Collections Automation and DSO Reduction: Method and Architecture
65 days. That is the average DSO for French mid-market companies in 2025 (source: Agicap Data Pulse, April 2026). Translated into cash terms: for a company carrying €1 million in receivables, every unnecessary day of DSO represents €2,750 of cash that isn't available to operate the business. Bringing DSO from 65 to 55 days frees up €27,500, without additional revenue, without renegotiating credit lines, without external financing.
This isn't a theoretical calculation. It's a process mechanic. And that's precisely where most Finance teams get stuck: DSO is treated as a commercial discipline problem ("sales needs to negotiate better payment terms") or an accounting rigour problem ("we need to chase faster"). Both levers are real. Neither is sufficient as long as the underlying collections process remains unstructured.
Why Spreadsheets Break Down Past a Certain Threshold
68% of French SMEs and mid-market companies still manage their collections in Excel (source: PwC, cited by LexRecouv). Below €200,000 in outstanding receivables or 500 active invoices, that choice is defensible. Beyond those thresholds, it becomes a structural constraint, for three concrete reasons.
Visibility degrades with volume. A spreadsheet tracking 800 open invoices, with columns for due dates, reminder history and manual notes, is unmanageable without a flawless update discipline. And that update discipline consumes exactly the time you're trying to save.
Reactivity is structurally limited. An accountant managing collections alongside other responsibilities checks the spreadsheet two or three times a week. A customer who paid Friday afternoon may still receive a reminder Monday morning if no one has updated the file. This kind of error is common. It carries a real relational cost with strategic accounts.
Escalation is manual by default. When an invoice crosses 45 days overdue, someone needs to decide whether to escalate to the CFO, the account manager, or legal. That decision, made invoice by invoice, is inconsistent from one month to the next and time-consuming at scale.
The Three Gaps That Standard Reminder Tools Don't Close
Automated reminder tools exist: built into ERPs or sold as add-on modules. They solve the volume problem well: automatically sending an email at D+5, D+15 and D+30 past the due date. But they fall short on three dimensions that actually determine whether collections performance improves.
Tone that doesn't fit strategic accounts. A mass reminder tool sends the same template to a customer representing €400,000 in annual revenue and to a one-off client at €3,000. An automated nudge at D+5 to a long-standing account, where payment delays typically reflect an internal approval process rather than bad faith, can damage a relationship that took years to build. Effective collections distinguishes customer segments and adjusts register accordingly.
Chasing customers who have already paid. This is the most common mistake in automated collections, and the most damaging to credibility. It happens when data isn't synchronised in real time between billing, CRM and the collections tool. A payment received Friday doesn't update the invoice status in the reminder system before Monday, and the automated email goes out anyway. This generates immediate, unnecessary friction.
No contextual escalation. A fixed-cadence reminder tool sends progressively firmer emails on a preset schedule. It cannot identify that an invoice is blocked because a credit note is pending on the seller's side, that the customer disputes a line item, or that the usual contact is out of the office. Escalating to the right person, at the right moment, with the right context remains a human judgment that standard tools don't support.
The Smart Automation Architecture: Tiers, Context, CRM Sync
An effective automated collections process isn't a sequence of scheduled emails. It's a conditional workflow that distinguishes situations and routes them accordingly.
The tiered structure remains the right foundation, but each tier must be conditional on current invoice status.
D+3 after the due date: a courteous reminder, neutral tone, short format. The message doesn't presuppose a problem; it simply flags the missed due date. It sends automatically if and only if the invoice status is still "open" in the billing system and not flagged as "disputed" or "credit note pending."
D+10: formal follow-up. Clearer tone, explicit reference to the outstanding amount and due date. The message includes a direct payment link where infrastructure allows. At this stage the workflow also checks whether a partial payment has been received since D+3: if so, the message reflects that rather than ignoring it.
D+20: manager escalation. The reminder no longer comes from accounting but from the account manager or CFO, depending on the customer segment. The change of sender is documented as an effective response lever (source: Agicap Data Pulse, 2026). At this tier, an internal alert is simultaneously generated for the account manager responsible for the relationship.
Beyond D+30: human judgment takes over. Possible referral to collections agency, negotiating an instalment plan, or initiating legal proceedings. The tool compiles a complete case file (communication history, exact outstanding amount, status of any disputes) and escalates it to the right person.
The condition running across all tiers: automatic status update in the CRM at every event. A received payment stops the reminder sequence immediately. A logged dispute routes the invoice to a separate workflow. An invoice whose status is inconsistent between billing and CRM triggers an alert rather than a reminder.
What 10 Days Less DSO Actually Changes
DSO reduction has two direct effects that aren't equivalent in importance.
The first is most commonly cited: recovering late payment interest and charging penalty fees. These amounts are real but rarely material for a mid-market company.
The second is structurally more significant: reducing the working capital requirement and improving cash flow predictability. Moving from 65 to 55 days of DSO on €1 million in receivables permanently frees up €27,500 of cash. This isn't a one-time gain: it's a permanent improvement in the baseline cash position, reducing dependence on overdraft facilities, improving the ratios presented to lenders, and creating room for investment decisions that would otherwise require external financing.
Predictability is the other benefit that tends to be undervalued. An automated collections process produces data: response rate by tier, average time to payment after the first reminder, resolution rate before D+20. These metrics enable 30- to 60-day cash flow forecasts that are significantly more reliable than anything producible from a manually updated spreadsheet. For the CFO, that forecast reliability is often worth more than the interest income recovered from late payers.
How to Assess Whether Your Process Is Ready to Automate
Before committing to automation, three questions reveal the real state of the existing process.
What is the single source of truth for invoice status? If the answer is "a spreadsheet maintained by one person," automation will start by solving a data problem, not a collections problem. An automated workflow can't be more reliable than the data it consumes.
How many disputed or contested invoices are currently tracked in the same tool as invoices that are simply overdue? These two populations need radically different treatment. Mixing them into the same automated reminder sequence produces inappropriate reminders and creates friction with customers who have a legitimate reason to withhold payment.
Are the billing system and CRM synchronised in real time? If a payment received in the billing system doesn't update the CRM status within 24 hours, collections automation will generate errors that your customers will see. Data synchronisation is a prerequisite, not a nice-to-have.
DSO is not a metric you improve through effort and discipline alone. It is the mechanical output of a structured process, or the absence of one. Collections automation isn't about sending more reminders. It's about sending the right reminder at the right moment with the right tone, and stopping automatically when the situation no longer warrants it. That precision, not volume, is what drives DSO down and keeps it there.