Because only one of them talks to you, and a CRM records conversations rather than decisions. The other five show up in the account history after the deal is lost, if at all.
A B2B purchase of any size is a group decision. The person who raised it, whoever owns the budget, the team that has to use it, and lately somebody from security or legal or finance who appears late and is not pleased about it.
One of those people emails you. That is the contact on the deal, and from the CRM's point of view that contact is the account. Everything looks engaged right up until it does not.
The failure is invisible until it is terminal
A single-threaded deal does not look risky. It looks like your best deal: one responsive contact, fast replies, good meetings, genuine enthusiasm.
What the record cannot show is that the enthusiasm belongs to one person who has to go and win an internal argument you are not in the room for, armed with whatever they remember from your call.
| What the CRM shows | What is happening |
|---|---|
| Contact replies within the hour | Champion is engaged. True, and not the question |
| Three meetings in two weeks | All with the same person |
| Stage: evaluation | Finance has not seen a number |
| No activity for eleven days | The internal argument is being lost |
| Closed lost, reason: budget | The champion could not get it funded alone |
"Budget" is the most over-recorded loss reason in B2B sales and it is usually shorthand for a conversation that happened without you.
Two numbers worth having
Almost no team calculates these and almost every CRM can produce them.
Average contacts engaged on won deals. Not contacts in the database. People who actually exchanged something with you.
The same figure on lost deals.
The gap between those two numbers is a better predictor of deal health than any stage, and it gives you a threshold you can act on. If your won deals average four engaged contacts and your lost deals average 1.4, then a deal sitting at one contact in week three is a deal with a known problem, and you have a reason to act that is not a hunch.
It also gives a sales manager something useful to ask in a pipeline review other than whether the deal will close. "Who else have we spoken to" is answerable, specific and leads somewhere.
Asking without going over anybody's head
The reason reps do not multi-thread is not that they do not know they should. It is that it feels like going behind the champion's back, and done clumsily it is.
The version that works asks the champion to widen the circle rather than going around them, and frames it as protecting them, which is accurate:
"When this gets to finance, they'll ask about X. I'd rather they hear the answer from both of us than from a forwarded email. Can we get twenty minutes with them before it goes up?"
That is not a request for an introduction. It is an offer to help them not get caught out, and champions generally accept it, because being caught out is the thing they are most worried about.
When they refuse, that is information too. A champion who will not introduce anybody is either not as senior as the deal needs or not as committed as the meetings suggest.
What to record, and why titles are the wrong field
Job title tells you what somebody does for their company. Decision role tells you what they do to your deal, and they are often different people: the most influential person on a software purchase is frequently the one who has to migrate the data, whose title says nothing about authority.
Worth holding per person:
- Their role in this decision: raised it, owns the budget, will use it, can veto it
- When they last engaged, and with what
- Whether they have ever been in a conversation with you directly
In Empiraa Signal, accounts and contacts is where the committee lives rather than a single contact record attached to a deal, so a deal with one engaged person is visibly a deal with one engaged person. The reason it matters that this sits in the CRM rather than in a spreadsheet is that the activity, the sequences and the proposal are all on the same record, so "who has actually engaged" is a fact rather than a recollection.
The thing that makes this more urgent than it was
Buyers now arrive having researched the category without you, which means more of the internal discussion happens before first contact and more of it continues without you afterwards. Buyers are shortlisting you inside ChatGPT covers what that does to the shape of the pipeline; the committee consequence is that the room you are not in has already met.
A deal where you know one person out of six was always risky. It is riskier when the other five have formed views before you arrive.
Do this one thing on Monday
Open your last twenty closed deals, ten won and ten lost. Count the people who actually engaged on each.
If you cannot tell from the record, that is the finding, and it is worth more than the number would have been.
Related reading: buyers are shortlisting you inside ChatGPT, and the proposal is the forecast.
Signal includes unlimited users on every plan, so adding the people who should be in the account costs nothing.
