NetSuite DSO Reconciliation: Paid-Invoice Days vs Ratio

Executive Summary

"NetSuite DSO" (days sales outstanding) is not one metric. Oracle's Benchmark 360 documentation defines monthly DSO as the average days open of fully paid invoices for the month ^[1], and describes it as the average number of days customers take to pay invoices ^[1]. The finance-textbook ratio instead divides accounts receivable by sales and multiplies by the days in the period, using either an ending balance or the mean of beginning and ending balances ^[2]. The first measure looks backward at invoices that finished the payment cycle. The second measures a balance that includes every invoice still unpaid at the cut-off.

That difference explains most "why doesn't DSO match average days to pay" disputes. In the three fictional scenarios built for this report (all figures illustrative), the paid-invoice measure moves between 26.3 and 34.0 days while the ending-balance ratio moves between 30.0 and 53.3 days, and the two can move in opposite directions. A month in which only a few fast-paying invoices close can look healthy on the paid-invoice measure while unpaid balances grow.

The practical rule is to document the population and denominator before comparing any DSO figure. Public companies show how much room exists: filings define DSO on quarter-end receivables, on average receivables, on 91.25-day or 365-day conventions, and with or without unbilled amounts ^{[3] [4] [5]}. The Securities and Exchange Commission (SEC) expects a disclosed metric to come with a clear definition and its calculation ^[6], and the Credit Research Foundation (CRF) calls DSO one of the most often misused and misunderstood performance measures ^[7].

Recommended use, as of October 2026: use a balance-based DSO (ending or average receivables over matched-period credit sales) for board and working-capital reporting; use best possible DSO and average days delinquent for collections targets ^[7]; use paid-invoice days only as a lagging payment-behavior indicator, never as a substitute for the balance-sheet figure and never as a cash-release calculator. Targets should come from the company's own history and like-for-like peers: published sector figures use different definitions and samples ^{[8] [9]}.

Before sign-off, tie receivables to the general ledger, tie the sales denominator to a named report, tie the paid-invoice population to a workbook, disclose every exclusion, and save a frozen monthly evidence snapshot. Field names, eligibility rules and report filters described here need testing in the specific NetSuite account. Where invoice populations cannot be reconciled, a scoped reporting and configuration engagement is the sensible next step.

Introduction and Background

Controllers, accounts receivable (AR) managers, financial planning and analysis (FP&A) leads and chief financial officers (CFOs) at NetSuite-using scale-ups, software-as-a-service (SaaS) businesses and distributors regularly meet two DSO numbers that disagree. One comes from a NetSuite dashboard or saved search. The other comes from the trial balance and an income statement. Both are labeled "DSO," and both can be correct for the question each answers. NetSuite's own educational content defines DSO, also called debtor days, as the average number of days it takes a company to convert credit sales to cash ^[10].

This report is a definition-reconciliation reference. It is not a collections-automation guide, a payment-prediction article or a generic cash-conversion-cycle explainer. It covers five things: the primary-source definitions side by side; the invoice-level evidence needed to rebuild each figure; a fictional ledger that computes both from the same month; a sign-off routine; and guidance on choosing a governed target. The audience is US and Canadian English-speaking finance teams, and "as of" dates refer to October 2026.

Three structural facts frame the problem. First, Oracle's own documentation uses the DSO label for more than one construct: Benchmark 360 averages days open of fully paid invoices, while the Planning and Budgeting forecast form treats DSO as an input that converts credit sales per day into a closing receivables balance ^[11]. Second, public-company disclosures show no single convention. Third, the largest published benchmark programs define the metric differently from one another, so "the benchmark" depends on whose it is. Houseblend, a Montréal NetSuite consultancy, works on reporting and configuration problems of exactly this kind; this report keeps that perspective to the sections on implications and engagement scoping, and claims about third-party products and data cite independent sources.

Oracle documentation pages were re-read on October 27, 2026. Where documentation is silent (for example, on how partially paid invoices are treated in the Benchmark 360 DSO workbook), this report says so rather than inferring. Nothing here is accounting or audit advice; readers should consult their auditors and qualified accountants on policy choices that affect reported results.

Key Differences Between the Two DSO Definitions

Table 1 below sets the competing definitions side by side before the sections that follow walk through each difference in turn.

