and several have none, which is uncomfortable and useful. The second is that a handful of names appear against every objective, and those people are almost always undervalued in how the founder allocates their attention.

The exercise takes an hour. It is the difference between attending events and building something.

A network is not who you know. It is who could describe you accurately to somebody you have not met, next Tuesday, without being prompted.

03How do networks depreciate?

Faster than people assume, and unevenly.

A relationship you have not touched in two years is a name rather than a contact. You can usually still make the call, and it will be a cold conversation with a warm opening, which is not the same as a warm relationship. The information both of you hold about the other is out of date, which is precisely the information that makes an introduction useful.

The decay is not linear. Relationships built through doing something together, a project, a difficult negotiation, a shared board, hold their value for years. Relationships built through proximity, an event, a shared table, a conference, decay within months unless something follows.

Which suggests a simple allocation rule. Spend your maintenance effort on the relationships that were built through work, and stop trying to maintain the ones that were built through attendance. Most founders do the opposite, because attendance-based relationships come with a diary invitation and work-based ones do not.

04Why do warm introductions matter so much in mid-market deals?

Because they transfer something that cannot otherwise be established quickly, which is the assumption that you are not going to waste the other person's time.

In transactions particularly, the difference is stark. An approach that arrives cold gets assessed on its merits, slowly, in a queue. The same approach arriving through somebody the recipient trusts gets a meeting in a fortnight and starts from a position of assumed competence. Nothing about the underlying proposition changed. What changed is who is standing behind it.

The same holds for senior hires, for board appointments, for partnership conversations, and for most of the things that matter at the point when a business is trying to move up a level. It is not a shortcut around quality. It is a shortcut around the time it takes to establish that quality exists.

05What makes somebody worth introducing?

Four things, and they are all learnable.

You are specific. People who ask to be introduced to "anyone in retail" do not get introduced. People who ask for an introduction to a named person, with a reason, and a sentence explaining what they want from the conversation, get introduced most of the time.

You close the loop. Tell the person who made the introduction what happened. Almost nobody does this, which is remarkable given how cheap it is and how much it affects whether they do it again. Two lines is enough.

You are useful in the other direction. The people who receive the most introductions are the ones who make them. Not as a transaction, but because the habit of thinking about who two other people should meet is what makes somebody worth knowing.

You do not waste the meeting. Arrive knowing what you want, be clear about it in the first five minutes, and be honest if it turns out there is nothing here. The person who says "I do not think this is a fit, but thank you, and here is somebody who might be" gets introduced repeatedly for years.

06What this looks like in practice

Twenty minutes a fortnight, spent deliberately rather than reactively.

Look at your three objectives. Look at the names against them, and the gaps. Then do one thing: reconnect properly with somebody who has decayed, ask for one specific introduction, or make one for somebody else.

That is it. It is not a networking strategy and it does not require attending anything. Over two years it produces a network that is shaped like your objectives rather than shaped like your history, which is the whole difference between a network that accumulates and one that works.

The founders I know with genuinely valuable networks are not more sociable than everybody else. Several of them are notably less so. They have simply been deliberate about a small number of relationships for a long time, and they have never once asked to be introduced to anyone in retail.