Working capital is not a plug. It is not a percentage of revenue. It is a set of operational drivers — Days Sales Outstanding, Days Inventory Outstanding, Days Payable Outstanding — that translate a company’s commercial reality into cash timing. Get the drivers wrong and your cash flow statement is fiction, regardless of how clean the income statement looks.
There is a question that comes up in almost every financial modeling interview in New York: “Walk me through how working capital flows into the cash flow statement.” Most analysts can define NWC as current assets minus current liabilities. Far fewer can explain why an increase in accounts receivable is a use of cash, or how to derive a DSO assumption from historical financials, or what happens to the NWC schedule when revenue assumptions shift by 20%.
This guide covers the mechanics of working capital modeling the way a practitioner builds it — with the driver logic, the sign conventions that trip people up, the schedule structure that survives a model audit, and the decisions that depend on what kind of company you are modeling.
What Working Capital Actually Measures — and Why the Simple Definition Fails
Net Working Capital is current assets minus current liabilities, excluding cash and short-term debt. That is the textbook definition. It is accurate and nearly useless for building a financial model.
What NWC actually measures is the cash timing gap between when a business incurs costs and when it collects revenue. A company that collects cash from customers in 90 days but pays suppliers in 30 days has a large cash timing gap — it needs to fund 60 days of operations from its own balance sheet or a revolving credit facility. A company that collects in 30 days and pays in 60 days generates cash from its working capital cycle, which is why some retailers effectively have negative NWC.
The reason this matters for modeling is that NWC drives cash generation independent of profitability. A company can be highly profitable and still consume cash if it is growing fast and collecting slowly. A company can be barely profitable and generate strong cash flow if it manages its working capital cycle aggressively. These dynamics only appear in the model if the working capital schedule is built from operational drivers — not if it is hardcoded or set as a fixed percentage of revenue.
The Three Drivers: DSO, DIO, DPO
Every working capital schedule is built from three ratios applied to the income statement. These ratios convert operational behavior — how quickly customers pay, how long inventory sits, how long the company takes to pay suppliers — into balance sheet positions.
Days Sales Outstanding (DSO) measures how many days of revenue are sitting in accounts receivable at any point in time. The formula is accounts receivable divided by revenue, multiplied by 365. A DSO of 45 means the company takes an average of 45 days to collect cash after recognizing revenue. In the model, accounts receivable in any period equals revenue divided by 365, multiplied by DSO.
Days Inventory Outstanding (DIO) measures how many days of cost of goods sold are held in inventory. The formula is inventory divided by COGS, multiplied by 365. A DIO of 60 means 60 days of production costs are tied up in unsold goods at any given time. In the model, inventory equals COGS divided by 365, multiplied by DIO.
Days Payable Outstanding (DPO) measures how many days the company takes to pay its suppliers. The formula is accounts payable divided by COGS, multiplied by 365. A DPO of 30 means the company pays suppliers within 30 days of receiving goods. In the model, accounts payable equals COGS divided by 365, multiplied by DPO.
The Cash Conversion Cycle ties these together: CCC equals DSO plus DIO minus DPO. A lower CCC means the company converts its investments in working capital to cash more quickly. A negative CCC — common in businesses like large grocery retailers or subscription software — means the company is effectively financed by its suppliers and customers.
Deriving Driver Assumptions from Historical Financials
Before you set a DSO assumption in the model, you need to know what the company’s historical DSO actually is — and whether it has been stable, improving, or deteriorating.
The derivation is mechanical. Pull three to five years of accounts receivable and revenue from the financial statements. Calculate DSO for each period: AR divided by revenue, multiplied by 365. If the number is stable — say, 42 to 45 days across all periods — your forecast assumption is straightforward. If it is trending in one direction, you need to understand why. Has the company extended payment terms to win customers? Is collections discipline deteriorating? Is the business mix shifting toward slower-paying segments?
The same process applies to DIO and DPO. Build a historical driver table in the assumptions tab before you set any forecast assumptions. This is the difference between a model that can be defended in a data room and one that cannot. When a counterparty asks “why is your DSO 45 days in year three,” the answer should not be “that seemed reasonable.” It should be “the historical average over the last four years is 43 days, with a slight upward trend — we assumed 45 days as a conservative base case.”