Table 1 shows that the measures differ on at least four independent dimensions: which invoices are in, how they are weighted, which clock is used, and what the denominator is. Only the first row is a property of closed invoices. The other rows are properties of a balance at a cut-off. A reconciliation that tries to bridge them with a single adjustment will fail; the bridge has to be built dimension by dimension.

Change 1: Population (closed invoices versus open balance)

Benchmark 360 draws only on invoices that reached full payment in the month. NetSuite itself marks an invoice as closed when it is paid in full but displays the status Paid in Full rather than Closed ^[15]. An invoice that is still open, or partly paid, has no place in a paid-invoice average until the final payment is applied.

The ending-balance ratio does the reverse: it counts every dollar still open, however old. Wall Street Prep, an investor-training publisher, describes DSO as the average number of days it takes a company to collect cash from credit purchases ^[2], which is the question the balance answers. Oracle's description of Benchmark 360 answers a narrower one: how long did the invoices that finished take?

The consequence is a built-in survivorship effect. A cohort of invoices that stays unpaid for 90 days contributes nothing to the paid-invoice measure in the month in which it should be most visible. When it is finally paid, it raises a later month's figure. A paid-only measure can therefore look healthy while the ledger deteriorates, a pattern quantified in the worked example later in this report.

Change 2: Weighting (invoice count versus dollars)

Oracle's example takes three invoices closed on May 31 and averages their days open without regard to amount. The ratio form is naturally dollar-weighted, since a $500,000 invoice and a $500 invoice contribute in proportion to their balances. A company that pays a few large invoices late and many small ones promptly shows a short simple average and a long dollar-weighted one.

Practitioner literature stresses a related point: because DSO is an average, early payers can mask late ones ^[16]. This is why collections teams add delinquency measures instead of relying on a single average. The invoice-weighted and amount-weighted variants of "average days to pay" are not defined by a standards body; where a dashboard offers either, the definition has to be read from the report configuration. For example, Oracle documents a Customer Dashboard indicator for average days between invoice date and payment date per customer, which is a customer-level view and not the same population rule as the Benchmark 360 monthly figure.

Change 3: Time basis (closed-in-month versus balance at cut-off)

The Benchmark 360 value is calculated for the current month, and the dashboard value comes from a workbook chosen on the preference page ^[1]. If no workbook, or the wrong one, is selected, the dashboard shows no data ^[17]. Which role can set those preferences should be verified in the account. Whichever workbook is selected therefore defines the number, and the workbook's filters belong in the evidence file.

The ratio form depends on the denominator's period. Public filings show three conventions: quarter-end receivables over that quarter's revenue times the days in the quarter ^[18], average receivables times 91.25 days ^[3], and year-end receivables over annual revenue using 365 days. A company moving from monthly to quarterly or annual denominators changes the reported number without changing collections.

Change 4: Denominator (total sales, credit sales and unbilled amounts)

DSO is normally calculated on credit sales, with cash sales excluded because payment arrives at purchase ^{[10] [19]}. Allianz Trade, a credit insurer, notes that cash sales have an effective DSO of zero ^[20]. In practice many companies do not report credit sales separately, so total net sales are substituted ^[21], and Wall Street Prep notes that credit sales alone are technically more accurate ^[2].

Receivables definitions also differ. One public contractor includes contract assets net of contract liabilities in its DSO ^[22]; another counts billed and unbilled balances. A SaaS business with annual upfront billing and deferred revenue faces a similar choice about whether the numerator is invoiced balances and the denominator billings or recognized revenue. The reconciliation worksheet must state the choice explicitly.

Oracle's standard key performance indicators (KPIs) show how the same label can drift: the standard Receivables KPI is "the total impact on A/R accounts during the selected" periods, which reads as a movement measure rather than a closing balance (verify the definition in the account). A formula assembled from KPI components therefore needs its inputs checked before it is called balance-sheet DSO.

Change 5: Day count and averaging convention

Day counts are 30, 90 or 365 depending on the period ^[10]; the principle is the number of days in the period measured. Some textbooks use a fixed 365 in the numerator of a receivables-turnover formula ^[21]. The CRF's survey formula uses month-end balances and a 30-day month ^[9], and credit-management training material describes a trailing form that measures the average aggregate time that receivables are open ^[23]. A 30-day versus 31-day month changes the result by roughly 3 percent on its own, enough to trigger a variance query.

