Insights

How to Reduce DSO: 12 AR Best Practices That Speed Up Cash Flow

15 min read
How to Reduce DSO: 12 AR Best Practices That Speed Up Cash Flow

Revenue can look healthy on paper while the bank balance tells a different story. A deal marked closed won becomes cash only when the customer pays, and every day between invoice and payment is working capital the business has earned and cannot use.

Days sales outstanding (DSO) is the number that measures that gap. This playbook covers 12 AR best practices that shorten it, organized by where they act in the order-to-cash cycle. Each one opens with a one-line takeaway you can hand to your team, followed by how to put it in place.

What is DSO, and why does a lower number matter?

DSO is the average number of days it takes to turn a sale on credit into cash. The standard formula is accounts receivable divided by credit sales for the period, multiplied by the number of days in the period.

Each day of DSO is roughly one day of revenue sitting on the balance sheet. For a company with $20M in annual credit sales, one day is about $55,000, so taking 33 days out of DSO would free up roughly $1.8M in working capital. That cash can fund hiring, reduce reliance on a credit line, or simply give the business room when a large customer pays late.

Tesorio customers see a 33 day average DSO reduction. It is an average across customers, and individual results depend on where a team starts: its terms, its customer mix, its invoice quality and how much of the process is still manual. The practices below are the ones behind that movement.

How far you have to go depends on your industry. In Tesorio's DSO benchmark report, drawn from more than billion in receivables, average days to collect runs from 26 days in logistics and supply chain to 48 in software, 60 in manufacturing and 65 in business services. The fastest industries collect 1.4x to 2.5x faster than the slowest. Speed and overdue exposure also diverge: financial services has just 11% of AR overdue, while energy and utilities has 50% past due.

What are AR best practices?

AR best practices are the habits and controls that move an invoice to cash with the fewest days and touches in between. They span the full cycle: extending credit, building and delivering the invoice, making payment easy, following up, applying cash and forecasting what comes next.

Most DSO problems start upstream of collections. A sales rep closes a deal and types the amount, billing address, purchase order number and AP contact into the CRM in a hurry. Lawyers on both sides spent weeks making the contract precise, yet the invoice gets built from the rushed CRM entry. When the PO number is wrong, the customer's AP team cannot pay it, and the delay looks like a collections problem when it began in billing.

That is why the 12 practices split into two groups. Six prevent delays before the due date. Six act on delays once they happen.

Matrix of 12 AR best practices: six that prevent delays before the due date and six that act on delays after it
The 12 practices in this playbook, split by where they act in the order-to-cash cycle.

How do you reduce DSO before an invoice is due?

1. Set credit policy and payment terms by risk

Takeaway: the cheapest overdue invoice to collect is the one you never extended on the wrong terms.

How to do it:

  • Use a standard credit application and a consistent evaluation of every new customer.
  • Set credit limits per account from financial data and payment history, and review them on a schedule.
  • Decide in advance when to require deposits, guarantees or letters of credit.
  • Match terms to risk. Offer an early payment discount such as 2/10 net 30 where margin allows, and apply late fees where contracts and law permit.
  • Revisit terms as a customer's payment behavior changes, in either direction.

2. Build invoices from the contract, and check them before they go out

Takeaway: a large share of late payments are self-inflicted by missing or incorrect invoice details.

Wrong PO numbers, outdated AP contacts and line items that do not match the purchase order all send an invoice to a queue where nobody can approve it. How to do it:

  • Pull billing details from the signed contract, which lawyers already checked.
  • Confirm every invoice carries the correct PO number, billing address, AP contact, terms, due date and tax treatment.
  • Deliver electronically, in the format the customer requires (PDF, EDI or direct submission to their AP portal).
  • Track delivery and opens, so a "we never got it" surfaces on day two instead of day forty.

3. Make paying you the easiest task on the customer's list

Takeaway: every extra step between an invoice and a payment adds days.

