About sharynneely522

How to build an FP&A function: start with a 13-week cash view

The common mistake when deciding how to build an FP&A function is starting with software and a twelve-month budget before anyone can state the cash you will hold in ninety days. The consequence is a planning process that makes slides while a payment bounces. Start smaller: a 13-week cash view, updated weekly, that feeds decisions you actually make.

What a 13-week cash view gives you that a budget does not

A 13-week cash view states the bank balance you can expect at the end of each week for the next quarter, in the order money actually moves. A budget, by contrast, records intentions over a year, spread across accrual accounting. Both are useful, and they answer different questions.

The budget can look healthy while week seven dips below zero. Revenue enters the books when it is earned, not when it is deposited, so a profitable quarter can still contain a week where payroll lands before a large customer pays. The cash view exposes that week while you still have six weeks to do something about it.

Thirteen weeks is one quarter, and that length is deliberate. It is long enough to act on what you see: collecting an invoice, moving a purchase, arranging short-term credit. It is also short enough that the inputs are knowable, because payroll dates, rent, and loan payments are already scheduled and the receivables already exist.

The mechanics fit on one page. Start with the opening balance, add expected receipts by week, subtract payments by week, and the result is a closing balance for each of the thirteen rows. Track the actual movement of money, which planners call the direct method, rather than converting profit into cash through a chain of adjustments.

One habit keeps the view honest: treat it as a working picture, not a promise. When a customer changes a payment date, change the line. When hiring slips, move the expense. The forecast earns trust by being corrected in public, on schedule, every week.

This view is not a full profit forecast, and it does not replace your accounting system. It is the first module of the function because every later piece, from scenario planning to investor reporting, depends on knowing where cash stands now and where it heads next.

Build the first version from what you already know

Everything the first version needs already sits in systems you run today. Bank balances come from the bank, scheduled payments from the accounting software, expected receipts from the receivables aging report, payroll dates from the payroll calendar. The work is assembly, not creation.

Start with outflows, because they are the most certain. List payroll by date, rent, loan payments, insurance, and the recurring software subscriptions that quietly add up. Then add variable spending from recent history, such as inventory purchases and contractor invoices. Certainty first, estimates second.

Receipts need the most honesty. Use the date a customer is likely to pay, based on how that customer has actually paid, not the due date printed on the invoice. If a large account historically runs three weeks late, put the money in the third week of lateness. Optimistic dates make the whole view useless.

The arithmetic stays simple: closing balance equals opening balance plus receipts minus payments, rolled forward week by week. Excel or Google Sheets handles this comfortably. Keep categories coarse at first, maybe ten lines of payments, and refine only when a category starts hiding decisions from you.

Before you trust the first version, run this check:

• every number traces to a source you can open;
• receipts are dated by expected deposit, not invoice date;
• the lowest week of the quarter is visible at a glance;
• one named person is responsible for the weekly update.

Expect the first pass to take a day, Cfo Services not a month. Depth comes from the weekly updates, when reality tells you which lines were wrong and by how much.

Clean the inputs before you trust the output

A forecast is only as good as the records underneath it, and the first weeks usually surface the same three problems. Invoices sit unrecorded, so the receivables list understates what customers owe. Bills are entered late, so next week’s payments look lighter than they are. Bank feeds lag, so the opening balance is a guess rather than a fact.

Fix the receivables aging first, because receipts drive most of the view. Match every open invoice to a real expectation: the customer’s payment history, a promise in writing, or a dispute still open. Anything you cannot match to an expectation gets its own line, dated conservatively, until someone resolves it.

Cut-off matters more than precision. A number that is a week old and labeled as such beats a fresh-looking number that quietly misses two days of activity. Set a rule for when the books close for the week, and update the view from that snapshot, so every figure in the file has a single point of origin.

While cleaning, watch for signals about the business itself:

• customers whose payments stretch a little further each month, which belongs in the receipt dates;
• supplier invoices that arrive after their due date, which hides future payments;
• recurring subscriptions nobody remembers approving, which are the easiest savings you will find.

Simple separation helps when more than one person touches the numbers. The person who enters bills should not be the only person who reconciles the bank account, even in a small team, because the forecast inherits whatever slips through. Where the team is one person, a monthly review by an outside bookkeeper or accountant serves the same purpose.

None of this cleanup is glamorous, and that is the point. Tracing every input to a source is the first internal control the business gets from the exercise, and it costs nothing but attention.

The questions that set the size of the function you need

How much FP&A you need depends on how often finance information changes a decision, and on what events are coming, not on revenue alone. A business with one bank account, rent, and payroll needs a lighter function than its revenue suggests. A business buying inventory on thin margins needs more.