On averaging, the Chartered Financial Analyst (CFA) Institute's investment-education blog argues that balance sheet figures should be averaged over the period when paired with income statement flows ^[12], while Wall Street Prep notes that ending balances are the more common practice for simplicity. Both are acceptable if documented. A mixed convention, such as an ending balance over a trailing-twelve-month revenue base, is also in use in public reporting ^[24], which is one reason a published DSO cannot simply be compared to an internal one.

Which Question Each Measure Answers

The paid-invoice measure answers: "For invoices that finished this month, how long did the customer take?" The ratio answers: "How many days of sales are sitting in receivables at the cut-off?" Best possible DSO answers: "What would DSO be if every customer paid on time?" Average days delinquent, the gap between the two, answers "how late is the average past-due dollar" ^[23]. Best possible DSO uses only current, non-delinquent receivables ^[14], and one practitioner publication states it as current AR divided by total credit sales, multiplied by the number of days ^[16].

Wikipedia's summary of the metric warns that DSO is often misinterpreted as the average time to collect each payment, when it is better read as days of average sales still outstanding ^[25]. That cautionary reading suits board reporting. For operational collections, the useful question is different: which customers, invoices and buckets drive the gap? NetSuite's own aging report ages items either by due date or by transaction date depending on an accounting preference ^[26], and, as the A/R Aging Detail is understood here (to verify in the account), treats Age as the number of days past due. Those fields describe delinquency, not the elapsed days that Benchmark 360 averages.

Sales volume also moves ratio DSO independently of collections. Practitioners note that a spike in credit sales improves the ratio numerically even if no extra cash has been collected ^[16], and a decline in sales alone raises DSO ^[25]. Seasonality compounds the effect, because with seasonal sales the average receivable figure may be unusually high or low at the measurement date ^[27]. NetSuite's own educational content gives the example of a seller of holiday goods with a high DSO in the fourth quarter ^[10], and recommends comparing like quarters. Others suggest a rolling quarterly calculation to mitigate it ^[27]. Wall Street Prep flags seasonal companies as an exception where the standard approach breaks down.

Forecast dates are a third family. Oracle's Planning and Budgeting module forecasts closing receivables by applying an assumed DSO, which "is based on your past collection history", to credit sales per day. Its Cash Cycle Dashboard shows DSO, days inventory outstanding, days payable outstanding and cash conversion cycle trends for actual and forecast periods, with a selectable version, subsidiary and currency (to verify in the account). That dashboard belongs to the planning product, not core enterprise resource planning (ERP) reporting, and the page read does not define its formulas.

Evidence Extraction Design in NetSuite

The goal is an invoice-level evidence register from which both measures can be rebuilt by anyone. Field names and eligibility rules below are a design to test in the target account, not verified specifications. The Oracle documentation read for this report does not describe how the predefined B360 DSO workbook treats partial payments, credit memos, write-offs or currency, so those must be established by inspecting the selected workbook and dataset directly. Oracle documentation indicates that the SuiteApp installs predefined workbooks and datasets for the KPIs, that custom workbook filters must follow a prescribed order starting with the date column, and that the dashboard KPIs themselves are fixed; confirm each point against the current help pages for the account's release.

Table 2 below lists the evidence fields, where each normally lives in NetSuite, and the question to settle.

Table 2 makes the point that several rows are where reconciliations typically break: the closing event, application mechanics and currency. An invoice may close because a payment, a credit memo, a deposit application or a combination was applied. Applying a credit memo is expected to deduct the credit from the amount due, and auto-apply is expected to direct credit to the oldest outstanding invoices (both to verify in the account). A policy for which of these closures count as "paid" has to be recorded.

Candidate saved-search design

A transaction saved search on invoices can produce the paid-invoice population. A search with Main Line set to true returns one row per transaction. Typical criteria, to be validated in the account, are status equals Paid in Full, close date within the month, and the relevant subsidiary. A numeric formula column subtracting the invoice date from the close date gives days open per invoice, and summary columns give the count, simple average and amount-weighted average.

The balance side should come from the general ledger, not the invoice search. The A/R Aging Detail report shows transactions by the date they are closed rather than applied, and certain journal entries against identical accounts and entities do not appear in it. Items with no customer or project appear on a separate line of the A/R Aging Summary, and the A/R Register combines data from all receivable accounts. Each is a legitimate reconciling object, and none is identical to the invoice-level population.

Controls on changes after the fact

Voiding sets a transaction's total and line items to zero while keeping the record, which Oracle describes as the preferred cancellation route because the audit trail is preserved; deleting a transaction retains no details. Voiding a customer payment is expected to reopen its invoices, and changes to applications in a closed period are expected to require reopening it (both to verify in the account). A paid-invoice population drawn today for a prior month can therefore differ from the one drawn at close. That is the reason to freeze evidence.

