CPG 13-Week Cash Flow Forecast: A Purchase Order Is Not Cash

In the last week of December, we had temporary workers ready, dried mango in the warehouse and approximately $135,000 of customer purchase orders scheduled to ship.

We were missing the pouches.

The product could not be packed, the orders could not be picked up and the cash clock never started. A sales forecast would still show the demand. The P&L might eventually show the revenue. Neither one could pay payroll that week.

A growing CPG company needs a 13-week cash flow forecast because sales demand and usable cash can diverge that quickly.

A 13-week forecast is a weekly operating document connecting bank cash, available credit, customer collections, supplier payments, inventory commitments, production and the decisions management still has time to change. Shrinking the annual budget into weekly columns will not produce one.

If it is maintained only by accounting, it will be wrong. The information that determines next Friday’s cash is usually sitting with sales, operations, sourcing and the people speaking to customers and suppliers.

Why 13 weeks works

Thirteen weeks is long enough to expose a liquidity problem before it becomes an emergency and short enough to force operational detail. J.P. Morgan describes the same horizon as a practical balance between detail and planning horizon.

In practice, the confidence level changes across the forecast:

• Weeks 1–2: invoice-level collections, scheduled payroll, known supplier payments and committed production.
• Weeks 3–6: informed estimates tied to the operating plan.
• Weeks 7–13: directional visibility, with more assumptions and room to act.

Pretending week 12 has the same precision as next Friday creates false confidence. Weeks 7 through 13 expose the direction of travel; the next two weeks drive payment decisions.

Start with usable liquidity, not the P&L

The opening position should include:

• Cash available across operating accounts
• Confirmed incoming transfers
• Available revolver capacity
• Any restrictions or minimum liquidity requirements

Do not count the face amount of a credit line if it is fully drawn, frozen, limited by a borrowing base or otherwise unavailable. A facility is liquidity only when the company can actually use it.

I kept two weeks of payroll plus one month of rent as our minimum operating cushion. It was our internal danger line, not a universal formula. Once the projected balance fell below it, management had to act. Waiting for the bank balance to approach zero would have left no time to collect receivables, negotiate with suppliers or change production.

The weekly model should be structured so management can see the cash bridge without opening supporting schedules:

Ending available liquidity equals ending bank cash plus revolver capacity the company can still draw. Keeping those lines separate prevents management from mistaking borrowed availability for cash already in the bank.

The rows beneath those totals hold the real work: customer-level collection assumptions, invoice-level AP, production commitments and named owners. The summary tells management where to look.

The arithmetic is simple. Deciding which numbers deserve to be in the forecast requires operating judgment.

Forecast customer behavior, not contractual terms

An invoice due date is a starting point. Historical behavior is a better cash assumption.

UNFI was supposed to pay us 10 days after picking up product from our Los Angeles warehouse under our FOB terms. In practice, it often appeared to start the clock when the product reached its warehouse, four to seven days after pickup. That difference mattered when payroll or a raw-material payment fell in between.

The forecast used the date UNFI was likely to pay, not the date the contract said it should pay.

Distributor receipts also needed a deduction haircut. We could estimate the overall percentage from experience, but not which promotions, shortages, returns or old claims would hit a particular remittance. Forecasting the full invoice value would have been fiction.

Every customer needed its own logic:

• Costco generally paid on time and could be forecast with reasonable confidence.
• CVS generated substantial deductions, so invoice value was not a dependable cash estimate.
• Some smaller customers prepaid. I was willing to offer a meaningful discount when earlier cash was more valuable than nominal margin.
• Amazon and DTC paid faster than grocery distribution, but fees, fulfillment, advertising, reserves and logistics still had to be understood. Once the engine worked, those channels could improve the cash cycle materially.

Disputed deduction recoveries stayed out of the base case until confirmed. The claim might be valid and the company might eventually win, but an unresolved dispute cannot fund this week’s payroll.

This is where the gross-to-net process feeds directly into cash forecasting. A company that cannot explain its distributor remittances cannot forecast them reliably.

Use scenarios for uncertainty, not wishful averages

We used one working forecast built from the best available information and refreshed it every week. For annual planning, we also tested different scenarios and assigned probabilities. Inside the 13-week view, management still needs to know what changes if the uncertain receipts and shipments do not arrive.

Do not average the three into a number that describes none of them. Keep the operating forecast honest, then show management the downside and the response.

This is also how hiring and discretionary spending should be approved. A new hire may fit the annual P&L and still push the 13-week low point below the company’s operating cushion. The start date, contractor engagement or marketing program should move when the cash says it has to move.

A purchase order does not belong in cash receipts

Sales teams want the forecast to reflect every promising meeting and expected order. Finance has to be more disciplined.

A retailer’s verbal interest is not cash. An expected purchase order is not cash. Even a purchase order is not a receivable until the product ships or is picked up under the applicable terms.

Before the shipment, too much can still change: timing, quantity, inventory availability, packaging, retailer approval or the buyer’s own plan.

The $135,000 December order is a good example. We had demand, labor and the primary ingredient. Missing one packaging component stopped the entire conversion from purchase order to shipment to receivable to cash.

I would not authorize a meaningful inventory or production commitment for a new account without a written customer commitment. Sales enthusiasm is not enough to spend cash against.