How to do it:

  • Accept ACH, wire, card and check, and put a pay link on every invoice.
  • Offer a self-service portal where customers can see open and past invoices, download statements and raise disputes.
  • Offer autopay for recurring billing.
  • Make your AR contact details visible on every invoice and statement.

4. Segment customers by value, risk and behavior

Takeaway: one cadence for every customer wastes collector time on accounts that would have paid anyway.

The old default was an ERP nudge that said "15 days late" to everyone. Segmentation replaces it with treatment that fits the account. How to do it:

  • Segment on revenue, payment history, industry risk, geography and strategic importance.
  • Automate low-value, low-risk accounts end to end, with standard terms and minimal manual touches.
  • Give high-value accounts a named owner, proactive contact and some flexibility on terms.
  • Put high-risk and chronic late payers on shorter terms, deposits, faster follow-up and more frequent review.

5. Start collections before the due date

Takeaway: the easiest collection happens before the invoice is overdue.

How to do it:

  • Send a reminder about a week before the due date, with the invoice and pay link attached.
  • For large invoices, confirm with the customer's AP team that the invoice is approved and scheduled.
  • Flag and resolve disputes, missing POs or approval blockers while there is still time before the due date.
  • Explain your billing process to new customers during onboarding, so the first invoice does not surprise anyone.

6. Give sales, customer success and finance one view of the account

Takeaway: collections move faster when the person with the relationship knows the invoice is late.

How to do it:

  • Share payment status and history with account owners inside the tools they already use, such as the CRM or Slack.
  • Log every customer conversation about billing in one place.
  • Define who owns what across AR, sales and customer success, and when an issue escalates from one to the next.
  • Alert the account owner automatically when a strategic customer slips past due.

How do you collect faster once an invoice is overdue?

7. Rank the collections queue by likelihood to slip

Takeaway: collectors should start each day on the accounts most likely to pay late.

Rules-based dunning sends the same sequence on the same schedule. The shift under way now is toward software that decides: it ranks the book by how likely each invoice is to slip, drafts outreach shaped by how that customer has paid before, and revises the ranking as behavior changes. How to do it:

  • Automate reminders for low-value accounts and keep a personal touch on high-value ones.
  • Write different sequences for 30, 60, 90 and 120+ days past due.
  • Use payment history to predict pay dates, and prioritize the accounts where a call changes the outcome.
  • Connect collections to your ERP, CRM and email, so outreach reflects the current balance and conversation.

Tesorio customers see 3x collector productivity, because collector time goes to the accounts that need a person.

Veeva Systems automated collections for accounts under $1,500 and let prioritization decide where collectors spent their day. Time spent on low-priority accounts fell from a full day a week to under two hours. The result, per Veeva's case study: 90-day aged receivables cut in half, and 75% of the team's time freed from manual collections tasks.

8. Use a clear escalation ladder, including credit holds

Takeaway: customers pay the vendors whose consequences are predictable.

How to do it:

  • 1 to 30 days overdue: automated reminders plus direct contact with AP.
  • 31 to 60 days overdue: more frequent outreach, account owner involved, credit hold on new orders.
  • 61 days and over: leadership review and a decision on third-party recovery.
  • Write the ladder down, share it with sales, and apply it consistently.

9. Apply cash the day it arrives

Takeaway: an unapplied payment keeps a paid invoice open, and a paid customer keeps getting reminders.

How to do it:

  • Match payments to invoices automatically, using remittance data from email, bank files and portals.
  • Route unmatched payments to a short exception workflow with a clear owner.
  • Apply same day where possible, and add controls to catch misapplied payments.
  • Sync applied cash back to the ERP and to the collections queue so nobody chases a paid invoice.

10. Forecast cash from invoice-level history

Takeaway: a forecast built from every invoice and payment is more reliable than one built from sales multiplied by DSO.

Applying an average DSO to revenue assumes every customer behaves alike. Looking at each customer's payment history on every past invoice, and predicting when each open invoice will be paid, gives a forecast that reflects your actual book. How to do it:

  • Bring receivables, payables and bank activity into one view.
  • Predict pay dates per customer and per invoice from past behavior.
  • Update the forecast each time payments land.
  • Model scenarios, such as a large customer paying 30 days late, to see the effect on cash before it happens.

