AR & Collections

Accounts receivable is the money customers owe a business for goods or services already delivered on credit — a current asset on the balance sheet until it's collected. AR and collections, as a practice, is everything a finance team does to turn that asset into cash on time: issuing invoices, tracking which are open and how overdue, deciding which to chase first, applying incoming payments against them, and eventually writing off the small percentage that never gets collected. This cluster covers that whole lifecycle, but it leans hardest on the two places most teams actually lose time and money — the aging report, which is the single tool almost every AR decision runs through, and days sales outstanding, the metric that most directly reflects whether collections is actually working.

The AR lifecycle, end to end

1. Invoice — an invoice is issued once the underlying good or service has been delivered, starting the clock on payment terms (net 30, net 60, and so on).
2. Age — every open invoice moves through aging buckets (current, 1 to 30 days past due, 31 to 60, 61 to 90, over 90) as time passes without payment, and the aging report is the single view that shows the whole portfolio's health at once.
3. Chase — collections outreach (reminders, calls, dunning sequences) targets whichever overdue invoices are prioritized, using some mix of amount, days overdue, and customer payment history.
4. Apply — an incoming payment is matched to the invoice or invoices it settles and the balance is closed, partially or in full.
5. Resolve — anything that stays unpaid past a threshold gets escalated, disputed, or eventually reserved and written off as bad debt.

Cash application — the mechanics of matching a payment to an invoice — is closely related to AR but is covered in depth in its own cluster; this one focuses on everything upstream and downstream of that single step: getting the invoice out, tracking it, deciding who to chase, and reading the aging report correctly once you have one.

Why AR data goes wrong

An aging report is only as reliable as the customer and invoice data feeding it. The most common ways it silently goes wrong: a customer record update (a new billing address, a merged duplicate account) that doesn't sync between the CRM and the accounting system, leaving invoices attributed to a stale record; a payment applied to the wrong invoice, which makes one account look overdue and another look paid when neither is true; and a credit memo or dispute that's tracked in a support tool but never reflected in the AR subledger, so the "overdue" balance includes an amount the customer has already disputed and won't be paying as invoiced.

Typical DSO ranges by business model

Days sales outstanding varies enormously by how a business actually sells, which is why a single "good DSO" benchmark is close to meaningless without knowing the underlying billing model:

The number that actually matters isn't how a company's DSO compares to a generic industry average — it's how a given period's DSO compares to that same company's own trailing average. A DSO that's climbing month over month, at a stable customer mix and stable payment terms, is the earliest reliable signal that either collections effort has slipped or something upstream (invoicing accuracy, aging report data quality) has started to break.

Reading an aging report: a worked example

An aging report lists every open invoice with the customer, amount, due date, and which bucket it falls into. Say a mid-market SaaS company's report shows $400,000 total AR: $250,000 current, $80,000 in 1-30 days past due, $40,000 in 31-60, $20,000 in 61-90, and $10,000 over 90 days. Read in isolation, that looks fine — 62.5 percent is current, and the 90-plus bucket is a small fraction of the total. The number that actually matters is how that distribution compares to the same report three months ago. If the 61-90 and 90-plus buckets have been growing month over month while current and 1-30 stay flat, that's a customer or process problem developing well before it would show up in a single month's DSO figure, because DSO is a blended average that can mask a small, worsening pocket of genuinely uncollectible accounts sitting inside an otherwise healthy portfolio.

The other thing worth checking every time: whether the total on the aging report actually ties to the AR balance on the general ledger. When it doesn't — and this is one of the more common diagnostic questions in this cluster — the mismatch is almost always a payment applied to the wrong invoice, a credit memo that posted to the GL but never updated the aging source, or a customer record that didn't sync between whatever system generates invoices and whatever system the aging report is pulled from.

The upstream lever most teams underuse: credit policy

Collections activity gets most of the attention because it's the most visible, active part of AR — but a meaningful share of collections problems are decided before an invoice is ever sent, by credit policy: what payment terms a customer is offered, what credit limit they're extended, and whether a deposit or prepayment is required before service starts. A customer with a history of slow payment who's offered the same net-60 terms as a reliable one will predictably show up in the aging report's older buckets regardless of how well collections chases them afterward. Teams that review and adjust credit terms based on actual payment history — shortening terms, requiring deposits, or tightening credit limits for repeat late payers — consistently see better aging distributions than teams that only ever act after an invoice is already overdue.