Worked Example: One Fictional Ledger, Three Scenarios (Hypothetical Example)

All figures in this section are invented for illustration. They do not describe any real company, and they simplify away partial payments, credits and currency. The month has 30 days. Opening receivables are $1,000,000 in every scenario. Collections equal opening receivables plus sales minus closing receivables, and for simplicity every collected dollar settles a fully paid invoice.

• Scenario A (steady state): sales $900,000; closing receivables $1,000,000; five invoices totaling $900,000 close after 28, 34, 20, 45 and 31 days (amounts $100,000, $300,000, $50,000, $250,000 and $200,000).
• Scenario B (stale balances accumulate): sales $900,000; closing receivables $1,600,000; only three invoices totaling $300,000 close, after 28, 20 and 31 days (amounts $100,000, $50,000 and $150,000). A $700,000 batch of invoices remains unpaid.
• Scenario C (sales surge): sales $1,500,000; closing receivables $1,500,000; five invoices totaling $1,000,000 close after 38, 30, 25, 42 and 35 days (amounts $400,000, $200,000, $150,000, $150,000 and $100,000).

Table 3 below computes each measure from those inputs.

Table 3 shows three results that matter for governance. First, in Scenario B the paid-invoice average is the lowest of the three scenarios (26.3 days) while the ending-balance ratio is the highest (53.3 days). Only the fast invoices closed, so the paid-only measure improved while $700,000 of invoices sat unpaid. Second, in Scenario C the order reverses: the paid-invoice average (34.0 days) exceeds the balance ratios (25.0 to 30.0 days) because sales surged, which enlarges the denominator of the ratio even though customers paid no faster. Third, the amount-weighted paid-invoice figure sits above the simple average in all three scenarios, since the larger invoices in these ledgers were also the slower ones. These are properties of the invented ledgers, not general findings.

The countback figure for Scenario B follows the logic Oracle documents for its JD Edwards product: the loop subtracts sales from the balance, period by period, until the remainder is smaller than a period's sales. Here $1,600,000 minus the current month's $900,000 leaves $700,000, which is 700 / 800 of the prior month, or 26.25 days, added to the 30 days already counted, giving 56.25 days. Countback methods are described as accounting for sales fluctuations ^[14], and JD Edwards lets the user choose among countback, average balance and current balance. Treat the countback result as an illustration of method, because the prior-month sales figure is an assumption.

One caution on working capital. Under the Scenario B assumptions, one day of ending-balance DSO corresponds to roughly $30,000 of receivables ($900,000 of sales over 30 days). A reduction in balance-sheet DSO can imply released working capital if the sales run rate is unchanged, but it is not revenue and not incremental profit. Paid-invoice average days carry no such conversion: they are an elapsed-time average on a closed population, with no receivables balance behind them, so they should never be multiplied by daily sales to estimate cash release.

Implementation Considerations and Process Changes

Reconciliation and sign-off routine

A repeatable close routine ties each component of the reported DSO to a named object. Table 4 below gives the reconciliation bridge.

The sign-off step is the one teams skip. The preparer should record who owns each definition and which use it is approved for (operational collections, working-capital analysis or board reporting), and a reviewer should be able to rerun the register against the snapshot and reproduce the figure.

The SEC's guidance on key performance indicators in management's discussion and analysis offers a useful standard even for private companies. It generally expects a clear definition of a metric and how it is calculated ^[6], says metrics should not deviate materially from those management uses to run the business ^[6], and indicates that when a calculation changes, prior-period metrics should be considered for recasting ^[6]. Applied internally: once the board DSO definition is chosen, changes require a recast of history and a note.

Permissions, roles and report filters

Oracle documentation indicates that Benchmark 360 requires an appropriate role and supports only specified standard centers, and access to preferences depends on role and should be verified in the account. A controller who sees a different number from the administrator may be looking at a different workbook, subsidiary filter or role-limited view. Record the role and filters used for each pull.

Currency, subsidiaries and consolidation

Applying a payment or credit memo to a foreign-currency invoice posts exchange gain or loss when the rate has changed, and revaluation of open balances is a separate month-end process. A base-currency balance-sheet DSO therefore embeds revaluation effects that an elapsed-days measure never sees. Decide whether DSO is reported per subsidiary, consolidated or both, and keep the currency basis constant across periods.