Building the NWC Schedule in Excel
The schedule has a simple structure. Each working capital item gets its own row. For each item, you calculate the balance using the driver formula, then calculate the period-over-period change. The changes feed the cash flow statement.
A clean NWC schedule looks like this:
| Line Item | Formula | Feeds |
|---|---|---|
| Accounts Receivable | Revenue ÷ 365 × DSO | Δ AR → Cash Flow (Operating) |
| Inventory | COGS ÷ 365 × DIO | Δ Inventory → Cash Flow (Operating) |
| Other Current Assets | % of Revenue × Revenue | Δ OCA → Cash Flow (Operating) |
| Accounts Payable | COGS ÷ 365 × DPO | Δ AP → Cash Flow (Operating) |
| Accrued Expenses | % of OpEx × OpEx | Δ Accruals → Cash Flow (Operating) |
| Net Working Capital | AR + Inventory + OCA − AP − Accrued | — |
| Change in NWC | NWC(t) − NWC(t−1) | Direct input to Cash Flow Statement |
Every driver — DSO, DIO, DPO, the percentages for other items — lives in the assumptions tab, not in this schedule. The schedule contains only formulas that reference the assumptions tab and the income statement. No hardcoded numbers in forecast periods.
The Sign Convention That Trips Up Analysts
This is where models most commonly break — and where interviewers most commonly test.
An increase in accounts receivable is a use of cash. The company has recognized revenue but not yet collected it. Cash has not arrived. In the cash flow statement, this appears as a negative number: subtract the increase in AR from operating cash flow.
A decrease in accounts receivable is a source of cash. Collections exceeded new revenue recognition. Cash came in. This appears as a positive number in the cash flow statement.
An increase in accounts payable is a source of cash. The company owes more to suppliers but has not yet paid. It is effectively borrowing from suppliers. Add the increase in AP to operating cash flow.
A decrease in accounts payable is a use of cash. The company paid down its supplier obligations. Cash went out. Subtract the decrease in AP.
The clean way to handle this in the model: calculate the change in NWC as current period NWC minus prior period NWC. Then subtract this number from operating cash flow in the cash flow statement. If NWC increased, the change is positive, and subtracting it reduces operating cash flow — which is correct. If NWC decreased, the change is negative, and subtracting a negative number adds to operating cash flow — also correct.
What practitioners do NOT do: separately sign-adjust each line item in the cash flow statement. That introduces formula complexity and audit risk. One row in the NWC schedule, one row in the cash flow statement, and the sign convention takes care of itself.
Days-Based vs. Percentage-of-Revenue: When to Use Which
The days-based approach — DSO, DIO, DPO — is the right default for any business where the working capital cycle is operationally meaningful: manufacturing, retail, distribution, healthcare, industrials. These businesses have real inventory turns, real collections cycles, and real supplier payment terms that can be benchmarked against industry peers.
The percentage-of-revenue approach is appropriate in two situations. First, for line items where a days-based driver does not make operational sense — prepaid expenses, accrued liabilities, deferred revenue. These are better modeled as a percentage of the relevant income statement line. Second, for simplified models under time pressure — a back-of-envelope LBO model where working capital is not the critical variable, and a rough percentage-of-revenue assumption is sufficient for the purpose.
The error we see most often is the reverse: analysts use percentage-of-revenue for everything, including accounts receivable and inventory, because it is faster to set up. The problem is that a percentage-of-revenue model does not reflect what actually happens when the business changes — when payment terms shift, when inventory management improves, when the revenue mix moves toward faster or slower-paying customers. In those situations, the days-based model captures the dynamics correctly and the percentage-of-revenue model does not.
What Interviewers in New York IB Roles Actually Ask
In investment banking interviews in New York, working capital questions come in two forms. The first is conceptual: “If accounts receivable increases by $10 million, what happens to cash flow?” The answer — operating cash flow decreases by $10 million, all else equal — needs to be immediate and mechanically explained, not just stated.
The second form is model-based. You are given a set of financials and asked to derive the NWC schedule. Interviewers look for whether you know to use revenue as the denominator for AR and COGS as the denominator for inventory and payables. Analysts who use revenue for all three line items — a common error — will produce an AP balance that is systematically too large or too small relative to the company’s actual payment terms.
