DSO – The Metric Everyone Reports, and Almost Nobody Moves

By Elaine Nowak, Global VP of Product Marketing at Sidetrade 23 September 2026

Ask any credit or collections professional what their most-watched metric is, and the answer is almost always the same: Days Sales Outstanding. It is the number that tells a CFO how long the business waits to collect cash it has already earned, and it appears in every treasury review and board-level cash conversation.

Yet for most organizations, DSO barely moves year over year. Teams add headcount, run dunning campaigns, and invest in invoice automation, and the number still does not respond as well as expected.  The Hackett Group’s 2025 U.S. Working Capital Survey found $1.7 trillion trapped in excess working capital across the top 1,000 U.S. public companies. Accounts receivable was the largest contributor, an estimated $600 billion opportunity, as the report showed DSO degraded for a second consecutive year. ^[1]

This article breaks down why DSO stalls, and the five levers using AI that actually move it.

The Formula and What It Hides

The standard calculation is simple:

DSO = (Accounts Receivable Balance / Total Credit Sales) x Number of Days.

A company carrying $18 million in receivables against $120 million in annual credit sales lands at roughly 55 days, and against Net 30 terms that gap represents about $8.2 million in working capital sitting where it should not be.

The formula gives a number, not a reason. It averages across customers who may behave very differently, and it is a lagging indicator, reflecting invoices issued 6-10 weeks ago. It also conflates collection performance with invoice quality, so a 15% dispute rate driven by billing errors will elevate DSO no matter how well or how much the collectors work. Invoice errors that trigger disputes are a consistently cited driver of elevated DSO and Credit Pulse puts the broad industry median at 40.5 days against a 59-day global average. ^[2]

How AI Helps Reduce DSO at the Source

Modern AI systems connected to an ERP deconstruct DSO in real-time by customer segment, invoice type, product line, dispute category, and collector. Instead of a single number that averages the signal, a breakdown is provided that shows exactly which accounts, processes, and invoice types are driving the gap. That is the diagnostic layer that most AR teams currently build manually in spreadsheets, if they build it at all.

The Five Levers That Actually Reduce DSO

DSO is an output. It responds to changes in the inputs that determine how quickly invoices move from issued to paid. Here are five levers that drive the most movement, in the order that matters for implementation.

Lever 1: Invoice Accuracy at the Point of Delivery

An invoice with the wrong PO number, a missing contract reference, or a price error will not be paid on time, and no amount of collection effort compensates for it.

AI systems validate invoices against contract terms, purchase orders, and customer-specific billing requirements before delivery, flagging discrepancies for correction. Some identify invoice characteristics that have historically resulted in a dispute in a particular customer segment, flagging them proactively. Reducing dispute volume at the source delivers the highest-leverage DSO improvement available, and it moves the number without adding any collection effort at all.

Lever 2: Collection Timing and Prioritization

Most collections efforts concentrate on the oldest, most overdue accounts. That is the right place for recoveries, but the wrong place for DSO reduction, since the damage is already done by day 60.

AI agents run proactive, personalized outreach across the full portfolio before invoices are even due, and prioritization models rank the daily worklist by payment risk rather than invoice age. They operate continuously, including outside business hours, which matters because a payment reminder sent early in the morning on the due date reaches a customer’s AP team at the start of their processing day rather than being buried in afternoon correspondence.

Lever 3: Dispute Resolution Velocity

Disputes stop the payment clock, and a resolution routinely crosses AR, billing, sales, and the customer’s AP team with no single owner. In most organizations, dispute resolution is the single largest controllable driver of extended payment cycles. There is rarely a single owner with end-to-end accountability, and resolution service level agreements (SLAs) are often defined but not enforced.

Dispute resolution is a collection activity. Dispute prevention is a DSO strategy.

AI systems categorize disputes at the point of receipt, route them to the correct owner immediately, and track resolution time against defined SLAs with automatic escalation when timelines are breached. More strategically, AI surfaces root cause patterns across the dispute portfolio, identifying the billing errors, contract mismatches, or process failures that are generating the highest dispute volumes. Fixing those upstream removes the disputes from the pipeline entirely.

Lever 4: Cash Application Speed and Accuracy

Cash received but not yet applied keeps invoices open, inflates DSO, and triggers collection activity against customers who have already paid, wasting collector time and creating friction with customers who have done nothing wrong.

AI trained on remittance formats and payment patterns automates most matching, applying correct payments instantly and surfacing genuine exceptions, often the fastest DSO improvement available for high-volume organizations. The effect on reported DSO would be immediate: the receivables balance begins to reflect actual open items rather than items awaiting processing, and collection activity focuses on invoices that are genuinely unpaid.

Lever 5: Credit Monitoring and Early Intervention

A customer drifting from 28-day to 42-day payment behavior is showing an early signal that annual credit reviews are not built to catch. When that signal goes unread, the account continues to be treated as a standard performer. Credit terms remain unchanged, and collection priority remains low. By the time DSO reflects the deterioration, the gap is already substantial.

AI systems monitor payment behavior across the entire customer portfolio and flag accounts where patterns are changing in real time. That flag triggers a credit reassessment, a change in collection priority, or a proactive conversation with the customer before the payment behavior becomes entrenched. The organizations that achieve sustainable DSO reduction are the ones that close the loop between credit management and collections, and continuous AI monitoring is what makes that operationally feasible at enterprise scale.

Why Traditional DSO Reduction Methods Fails

Traditional DSO reduction methods stall for three consistent reasons:

• They measure activity, call volumes, and email counts, rather than outcomes such as dispute resolution time and days to first contact
• They automate a broken process, which just executes bad sequencing faster
• They treat DSO as a collections problem instead of an end-to-end challenge that spans billing, credit, and finance operations, holding AR accountable for outcomes it cannot fully control

What Meaningful DSO Reduction Actually Looks Like

Organizations that achieve sustained, double-digit DSO reductions share four practices:

• Diagnose the gap first, rather than launching a reduction program before knowing what is driving DSO.
• Sequence invoice accuracy and cash application ahead of collection workflow redesign, since those levers carry the least implementation risk.
• Track leading indicators rather than waiting on DSO itself, which only reflects results six to ten weeks after the fact.
• Use AI as the execution layer across all five levers, not as a replacement for collections judgment.

The formula is simple. The work of moving it is specific and sequenced. Get the five levers right, in order, and the DSO number moves on its own.

This blog covers the diagnostic and the five levers at a summary level. A full breakdown, with the complete playbook for each lever, publishes soon as an article in the Credit Research Foundation Journal.

Sources

Elaine Nowak, Global VP of Product Marketing at Sidetrade

Elaine focuses on crafting compelling value propositions and targeted content that highlights the transformative potential of agentic AI. She collaborates closely with thought leaders, industry experts, and analysts to effectively communicate the benefits and value of these innovations.

With more than twenty-five years of educational content development, messaging, positioning, and thought leadership, she works cross-functionally to ensure Sidetrade evolves with market needs across the order to cash cycle to deliver the greatest impact to finance professionals.

FAQ