Last updated: August 3, 2026
Your revenue report says one thing. Your bank balance says another. The gap between them has a name: days sales outstanding.
DSO is the average number of days it takes to turn a credit sale into cash in the account. It is one of the few finance metrics that is simultaneously a cash-flow number, an operations number and a customer-relationship number, which is exactly why it is so often left to drift. Sales owns the deal, finance owns the invoice, and nobody really owns the thirty days in between.
Global average B2B DSO sits at roughly 45 days, against terms that are usually Net 30. That two-week gap is not a rounding error. For a company doing 50 million in annual revenue, every 10 days of DSO represents about 1.37 million dollars in working capital you have already earned and cannot yet spend.
This playbook covers what DSO actually measures, what good looks like in your industry, why the number drifts above your stated terms, and which levers move it most.
The formula is simple enough to run on a napkin:
DSO = (Accounts Receivable / Total Credit Sales) x Days in Period
Take 200,000 dollars sitting in receivables against 1,500,000 dollars of credit sales in a 90-day quarter. That is (200,000 / 1,500,000) x 90, or 12 days. Low DSO means fast cash conversion. High DSO means capital parked in unpaid invoices instead of your operating account.
What the raw number does not tell you is whether 12 days, or 42, or 62, is good. That depends entirely on the terms you offered in the first place.
Divide your actual DSO by your average payment terms and you get a DSO efficiency ratio, which is the single most useful diagnostic in receivables. A ratio between 1.0 and 1.15 is excellent, 1.15 to 1.30 is acceptable, and anything above 1.50 is a compounding cash-flow problem.
The reason this framing beats the raw number is that it separates two very different diseases. If your DSO is high but your ratio is near 1.0, your terms are the problem and that is a commercial decision. If your ratio is 1.6, your customers agreed to pay you and are not doing it, and that is an operations problem you can fix this quarter.
DSO varies more across industries than almost any other finance metric, because the structure of how work gets billed varies. Construction retainage and healthcare claim adjudication are not collections failures; they are features of those markets. Here is where the averages land:
| Industry | Typical DSO | What drives it |
|---|---|---|
| Retail / e-commerce | 5–20 days | Card settlement, not customer behaviour |
| Wholesale distribution | 30–50 days | Standard trade terms |
| SaaS | 30–45 days | Recurring billing, Net 30–45 enterprise contracts |
| Professional services | 30–60 days | Milestones, retainers and hourly work mixed together |
| Manufacturing | 45–60 days | Milestone-based invoicing |
| Healthcare | 45–70 days | Claim adjudication and denials |
| Construction | 60–90+ days | Retainage and multi-tier approvals |
Source: Credit Pulse DSO benchmarks by industry. One adjustment worth making: the same research finds that recessions push DSO up 15–25% across industries, with B2B sectors hit harder. If you are benchmarking during a downturn, comparing yourself to normal-condition averages will make a macro problem look like a team problem.
Almost every case of runaway DSO traces back to three causes, and only one of them is the customer.
Every disputed invoice restarts the payment clock. At a 3–5% error rate you add 5 to 10 days per disputed invoice, and above a 10% error rate you add more than twenty. The unglamorous fix is the highest-leverage one: get invoices out within 24 hours of delivery, with the correct PO number and a named AP contact. That alone is worth 5 to 8 days.
This is where most receivables value quietly leaks. Collection success is brutally time-sensitive: contact within 24 hours of a missed payment succeeds around 65% of the time, dropping to 45% at three days, 30% at seven days and 15% once you pass two weeks.
Almost no manual AR team can hit a 24-hour window consistently across a full ledger, because the accounts that need chasing are scattered and the work is unrewarding. That is precisely why automated payment reminders outperform manual follow-up by 12 to 18 days on average. The advantage is not cleverness, it is consistency.
Most companies skip screening until bad debt forces the conversation. But payment behaviour is predictable: customers in the fair credit band pay 15–25% past terms, and those below it run 40–60% over. A modest upfront credit check prevents a large share of write-offs later, and it also tells your collections team where to spend its attention.
Ranked roughly by days of DSO removed per unit of effort:
The uncomfortable part is that levers 2 and 3 are where the days are, and both are pure execution volume. A team of three cannot personally reach every account within 48 hours of every missed payment, every week, in the customer’s language and preferred channel. This is the specific gap AI agents close well. Rio, Darwin AI’s collections agent, works the full aging ledger on a fixed cadence across WhatsApp, email and voice, escalating only the accounts that need a human negotiation, so the 48-hour window stops being aspirational. If you are evaluating that shift more broadly, our guide to AI in accounts receivable walks through the operating model, and there is a separate breakdown of how voice agents handle collections calls for ledgers where email goes unread.
One caution worth stating plainly: automating collections badly is worse than doing it slowly. Aggressive, tone-deaf dunning recovers this quarter’s invoice and loses next year’s renewal. The discipline of raising recovery rates without damaging the relationship is a design constraint, not an afterthought.
DSO is one leg of the cash conversion cycle:
CCC = Days Inventory Outstanding + DSO − Days Payable Outstanding
Because DSO enters that equation directly, every day you remove is a day of working capital returned. At 50 million dollars of revenue, ten days of DSO is roughly 1.37 million dollars unlocked without signing a single new customer or raising a single price. At 10 million in revenue it is still north of 270,000 dollars.
That is the argument to bring to a CFO. Reducing DSO is not a cost-savings project competing with headcount; it is a source of non-dilutive capital that also happens to reduce bad debt. And the broader trend supports acting now rather than later: around 70% of companies had automated some part of their AR process by the end of 2025, which means the performance gap between automated and manual receivables teams is widening, not closing.
Collect on time without chasing every invoice by hand.
Rio works your entire aging ledger on a fixed cadence across WhatsApp, email and voice, and escalates only what needs you.
See how Rio reduces DSOOne that sits close to your own payment terms. Divide DSO by average terms: 1.0 to 1.15 is excellent and above 1.50 signals a real cash-flow problem. A SaaS company at 40 days on Net 30 terms is performing well; a wholesaler at 40 days on Net 15 is not.
Late-payment counts tell you something went wrong. DSO tells you the rate at which the business converts revenue into cash, which is a number you can benchmark, target and hold someone accountable for. Both matter, but only one is a management metric.
Faster than most teams expect, because the biggest levers are process rather than product. Invoicing within 24 hours and running a fixed reminder cadence can move the number inside one or two billing cycles; automated follow-up collects 12 to 18 days faster than manual chasing. Credit-policy changes take longer because they only affect new customers.
Chasing harder does. Chasing earlier and more consistently usually does not. A polite reminder on day one of delinquency reads as competence; a legal-tone letter on day forty reads as a rupture. The relationship risk lives in tone and timing, not in frequency.
Higher, structurally. Construction runs 60 to 90-plus days because of retainage and layered approvals, and healthcare 45 to 70 days because of claim adjudication. Benchmark against your sector, not the global average, or you will chase a number your market does not allow.