Period comparability checklist

• Same month length: compare 30-day months with 30-day months, or state the convention.
• Same AR basis: ending versus average receivables, stated and unchanged across the comparison.
• Same sales basis: credit sales versus total sales; same treatment of deferred revenue and billings.
• Same scope: identical subsidiaries, A/R accounts and customer exclusions.
• Same cut-off: snapshot date and time, with period lock status.
• Seasonality control: compare to the same quarter a year earlier, as NetSuite's own educational content advises.

Data Analysis and Evidence

Published benchmarks cannot settle a company's target because they use different formulas, samples and years. The sources below are the originators whose methods are visible.

Credit Research Foundation (CRF). The CRF's National Summary of Domestic Trade Receivables for the second quarter of 2026 reports independent medians ^[29]: DSO of 37.50 days, best possible DSO of 34.00 days and average days delinquent of 3.70 days. The CRF computes survey DSO from the last three month-end balances times 30 over quarterly credit sales. A company using ending receivables over monthly total sales and a 30-day month is on a different footing, so the comparison is indicative only. The page does not state a participant count.

NYU Stern (Aswath Damodaran). The January 2026 working-capital dataset gives accounts receivable as a percent of sales by sector, not as days ^[8]. It reports, for example, 16.84 percent for application software across 309 firms and 12.83 percent for distributors across 62 firms. Multiplying a percentage of annual sales by 365 gives a rough day-equivalent (approximately 61 days and 47 days respectively; this report's arithmetic, not the dataset's). Inclusion of financial firms changes the total-market figure from 12.39 percent to 48.40 percent, which shows how sample definition alone can swing a "benchmark."

Hackett Group. In its 2025 survey of the top 1,000 US publicly traded nonfinancial companies, the Hackett Group identified an 18-day DSO gap between top and median performers ^[30]. Its 2026 North American release reports DSO worsening for a third consecutive year, in 40 of 50 industries ^[31]. A trade-press summary of the 2025 survey reported DSO falling 0.7 days to 41.9 days ^[32]. These statements are not obviously comparable: samples (US versus North America) and statistics differ, and the releases do not publish the formula used. The discrepancy is left unresolved here.

Atradius. The 2024 US Payment Practices Barometer, an earlier edition than the October 2026 date of this report (later editions may report different figures), found that half of B2B invoices were overdue (Source: atradius.us) and that overdue invoices were collected on average 20 days past due (Source: atradius.us). Most US respondents set payment terms of about 30 days from invoicing (Source: atradius.us). The sample was 240 interviews (Source: atradius.us), concentrated in chemicals, electronics and steel manufacturing, so it is a payment-terms context for manufacturers rather than a software or distribution benchmark.

Academic data. Klapper, Laeven and Rajan analyzed 29,019 supplier contracts in a National Bureau of Economic Research working paper, finding average net days of 59.2 and a median of 60 ^[33]. They note that contract-level data avoids misreporting concerns attached to survey-based trade credit data ^[33], so the answer to "how long do customers take" depends on whether terms, contracts, invoices or balances are measured. Allianz Trade attributes regional differences to cultural payment behaviors ^[20].

Oracle's Benchmark 360. NetSuite takes the industry benchmark values for its dashboard KPIs from the American Productivity and Quality Center (APQC) ^[34], for five preselected KPIs. APQC's own DSO definition could not be retrieved for this research (the page returned an access error), so whether it matches Oracle's paid-invoice formula is unverified; confirm before treating the quartile comparison as like-for-like.

Public-company definitions. SEC filings show distinct conventions in plain text: quarter-end receivables over quarterly revenue; a 365-divided-by-4 quarter convention ^[35]; average receivables with 91.25 days; average receivables over average daily net sales ^[36]; period-end receivables including unbilled, less deferred revenue; and a trailing-twelve-month revenue base. Regulation S-X requires the allowance for doubtful accounts to be set out separately ^[37], a reminder to state whether receivables are gross or net.

Selecting a Governed Target

A governed target starts from the decision it supports. For operational collections, set targets on best possible DSO and average days delinquent, because they separate terms from lateness (the NACM session material defines best possible DSO as the DSO if all customers paid on time ^[23]). For working-capital analysis and board reporting, use a balance-based DSO with a stated averaging rule and day count. For a lagging view of payment behavior, retain the paid-invoice average from Benchmark 360 or an equivalent saved search, labeled as such.