Ask four questions. First, how often does cash position change what you do, weekly or only at quarter end? Second, is a financing round, a loan application, a sale process, or an acquisition planned within the next year? Third, how many entities, accounts, and currencies does the business hold? Fourth, who updates the numbers today, and how much time do they have?

The answers sort businesses into rough groups. Simple structures with predictable flows run well on one spreadsheet and a weekly ritual. Businesses facing an event, such as raising debt or preparing materials for a sale, need output that survives outside scrutiny from lenders, investors, or buyers.

Cost belongs in the decision too. Every option, from a software subscription to added help, is bought with money or with hours you currently spend elsewhere. Match the spend to the decisions the output will inform, and re-check the match when circumstances change.

A simple test keeps this honest: name three decisions of the next quarter that the output will feed, such as a hiring date, a purchase timing, or a credit line size. If the list will not fill, the function is decoration for now, and decoration can wait.

When the in-house spreadsheet is the right answer

The spreadsheet fits when flows are simple, the structure is one or two bank accounts, and someone inside the business can protect a fixed weekly slot for the update. Under those conditions it is the correct tool, with no apology needed.

Its strengths are real. You see every formula, so nothing hides in a black box. Changes take minutes. The cost rounds to zero, and anyone can learn the file in an afternoon, which matters when the usual owner takes vacation.

The limits show up as the business grows. Multiple people editing copies creates version problems. Manual re-keying from the accounting system invites errors. Outside parties, such as lenders reviewing a facility, eventually ask for formats and history that a loose file struggles to produce.

You have outgrown the spreadsheet when you maintain several versions and argue about which is current, when the weekly re-keying eats hours, or when the file cannot answer a scenario without being rebuilt. Those are process problems first, and buying software solves only the ones caused by repetition.

Document the file as if someone else will run it, because someday someone else will. A single page listing the source of each line, the day the update happens, and the two or three formulas that do the work turns a personal artifact into a business process. Vacation, illness, and turnover stop being risks to the forecast at that point.

Keep the file honest with three habits: one master file, one named owner, and a short log of assumption changes next to the numbers. A spreadsheet updated every week beats an expensive platform updated when someone remembers.

When a planning tool, part-time help, or a full-time hire fits better

Three upgrades exist once the manual version works: planning software, a fractional or interim finance lead, and a full-time hire. Each answers a different bottleneck, and choosing among them comes down to criteria.

Software fits repetition and structure: several entities or accounts, monthly or faster reforecasts, many people needing identical numbers, and errors caused by re-keying. Buy it after the manual process runs, because the tool automates steps you must first understand. A system fed with guessed inputs produces confident nonsense on schedule.

A fractional or interim finance lead fits two situations. One is a defined event on the horizon, such as preparing financing materials, supporting due diligence in a sale, or integrating an acquisition, where experienced hands are needed for a stretch, not forever. The other is a standing need a few days a month, when leadership wants a finance partner but the workload does not justify a permanent seat.

The full-time hire fits when the cadence is permanent: weekly reporting obligations to investors or lenders, continuous budget-versus-actual work, and a forecast that renews every month without fail. Hiring into an undefined process wastes the person, so define the routine first, even in a spreadsheet.

Order matters more than any single choice. Manual view first, tool second, extra hands around the decisions the numbers feed. Businesses that reverse the order, buying platforms before owning a process, pay twice: once for the subscription and once for the cleanup.

Turn the weekly view into a routine that drives decisions

The routine is what makes the view a function rather than a file. Pick a fixed slot, Monday morning works well because weekend bank activity has posted, and run the same steps every week in the same order.

The steps take under an hour once practiced. Pull actual balances from the bank, replace last week’s forecast with what happened, roll the window forward so it always covers thirteen weeks, and record the difference between forecast and actual for the week that just closed.

That difference, the variance, is the learning engine. A collection that lands two weeks late every month tells you to shift that customer’s line permanently. Spending that overshoots one category points to where estimates are weak. Patterns, not single misses, are what you act on.

Grow the function module by module. Once the cash view holds up for a couple of months, add a simple profit forecast, then a budget with monthly comparison, then scenario copies that ask specific questions: what if the largest customer pays a month late, what if hiring slips two quarters. Each addition should answer a question someone asked out loud.

Reporting benefits follow. Investor updates and lender conversations carry more weight when the numbers come from a weekly discipline instead of a scramble. One distinction earns respect fast: recognized revenue follows rules such as ASC 606, while cash arrives on its own schedule, and a finance function that explains the gap between the two sounds like it knows the business.