How teams decide who to chase first

With a finite amount of collections time and an aging report with dozens or hundreds of open lines, prioritization usually comes down to some combination of: dollar amount (chase the largest balances first, since they move the total AR number the most); days overdue (chase the oldest first, since collectibility drops the longer an invoice ages); and customer risk signal (a customer with a history of late payment, a support ticket about a dispute, or a recent change in company circumstances gets bumped up regardless of amount or age). Teams below AR automation maturity level 3 typically do this by eye, sorting a spreadsheet; more mature teams use a weighted score combining all three factors so the daily worklist is generated rather than assembled by hand each morning.

The metrics that matter beyond DSO

• Collection effectiveness index (CEI) — the percentage of collectible receivables actually collected in a period, which unlike DSO isn't distorted by revenue growth or seasonality.
• Aging bucket distribution — what share of total AR sits in each bucket. A healthy portfolio is weighted heavily toward current and 1-30; a growing 90-plus bucket is the clearest early warning of a collections or credit-policy problem.
• Bad debt as a percentage of revenue — the share of receivables that ultimately gets written off rather than collected, which reflects both credit policy and collections execution.
• First-response time on disputes — how quickly a customer's dispute or question about an invoice gets acknowledged, since an unaddressed dispute is one of the most common reasons a collectible invoice quietly ages into write-off territory.

When an overdue invoice becomes a write-off

Most invoices that age past 90 days still get collected eventually, which is why write-off is treated as a last resort rather than an automatic outcome of aging. A typical policy escalates in stages: past 60 days, a formal collections notice; past 90, involvement from a manager or account owner outside the collections team, since a relationship-level conversation often succeeds where an automated reminder hasn't; and only past a company-defined threshold — commonly 120 to 180 days, though this varies with deal size and customer relationship — does the balance move to a bad debt reserve and eventually get written off. Separately, a company sets a general bad-debt reserve based on historical collection experience across its whole portfolio, so that expected losses are reflected in the financials before any specific invoice is individually written off. Getting the aging report right matters here specifically because a reserve calculated against inaccurate aging data misstates the reserve itself, not just the individual invoice.

Why automation here is judged on relationships, not just speed

AR automation carries a specific risk that AP or bank reconciliation automation doesn't: a badly tuned dunning sequence damages a customer relationship in a way a badly tuned bank reconciliation never can. An automated reminder that fires after a customer has already paid, or escalates a dispute the customer already raised through a different channel, costs more in goodwill than the time it saves. The teams that automate AR successfully tend to automate the mechanical middle of the process first — aging, prioritization scoring, payment application — and keep a human in the loop on anything customer-facing until the underlying data (which invoices are genuinely open, which are disputed, which customers have a history worth treating differently) is reliable enough to trust with fewer manual checks.

Why AR automation and ERP integration are usually the same project

For any company running an ERP as its system of record — SAP, NetSuite, or similar — AR automation almost never means replacing the ERP's AR module. It means connecting a purpose-built automation layer (aging, prioritization, dunning, cash application) to the ERP's invoice and customer data, so the ERP stays the single source of truth for the ledger while the automation layer does the work of turning that data into a worklist and a set of outbound actions. This is why the integration question is so central to evaluating AR automation tools: a tool that can only read and write basic invoice status is far more limited than one that can also see credit terms, customer hierarchy, and dispute status, because prioritization and dunning both depend on that fuller picture, not just an open-or-closed flag on each invoice.

What's in this cluster

The pages below cover what accounts receivable and AR automation actually are; the specific, sourced diagnostics for integrating AR automation with ERPs like SAP and for automating collections without damaging customer relationships; the days-sales-outstanding questions practitioners ask most; and the aging-report questions — reading one, fixing one that's gone wrong, and forecasting collections from one — that sit underneath almost every other AR decision.