The test is not whether you can define DSO. It is whether you can build the schedule correctly from scratch, explain each driver assumption, and walk through how a change in DPO affects operating cash flow, net debt, and ultimately the valuation. That chain of logic — from operational driver to cash impact to enterprise value — is what a trained financial modeler can articulate and a self-taught analyst often cannot.
If you cannot yet walk through that chain confidently, that is exactly what a structured session at Financial Modeling New York is designed to fix — using your actual model or a live case study, not a pre-built template you watch someone else work through.
How Working Capital Connects to the Rest of the Model
The NWC schedule does not sit in isolation. It has three downstream connections that must work correctly for the integrated model to hold together.
The cash flow statement receives the change in NWC as a single line item in operating activities. If the NWC schedule is days-based and correctly structured, this line updates automatically when revenue or cost assumptions change.
The balance sheet receives the ending balance for each NWC line item. Accounts receivable, inventory, other current assets, accounts payable, and accrued liabilities all pull directly from the NWC schedule. These cells should never be hardcoded on the balance sheet.
The credit metrics and leverage analysis depend on the accuracy of the NWC schedule. Free cash flow — the basis for DCF valuation, LBO return analysis, and debt capacity analysis — is directly affected by the change in working capital. A model where NWC is a rough estimate produces a free cash flow figure that will not withstand scrutiny in a data room or a lender’s credit analysis.
Model Checks for the NWC Schedule
The NWC schedule needs its own set of sanity checks, separate from the balance sheet balance check.
First: implied DSO, DIO, and DPO checks. Back-calculate the implied drivers from the model’s output to confirm they match the assumption inputs. If DSO is set to 45 in the assumptions tab but the back-calculated DSO from the model’s AR balance is 47, there is a formula error somewhere.
Second: NWC as a percentage of revenue. Calculate this for both historical periods and forecast periods. If it jumps materially in a forecast year without a corresponding change in the driver assumptions, a formula is broken.
Third: the sign check on the cash flow statement. In a year of strong revenue growth, accounts receivable should increase and operating cash flow should be lower than EBITDA to a degree that reflects the AR build. If operating cash flow is higher than EBITDA in a high-growth year without an explanation in the payables or accruals, check the NWC sign convention.
Frequently Asked Questions
How do you calculate working capital in a financial model? Build a schedule with three days-based drivers: DSO for accounts receivable (AR ÷ Revenue × 365), DIO for inventory (Inventory ÷ COGS × 365), and DPO for accounts payable (AP ÷ COGS × 365). Derive each driver from historical financials, set forecast assumptions in a central assumptions tab, and let the balance sheet pull from the schedule.
Why does an increase in accounts receivable reduce cash flow? Because the company recognized revenue but has not yet collected the cash. Cash flow measures actual cash movement — when AR rises, revenue was booked but money did not arrive. In the indirect cash flow statement, subtract the increase in AR from operating cash flow. The sign reverses when AR decreases: collections exceeded new revenue recognition, adding cash.
When should you use percentage-of-revenue instead of days-based drivers? Use days-based drivers for accounts receivable, inventory, and accounts payable — any item where the collection or payment cycle has operational meaning. Use percentage-of-revenue for items like prepaid expenses, deferred revenue, and accrued liabilities, where a specific payment cycle does not apply. Using percentage-of-revenue for AR and inventory produces a model that cannot reflect changes in payment terms or inventory management.
How is working capital modeling tested in New York investment banking interviews? Interviewers test two things: the mechanical sign convention (does an AR increase reduce or increase cash flow, and why) and the driver derivation (can you calculate DSO from a balance sheet and income statement). The more advanced test is asking you to walk through how a deterioration in DSO from 40 to 55 days affects operating cash flow, free cash flow, net debt, and enterprise value. That full chain is what separates a prepared candidate from one who has only read about it.
For analysts preparing for New York IB interviews or finance teams rebuilding their working capital models from scratch, a structured session covers this build live in Excel with a real case. Reach out at financial-modeling.com to discuss what that looks like for your situation.