The Pool Nobody Sizes

Ask a portfolio company where next year's growth comes from and the answer tends to be new logos. Ask them to size their dormant base and many cannot, because it has not been counted.

When we did count it, the number was consistently larger than the net-new target it was competing with.

At one PE-backed distributor, the count came to more than two thousand accounts, fully churned, at risk, or in visible decline, carrying over $400M of prior revenue between them. That was ranked the single largest organic growth opportunity in the business, ahead of every net-new motion, and not because the revenue was recoverable in full. It was ranked first because the conversion rate tends to be structurally higher. The relationship already exists. Somebody there is likely to know your name.

At an industrial machine distributor, the same exercise split the dormant base by how long it had been quiet. Accounts dormant three to five years accounted for roughly $400M of historical revenue. Accounts dormant six to ten years accounted for roughly another $270M. Together, more than 40% of every dollar the business had ever recorded sat in customers who were no longer buying.

Neither company was failing. Both were spending real money on demand generation aimed at people who had never heard of them.

Dormant Is Not Churned

The industrial machine distributor is where this gets interesting, because the obvious read was wrong.

The instinct with a six-year-dormant account is usually to write it off. In industrial machinery that instinct tends to be backwards. Customers replace a large machine roughly once a decade. A customer who bought six years ago is probably not lost. They are more likely mid-cycle, and they are exactly who you want a relationship with when the cycle turns.

That reframes the list entirely. It is not a win-back list, it is an interim-purchase list, and it segments by dormancy band. Three to five years out, the offer is accessory machines, tooling and upgrades, the things that get bought between major purchases. Six to ten years out, the offer is trade-in programs and financing, because those customers are approaching a replacement decision and the competitor who is in the room when they make it holds the advantage.

Same data, two completely different motions, and neither of them is a discount email.

A Fifth of Your Churn List Is Not a Churn List

Before any of this runs, the list itself has to be cleaned, and this is where most of the surprise lives.

At a subscription services business, the first churn-save motion went out against roughly a hundred accounts. Working through the activity report afterward, about a fifth of them needed a completely different play. The account had been sold and the new owner was a live prospect nobody had contacted. The business had closed but the operator was running something else. The account was being served at a corporate level and was never independently churned in the first place. One had become a different business entirely.

Three rules came out of that, and they are the ones I would put on the wall.

Sold is not dead. When an account has changed hands, the motion is not a save, it is a new-logo call to the buyer. One of those accounts turned out to have a new owner already buying heavily under a different number.

Closed is not dead. The operator usually moved somewhere. Finding out where is a ten-minute exercise that can produce a warm introduction.

Dark is not dead. In the same business, close to a quarter of all accounts went quiet for three months or more and came back on their own. A 90-day inactivity definition does not measure churn. It measures a gap. Treat it as churn and you spend save-motion budget on customers who were never leaving, while the ones who actually left look identical in the report.

And the discipline underneath all three: "probably closed" after one unreturned voicemail is not a disposition. It is a rep guessing, written into your system of record as fact.

Nobody Knows Why They Left

Here is the part that makes this uncomfortable, and it showed up in each of these businesses.

At the distributor, across every closed opportunity in the system, well over a hundred thousand of them, there were no reason codes. None. Competitive intelligence was, in the assessment's own words, essentially unavailable. When we built the list of churn drivers, it had to be written as hypotheses with an explicit data-quality flag on it, because the honest answer to "why did these accounts leave" was that nobody had ever recorded it.

This matters more than it sounds. Without reason codes it is hard to segment the dormant base by cause, which makes it hard to match an offer to a cause, which tends to leave the re-engagement generic, and that is a large part of why win-back campaigns underperform and get abandoned.

So the reason-code fix is not really a reporting project. It is what can make the second year of this motion better than the first. Build it into the stage definitions now, as a required field on Closed Lost and on any dead-lead disposition, and the data starts accumulating from the day you turn the motion on.

What Re-Engagement Actually Looks Like

The mechanism looked much the same across these businesses, and it is not a discount.

It is an offer of tangible value, meaning something worth having that requires nothing from them: an audit, an assessment, a pricing review, a benchmark against what comparable operations are paying or achieving. It tends to work because it gives the dormant customer a reason to take the call that is not "we want you back," and it gives you a diagnostic conversation instead of a negotiation.

A discount tends to do the opposite. It suggests the relationship is worth less than it was, it trains the ones who come back to wait for the next one, and it tells you little about why they left.

Where This Lands in a Value Creation Plan

If you are underwriting organic growth, this is the line item to look for and it is usually missing.

Size the dormant base before you fund a net-new motion. Segment it by how long it has been quiet and by what a customer in that band would plausibly buy next. Clean the list before anyone calls, because a meaningful share of it is likely to need a different play. And make the first conversation a diagnostic, not an offer.

One more thing worth saying plainly, because it decides who runs this. The dormant base is not a sales project. Sizing and segmenting it is analysis. The offer and the sequence are marketing. Working the named accounts is sales. And serving the long tail of small dormant accounts at a cost that makes sense usually needs a digital channel, where the company has one. Hand the whole thing to any single function and it becomes that function's side project, which is roughly how it ends up unsized in the first place.

It tends to be the cheapest growth available, and in the handful of companies I have measured it in, it was also the largest.

Like what you are reading? Let’s talk.

If this reflects the kind of work you need, it might be time to bring in a leader who has done it before.