Sending the same five-day reminder sequence to every overdue invoice isn't a strategy. It's a hope. Here's why segmented dunning changes the math.
The Uniformity Problem
Most mid-market AR teams run some version of the same dunning sequence: a polite reminder at net+5, a firmer note at net+15, a phone call at net+30, escalation at net+45. The sequence is logical. The problem is that it applies the same logic to every customer in your portfolio — regardless of how they've actually paid in the past, what their relationship looks like with your sales team, or what their current cash position might signal.
Think about what's actually in a typical mid-market AR portfolio. You have a customer who's been with you for six years and pays 3-5 days late on every invoice without fail — but always pays. You have a new customer who just placed their first $85,000 order and is 12 days overdue with no prior history. You have a customer whose average payment time jumped from 28 days to 58 days over the last two quarters. And you have a spot buyer who purchased once and owes $4,200.
These are four fundamentally different situations. A dunning sequence designed for the average of them serves none of them well.
What Uniform Dunning Actually Costs
The cost of uniform dunning isn't obvious on any single invoice. It accumulates across your portfolio over months.
When you send an aggressive follow-up to the six-year customer who always pays slightly late, you risk damaging a relationship that's worth far more than any individual invoice. AR managers at mid-market distributors and manufacturers tell us this is the failure they feel most acutely — collections behavior that looks efficient on a spreadsheet but quietly erodes trust with accounts that have real long-term value.
When you send a gentle five-day reminder to the customer whose payment timing has been steadily deteriorating, you lose 10 days you could have used to get ahead of a potential write-off. The polite email that goes unanswered at net+5 becomes a harder conversation at net+60 — and by that point, your leverage has diminished considerably.
When you route both the $4,200 spot buyer and the $85,000 new account through the same sequence, you're spending roughly the same AR effort on both. The economics of that don't add up.
Why Segmentation Is Harder Than It Sounds
If the solution is obvious — segment your customers and tailor outreach accordingly — why do most mid-market AR teams still run uniform sequences?
Part of the answer is tooling. Standard AR modules in ERP systems like NetSuite, QuickBooks, and Sage Intacct generate aging reports and offer basic reminder scheduling. They don't easily support branching dunning logic based on customer payment history. Building custom segment logic on top of these systems requires either significant configuration effort or manual AR specialist judgment on every account.
Part of the answer is data. To segment meaningfully, you need a reliable view of each customer's payment pattern — not just their current balance, but their average days-to-pay, their variance over time, their invoice size relative to their typical purchase behavior. Most mid-market companies have this data scattered across ERP transaction records, but assembling it into a working segmentation model takes time that AR teams don't have.
And part of the answer is cognitive load. An AR specialist managing 200-300 open invoices can't manually assess each customer's payment profile before deciding how to route the next reminder. The practical default is: apply the standard sequence, flag anything egregiously overdue for individual attention.
This is not a criticism of AR teams — it's a description of a resource constraint that uniform tooling creates.
Four Segments Worth Building
We're not saying you need 12 customer segments and a different sequence for each one. That creates its own management overhead. But four segments covers most of the meaningful variation in a mid-market B2B AR portfolio:
Reliable-late payers — customers with a consistent history of paying 5-20 days beyond terms, but who pay. These accounts need lighter-touch dunning, earlier in the cycle. Sending them the standard aggressive net+30 sequence doesn't accelerate payment; it just creates friction. A pre-due reminder at net+0 and a brief follow-up at net+10 is often enough, because they're already planning to pay — they just have their own internal payment schedule.
New accounts with no history — first or second invoice from a customer with no established payment pattern. These need more attentive follow-up, not because they're high-risk but because you have no data yet. A quicker escalation to phone contact (net+15 rather than net+30) helps establish the relationship and gather early payment behavior data faster.
Deteriorating accounts — customers whose average days-to-pay has increased meaningfully over the last 2-3 invoice cycles. This is the segment that most uniform dunning sequences miss entirely, because they look at current aging rather than trend. A customer who paid at net+22 for three years but is now sitting at net+38 on two consecutive invoices is telling you something. This segment warrants a proactive outreach call before the invoice is overdue — not a reminder email after.
Low-value, low-relationship accounts — spot buyers or occasional customers where the cost of a long collections process may exceed the value of maintaining the relationship carefully. These are candidates for a shorter, more direct sequence that prioritizes resolution over relationship management.
The Data You Already Have
Building these four segments doesn't require external data sources. The payment history sitting in your ERP transaction records contains most of what you need: average days-to-pay per customer over 12-18 months, invoice volume and size history, recent trend in payment timing, and account age.
The challenge is extracting and operationalizing it. An AR aging report shows you current balance by bucket. It doesn't show you that a customer's average days-to-pay has moved from 31 to 47 over the last six invoices, or that three of your largest open invoices belong to customers whose payment timing has been accelerating in the wrong direction.
Consider a mid-size Midwest industrial supplier with roughly $18M in annual B2B revenue and about 140 active customer accounts. Their AR team of two specialists runs a single dunning sequence — email at net+7, email at net+21, call at net+35. About 60% of their open invoice balance at any given time is in the 30-60 day bucket, and they write off roughly 1.2% of AR annually to bad debt. When we look at their payment data, about 35% of that 30-60 day balance belongs to reliable-late payers who would have paid regardless — but who now receive the same treatment as the customers who are genuinely at risk. The AR team's time is distributed by aging bucket, not by risk profile.
Reorienting around payment behavior segments rather than aging buckets doesn't eliminate write-offs, but it concentrates AR effort where it actually influences outcomes — on the accounts that are genuinely at risk, earlier in the payment cycle.
The Limit of This Argument
We're not saying segmentation solves every AR problem. Customers who can't pay won't pay faster because you sent them a better-targeted email. A customer going through a genuine cash crisis needs a different kind of conversation — and often a payment plan or dispute resolution process — that no dunning sequence automates well.
The claim is more specific: for the majority of your AR portfolio, where the question isn't "will they pay" but "when will they pay and how much attention does this require," segmented dunning consistently outperforms uniform sequences. It concentrates effort, reduces friction with good accounts, and improves the signal-to-noise ratio for AR specialists trying to triage their workload.
Getting from a single sequence to a segmented model requires two things most teams struggle with: a reliable view of customer payment history and a way to act on it without adding hours to the AR specialist's day. That's the problem we're building toward at Cashvyne — and why the behavioral segmentation question is central to how we think about dunning, not peripheral to it.