Sales & PartnershipsFree
Deal Multithreading and Access Plan
Activity data gives one number for how concentrated the deal is, and your own won deals give the roles that should already be in it.
The advice is universally the same and it is useless in the form it is given. Do not rely on one contact. Aim for four to six engaged stakeholders. Below three is a single point of failure. Ask your champion for an introduction. None of it tells a rep whether their own deal is single threaded, because contact count is the wrong measure and every CRM reports it as though it were the right one.
Six contacts on an account can be one real thread with five bystanders attached to it. The measure that separates the two is concentration, not headcount: square each contact's share of total logged activity, sum the squares, and you get an effective thread count. In the worked example six contacts becomes 2.42, and a completely separate test, counting who has actually replied to anything in the last month, independently gives two. Those two numbers landing in the same place is the whole argument for the measure.
The second half is what to do about it, and it comes from the seller's own closed-won history rather than a generic template. Which roles appeared in your last several wins, and by what day each one first appeared. A role missing here past the day it normally shows up is the gap worth working. Built for the account executive whose deal has gone quiet on one contact. Who is in the buying group is the stakeholder map, whether the deal is real is the qualification scorecard, and the tool index has the rest.
The concentration measure already exists, borrowed from antitrust
Regulators have had this problem for decades. Counting firms in a market tells you nothing, because four firms holding 97, 1, 1 and 1 percent is not four competitors. The answer is the Herfindahl-Hirschman Index, calculated by squaring the market share of each firm and summing the results, which lands anywhere from near zero to 10,000. Above 1,800 the agencies treat a market as highly concentrated. It is exactly the same shape of problem as a deal with six contacts and one of them holding sixty percent of the conversation.
Run it on activity shares and the deal in the worked example scores 4,134, more than twice the threshold at which a regulator would call a market highly concentrated. Turn that into an effective count by dividing into 10,000 and you get 2.42 threads. Not six. The reason it works is that squaring punishes concentration hard, so a contact holding two percent of activity contributes almost nothing to the total, which is the correct treatment for somebody who was copied twice in April.
Then a second measure, built differently on purpose. Count only contacts who have sent you something unprompted in the last thirty days. In the worked example that is two, and two independent measures landing in the same place is what makes the number defensible in a deal review. A rep can argue with a rule of thumb about four to six stakeholders. Arguing with your own activity log is harder.
How it works
Paste the activity
Contacts, dates and direction, from a CRM export or from the email threads themselves.
Add your closed deals
Wins and losses of a similar size, so the role benchmark comes from your own history.
River counts threads
Concentration, effective threads and live replies, then the missing roles dated against your wins.
Send the asks
One message per missing role, each naming a person, a meeting and a reason to agree.
What you get
- One concentration number for the deal, computed from your own activity log
- The effective thread count, which is almost always lower than the contact count
- A second independent read from who has replied unprompted in thirty days
- The roles your own closed-won deals had, with the median day each was first touched
- Every missing role dated against that benchmark, so lateness is in days not adjectives
- One written ask per missing role, with the route and a reason the buyer agrees
Common questions
Our CRM activity logging is patchy.
Then it reads the mailboxes and the calendar instead, which are usually better records than the CRM anyway. Where coverage is genuinely thin the tool says which period it trusts and computes on that, rather than presenting a concentration figure built from a third of the touches as though it were complete.
Why square the shares instead of just counting people?
Because squaring makes a small share almost worthless, which is the right treatment. A contact holding two percent of the activity adds three points to a total of over four thousand. Counting them as a thread is how a deal running entirely through one person reports as having six-way coverage in a pipeline review.
We do not have enough closed deals for a benchmark.
Then it says so and falls back to what this buyer's own process implies, plus whatever you can describe from memory. A thin benchmark is labelled thin. The concentration half works regardless, because it only needs this deal's activity, and that half is usually the one that changes the conversation.
Our champion does not want us going around him.
Then the asks are written to go through him, which is a different message and a better one. Every route names who makes the introduction and why the buyer benefits from making it. Where a contact has blocked the same introduction twice, that pattern is in the activity log and gets reported, because it is a deal risk rather than a preference.
Is a high concentration always bad?
Not always, and small transactional deals run fine on one thread. It matters when the number is compared against your own wins at the same size. If your last eight wins all had an economic buyer by day 34 and this one is at day 63 with none, the concentration is not a style question any more.
What makes an ask work rather than sit in a plan?
A named person, a specific meeting and a reason the buyer gains from agreeing. Compare would you introduce me to your CFO with before your finance team sees numbers, can we spend twenty minutes with whoever owns that budget line. One is a favour and the other is self-interest, and only the second gets scheduled.
Where does this sit against a mutual plan?
Before it. This finds who is missing and gets the first meeting. Once several people are engaged and there is a sequence to run, that becomes the mutual action plan. Running a mutual plan through a single thread just gives one person a longer list.
Deal Multithreading and Access Plan
Fill in the form and your workspace opens with the work already underway.