The AP report and the cash forecast must tell the same story

Every Wednesday, once the distributor remittance notices arrived, we had reasonable visibility into the cash expected roughly two days later. That was when the week’s payment decisions were made.

The process combined:

• Current bank cash and available borrowing
• AR by customer, expected payment date and collection probability
• AP invoices and due dates
• Purchase orders requiring payment before release
• The production schedule
• Incoming ingredients and packaging
• Suppliers that might place the company on hold
• Payroll, rent, insurance, debt payments and other fixed obligations

Finance consolidated the information, but operations and sourcing made the forecast usable. They knew which supplier consistently delivered late, which representative would escalate a missed payment, which vendor could tolerate another week and which small unpaid invoice might stop a much larger shipment.

That knowledge rarely lives in the ERP.

In healthy periods, AP could be paid according to terms. During tighter years, I manually selected invoices and sometimes paid only part of an invoice. It was unpleasant, but it was the actual job. A cash forecast that assumes every payable will be paid on its due date may look clean while being completely disconnected from how the company is operating.

Communication with suppliers mattered enormously. We answered calls, explained the situation and gave dates when we could. Stretching AP without communication destroys trust faster than paying late with a credible plan.

One delayed broker payment proved how interconnected those relationships were. The broker spoke with another supplier, and that supplier placed us on hold until the balance was resolved. The amount looked small in the AP aging. The operational consequence was not.

Production planning belongs inside cash planning

Operations built the production and replenishment schedule from purchase orders, minimum production quantities and the lead times for ingredients and packaging. Finance converted that operating plan into payment timing. If those two schedules do not agree, the cash forecast is already wrong.

For most purchases, I used the total delivered cost to our warehouse. Splitting freight, duties and ingredient cost into more detail was useful only when the charges were large or payment terms differed enough to change the cash decision.

Inventory commitments were separated by certainty:

1. Purchase orders already issued
2. Expected purchases required by the production plan
3. Desired purchases that would proceed only if cash allowed

Long-term ingredient contracts added obligations beyond open POs. When we saw that annual quantities would not be pulled in time, we negotiated early to roll the balance into the next contract when the supplier relationship allowed it.

What gets paid when cash is tight

There was no elegant universal ranking. Payroll, rent and insurance came first. After that, the answer changed every week.

A supplier could move up the list because a shipment was waiting for release, production needed the material, terms were about to tighten or the relationship was close to breaking. Another vendor might accept a partial payment. Lawyers, brokers and expenses not directly tied to keeping the company operating could often wait.

Operations did not always see the urgency of bank interest and principal because those payments did not release inventory or keep a production line moving. Finance had to protect those obligations while understanding the operational consequences of every dollar withheld elsewhere.

AP prioritization belongs to management. Accounting can show what is due; it cannot decide which payment protects the business this week.

The weekly discipline

The forecast worked because it had a recurring rhythm:

1. Update receipts using customer-specific behavior. Start with open AR, expected remittances and realistic deduction assumptions.
2. Connect spending to operations. Include payroll, debt, inventory commitments, release payments and the production schedule.
3. Compare forecast with actuals. The miss matters. Was cash late, was the deduction larger, did a shipment move or was an expense omitted?
4. Make the decisions immediately. Chase AR, delay or split AP, reduce a production run, postpone discretionary spending or escalate the funding need.

I maintained the forecast weekly and communicated the broader outlook monthly. The CFO, COO and CEO saw the full picture. Other leaders received the information and decisions relevant to them.

The CEO does not need to study every cell when liquidity is healthy. The CEO needs to know when the forecast crosses the danger line, when a launch changes the low point or when management action is required.

The most common forecast failures

I do not trust a cash forecast that is simpler than the business it claims to represent.

The usual failures are predictable:

• Starting with the P&L and contractual due dates instead of actual cash behavior
• Treating purchase orders, disputed deductions or fundraising conversations as cash
• Ignoring debt principal, card settlements and the full cost of inventory and production
• Showing credit capacity the company cannot actually draw
• Using one collection assumption across customers with very different behavior
• Building the model inside finance without operational input or forecast-to-actual review

Founders also wait too long to chase receivables because they fear damaging the customer relationship. Professional collection is part of the relationship, and overdue AR is often the fastest financing available to the company.

When the 13-week forecast showed a problem, collecting AR was always the first action. Then came supplier conversations, smaller production runs, delayed discretionary expenses and, if necessary, debt or equity discussions.

We could usually see a serious liquidity issue about two weeks ahead. That was not much time, but it was enough to make decisions. Without the forecast, the same problem would have appeared when the bank balance was already gone.

Every CPG company should have one

A 13-week cash forecast is useful before a company reaches crisis. It can begin as a simple weekly schedule and grow with the business.

For a healthy brand, it protects inventory planning, retailer launches and hiring decisions. For a constrained brand, it becomes the operating map for the next payment cycle. For an investor or lender, it shows whether management understands how the business actually converts orders into cash.

At one point, our credit capacity was fully utilized. The forecast showed that we could not fund the slotting, inventory build and collection delay behind another major retailer or product line. Saying no was frustrating. A purchase order would not have financed the launch.

The forecast forces that decision while options still exist.

Netter Group helps growing CPG companies connect cash forecasting with receivables, inventory, production, retailer launches and supplier commitments. If your bank balance keeps surprising you even when sales are growing, contact julien@nettergroup.com.