Ownership stays singular. One person updates, one decision maker reviews, and the meeting produces notes: which invoice to chase, which payment to move, which hire to date. Ten minutes of decisions is the product; the spreadsheet is the input.

What the weekly record is worth in a financing or a sale

When a lender, an investor, or a buyer asks about cash, they are really asking whether the business knows itself. A company that can produce thirteen weeks of forecast history, with the misses explained, answers that question before it is asked. One that assembles numbers the night before answers it badly.

Lenders often ask for a cash forecast as part of working capital or asset-based facilities, and they often ask how the forecast has performed against actuals. The weekly file produces that comparison as a byproduct: each week’s variance, accumulated, is the track record. Explaining a miss with a reason and a correction reads better than a forecast that has never been wrong because it has never been checked.

A sale process raises the stakes. Diligence teams test receivables by comparing payment histories to the aging report, so the receipt dates in your forecast need to survive that test. The cleanup described earlier, matching every open invoice to a real expectation, is exactly the preparation that shortens the questions later.

After an acquisition closes, the same view becomes the integration instrument. The acquired company’s payroll, payables, and receivables can be folded into the thirteen-week window, and the combined picture shows where cash actually sits before the bank accounts and systems are merged. Planning cash that way avoids surprises in the weeks when control of the accounts is in transition.

The record also shapes the capital structure conversation. A business that can show when cash tightens, by how much, and for how long can argue about the right instrument: a line of credit for timing gaps, term debt for purchases that pay back over years, equity for a gap that only growth closes. Without the weekly picture, that argument defaults to whatever money is offered first.

None of this guarantees an outcome, and anyone promising one should be asked to explain the mechanism. What the discipline provides is readiness: shorter assembly times, fewer unexplained numbers, and questions answered from records instead of reconstruction.

What to do when the view shows a squeeze coming

Seeing a tight week nine weeks early is the payoff of the whole exercise, because early sight converts a crisis into a scheduling problem. Work the levers in order of least damage.

1. Call the customers behind your largest open invoices; a phone call moves payment dates more often than a reminder email.
2. Ask key suppliers about aligning payment dates with your receipt patterns; many prefer a slower sure payment over a fast missed one.
3. Shift discretionary spending, such as equipment, marketing campaigns, and hiring dates, by a few weeks.
4. Use short-term credit only when the gap is one of timing, not of a business losing money every quarter.

Talk to your bank before the tight week, not during it. A lender hearing about a gap eight weeks out, with a schedule showing how it closes, is in a different conversation than one hearing about it after a payment has bounced. The thirteen-week format travels well into exactly those discussions, so the artifact you already maintain is the one the conversation wants.

A worked example, entirely hypothetical, shows the arithmetic. Say a distributor holds $400,000 today. Payroll of $75,000 lands on the 15th and again on the last day of the month. An inventory purchase of $200,000 is scheduled for week nine, with receivables of $180,000 and $120,000 expected in weeks ten and eleven. Leaving the purchase in week nine takes the balance near $50,000 before the receivables land; moving it two weeks later keeps the lowest week comfortable.

The forecast also distinguishes a timing gap from a structural one. If every thirteen-week window closes lower than the last, the business is spending more than it collects, and no amount of scheduling fixes that. The view’s job is to reveal the pattern early; the response, whether repricing, cutting costs, or raising capital, is a separate decision.

Resist the urge to hide ugly weeks from the model. A forecast polished to look safe gets ignored the first time reality disagrees with it. The version that shows the bad week and the plan for it keeps its audience, including lenders, who generally prefer early bad news to late bad news.

The costliest mistake: a forecast nobody updates

The most expensive failure mode in this work is the abandoned forecast: a careful 13-week view built once, presented once, and left to go stale while decisions march on. The consequence is false confidence, which is worse than having no forecast at all, because it feels like information.

Abandonment rarely comes from laziness. It comes from missing ownership, no calendar slot, and updates happening only when fear spikes. When the scary week passes, the ritual dissolves until the next one, and each restart costs more than the last.

The avoidance is unglamorous and reliable. One named owner. A fixed weekly slot that survives busy weeks, shortened when needed rather than skipped. A short log of changed assumptions so the history explains itself. And a standing question at each review: which decision does this week’s update change?

Built that way, the function grows along the path described above: cash first, then profit forecast, budget, scenarios, and cfo services reporting, each added when a real question demands it. A year in, the weekly page still comes first, because it is still the fastest honest answer to the question every tight period asks.

Start with what you know: this week’s balance, next week’s payroll, the invoices actually in the drawer. Thirteen rows later you have the first piece of a finance function, and unlike software bought in a hurry, this one already works.