Compare the company with its own history first, using matched seasonality, and then with like-for-like businesses whose definitions are known. Avoid universal DSO benchmarks. The CRF describes DSO as one of the most often misused measures ^[7], and a session hosted by NACM, the credit-management association, quotes the CRF president calling it the most misunderstood and misused metric in the field ^[23]. Where industry medians are used, record the originator, year, formula and sample alongside the target.

Implications and Future Directions

For finance leaders, the implication is a governance step more than a technical one. Each DSO in a board pack, covenant calculation, bonus plan or collections scorecard should carry a one-line definition from a versioned register: calculation basis, reporting population, weighting, cut-off, exclusions, data source, owner and approved use.

Several uncertainties remain. Oracle may revise Benchmark 360 documentation or the predefined workbook; the pages cited here were read on October 27, 2026, and the Oracle DSO page did not disclose partial-payment, credit-memo, currency or write-off handling. Updates to benchmark data, if any, can change peer quartiles without any change in a company's own performance. And no public benchmark identified in this research publishes a field-level reconciliation of paid-invoice days to balance-sheet DSO, which is why internal documentation matters.

Options for executing the reconciliation

Companies can approach the work in different ways; Table 5 below compares neutral options. It describes service models only and does not rank them.

Table 5 is intended to help scoping, not to recommend a provider. The trigger for outside help is usually specific: two reports that cannot be tied after the Table 4 bridge has been attempted, or a definition change that must be applied consistently across subsidiaries, integrations and history.

Frequently Asked Questions (FAQs)

How is DSO calculated in NetSuite?

There is more than one calculation. Benchmark 360 averages days open of fully paid invoices for the month. A balance-based DSO is built from receivables and sales using a stated day count and averaging rule ^[7].

Is there a NetSuite days sales outstanding report?

The Benchmark 360 dashboard shows DSO from a workbook selected in preferences; the Planning and Budgeting Cash Cycle Dashboard shows DSO trends by subsidiary and currency. Standard aging reports show delinquency rather than DSO.

Can DSO be built as a saved search?

A paid-invoice average can be built from an invoice search with status, close date and a day-difference formula, as described above. The ratio form should be tied to the ledger, not to an invoice search. Field names must be validated in the account.

What is the difference between average days to pay and DSO?

Average days to pay measures elapsed days on invoices that were paid, for example a customer-level indicator of days between invoice and payment dates. DSO as a ratio measures days of sales sitting in receivables ^[25].

Why doesn't DSO match average days to pay?

Different populations (closed invoices versus open balances), weighting, cut-off and denominators. The Table 3 scenarios show the paid-invoice average falling from 31.6 to 26.3 days while ending-balance DSO rose from 33.3 to 53.3 days.

What is best possible DSO, and how does it differ from average days delinquent?

Best possible DSO uses only current receivables; average days delinquent is DSO minus best possible DSO and reflects how far past due invoices are on average.

What is the count-back method, and how does it differ from simple DSO?

Count-back subtracts earlier periods' sales from the balance until it is used up ^[38], so a sales surge in the latest month does not flatter the figure. Simple DSO divides the balance by the current period's sales only.

How should an AR aging report be reconciled?

Tie the aging total to the general ledger receivable account(s), and explain unapplied payments, journals and items without a customer. Record the aging basis, due date or transaction date.

Conclusion

Two valid NetSuite DSO reports disagree because they answer different questions on different populations. Benchmark 360 averages days open of fully paid invoices for the month; the finance ratio converts a receivables balance into days of sales ^[7]. A paid-only average can fall while unpaid balances grow, and a sales surge can lower the ratio without faster payment, as the illustrative scenarios show.

The decision framework is short. Use a balance-based DSO with a stated averaging rule for board and working-capital reporting; best possible DSO and average days delinquent for collections; and paid-invoice days as a lagging payment-behavior indicator. Document the population, weighting, cut-off, exclusions, data source, owner and approved use for each, tie the pieces to the ledger monthly, and freeze the evidence. Targets should come from the company's own history and like-for-like peers.

Where invoice populations or definitions cannot be reconciled, a scoped reporting and configuration engagement with a NetSuite specialist is a reasonable next step (disclosure: this report is published by Houseblend, a Montréal NetSuite consultancy, so the link to Houseblend is a self-reference, not independent evidence), beginning with account-level testing of the field names, eligibility rules and workbook filters described in this report.