Couchbase, whose revenue spans subscription, on-premise and cloud licensing, moved to this approach and now builds cash forecasts in hours where they previously took days of spreadsheet work. It also reduced DSO by 10 days, according to Couchbase's case study.

11. Track DSO alongside CEI and ADD

Takeaway: DSO alone can hide problems, so read it next to the metrics that explain it.

How to do it:

  • DSO: how many days of sales are still sitting in receivables.
  • Collection Effectiveness Index (CEI): how much of what was collectible in a period you actually collected.
  • Average Days Delinquent (ADD): how far past due the average account runs.
  • Bad debt percentage and cost per collection: what late payment costs once it becomes loss or labor.

Review them monthly by segment and by collector, and use them to find the bottleneck, whether it sits in billing, disputes or follow-up.

12. Build controls, and train the team that runs them

Takeaway: the practices above only hold if the process is documented and the people running it are trained.

How to do it:

  • Follow the rules that apply to you, such as fair collection practices, data protection and SOX controls, in every jurisdiction you bill.
  • Keep records of every collection activity, and review credit limits and terms periodically.
  • Watch customer concentration, and consider credit insurance for large receivables.
  • Train the team on your AR systems, customer communication and the metrics in practice 11, and cross-train so no critical task depends on one person.

What should the technology stack support?

None of these practices requires a specific product, but most get hard to sustain on spreadsheets and ERP reminders once volume grows. Whatever you use should:

  • Connect to your ERP (such as NetSuite, Sage Intacct, Workday or Oracle), your CRM (such as Salesforce or HubSpot), your bank feeds and your email and chat tools.
  • Handle invoice delivery, payment options, collections workflows and cash application on one record of the customer.
  • Report DSO, CEI and ADD in real time, by segment.

The distinction worth making is between a collections tool, which automates one step of the cycle, and an order-to-cash analyst, which works across the cycle from credit through forecast and coordinates sales, customer success and finance. A collections tool suits a team whose delays genuinely start and end in follow-up. If your delays start in credit, billing or cash application, a tool that covers only follow-up will leave those days in place.

How should you roll these practices out?

Phase 1: measure and plan (weeks 1 to 4)

  1. Calculate your current DSO, CEI and ADD, by segment.
  2. Map the process from contract to cash and mark where invoices stall.
  3. Audit your systems and integrations.
  4. Set specific targets for each metric.

Phase 2: quick wins (weeks 5 to 12)

  1. Fix invoice accuracy and delivery (practice 2).
  2. Review terms for new customers (practice 1).
  3. Turn on pre-due reminders and a basic dunning sequence (practices 5 and 7).
  4. Start monthly metric reporting (practice 11).

Phase 3: deeper changes (weeks 13 to 26)

  1. Segment customers and rank the collections queue (practices 4 and 7).
  2. Automate cash application (practice 9).
  3. Move forecasting to invoice-level history (practice 10).
  4. Formalize escalation, controls and training (practices 8 and 12).

Frequently asked questions

How much can AR best practices reduce DSO?

It depends on your starting point. Tesorio customers see a 33 day average DSO reduction. Teams with manual processes, inaccurate invoices or one cadence for every customer usually have the most room to move.

Which AR best practice has the fastest effect on DSO?

Invoice accuracy and pre-due reminders are usually the quickest, because they remove delays that were never about the customer's willingness to pay.

What is a good DSO?

A good DSO sits close to your standard payment terms. If most customers are on net 30 and your DSO is 60, roughly half of each invoice's life is time past due, and the practices above are where to look.

Is DSO the only AR metric that matters?

No single metric tells the whole story. Read DSO with CEI and ADD to see whether a change comes from collecting more effectively or from a shift in sales volume.

If you want to see what these practices look like running on your own receivables, see how Tesorio works on your data.

More from Insights