Learn how to calculate allowance for doubtful accounts, reduce bad debt write-offs, and automate credit risk management. See how modern AR teams do it faster.
Shyam Agarwal 
Nobody likes writing off bad debt. But if you're running AR for a mid-market or enterprise company in the US, you already know it's part of the job. The question isn't whether some invoices will go uncollected. It's whether you've estimated that number accurately enough to protect your financial statements, satisfy your auditors, and make smart credit decisions going forward.
That's where the allowance for doubtful accounts comes in. It's one of those accounting concepts that sounds simple in a textbook and turns into a genuine headache in practice. This guide breaks down how it actually works, why most companies get it wrong, and what better looks like when you have the right tools and processes behind you.
The allowance for doubtful accounts is a contra asset account on your balance sheet that offsets your gross accounts receivable. It represents your best estimate of how much of what you're owed today you'll never actually collect.
It's not a guarantee. It's not a write-off. It's an estimate, recorded in advance, so your financials reflect a realistic picture of what your AR is actually worth.
Under US GAAP, you're required to use the allowance method rather than the direct write-off method for companies whose bad debts are material. That means you can't just wait until a customer definitively stiffs you and then expense it. You record the expected loss in the same period you record the revenue. That's the matching principle at work.
So how do you arrive at that estimate in the first place? There are two widely used approaches.
The percentage of sales method takes a fixed percentage of credit sales during the period and records that as bad debt expense. It's simple and income statement-focused. If your historical write-off rate is 1.5% of credit sales and you did $20 million this quarter, you'd record $300,000. The problem with this approach, especially at higher revenue volumes, is that it can drift away from the actual aging reality of your receivables portfolio.
The aging of accounts receivable method is more granular and, in most cases, more defensible. You bucket your outstanding invoices by how long they've been unpaid (0-30 days, 31-60, 61-90, 90+) and apply different expected loss rates to each bucket. The older the bucket, the higher the percentage. A 90-day invoice is materially riskier than a 15-day one.
For a manufacturing company carrying $80 million in AR across 400 active customer accounts, the aging method gives you a far more accurate reserve and a much clearer signal about where your collectors need to focus. You can also blend both approaches as a cross-check.
There's a third method worth knowing: the specific identification method. This is used when you have a particular customer you believe is in serious trouble and you want to reserve against that specific balance. Think of a construction subcontractor whose GC just filed for Chapter 11. You're not waiting for the aging bucket to catch up. You're reserving against that balance now.
For more context on how receivables are classified and tracked, this guide on types of accounts receivable is worth reading.
Let's be honest: most companies underestimate their allowance for doubtful accounts. It's not always intentional. It's often a data problem.
Picture a $500 million wholesale distribution company managing invoices across 1,200 customers. Their AR team runs aging reports out of their ERP, manually exports to Excel, and applies rough percentage estimates that haven't been revisited since the CFO before the current one set them up. They've got a 1% reserve rate across the board, which made sense three years ago when their customer base looked different. But now they have 40 new accounts in a sector that's been hit hard by supply chain disruptions, and nobody's updated the risk profile.
When those accounts start going bad, the write-offs hit all at once. The income statement takes a surprise hit. The auditors ask uncomfortable questions. The CFO is explaining variance to the board.
That's a real problem. And the fix isn't complicated once you see it clearly.
Underreserving distorts your financial picture, which can affect everything from covenant compliance to investor confidence. Overreserving, on the other hand, hits your reported earnings unnecessarily. There's genuine skill in finding the right number.
The consequences of inefficient AR management go well beyond bad debt expense. They ripple into cash flow, credit decisions, and strategic planning.
Your allowance calculation is only as good as the data feeding it. And in most companies, that means your AR aging report is either your best friend or your biggest blind spot.
A clean, current aging report tells you exactly where your risk is concentrated. It shows which customers are slipping past terms, which accounts have disputed invoices holding up payment, and which segments of your portfolio are trending worse than usual. Without that visibility, you're estimating in the dark.
The problem is that in high-volume AR environments, aging reports are often stale by the time they get acted on. A distribution company processing 50,000 invoices a month doesn't have time to review each line. Collectors are triaging manually, escalating based on dollar size, and hoping the smaller-balance delinquencies don't add up to a material miss.
Real-time AR aging reports change this dynamic. When your collectors can see exactly what's aging and by how much, and when your finance team can pull that same data on demand, your allowance calculation goes from a quarterly guess to a continuously calibrated estimate.
This is where people sometimes get confused, especially when they're reviewing financial statements for the first time or preparing for an audit.
Bad debt expense shows up on the income statement. It reduces net income. The allowance for doubtful accounts shows up on the balance sheet as a contra asset, netting against gross AR to show what you reasonably expect to collect. Together, they reflect the same underlying reality from two different angles.
When your allowance account needs adjustment (because write-offs exceeded your estimate, or your portfolio got riskier), you run the adjusting entry through bad debt expense on the income statement. That's what makes auditors pay close attention to the adequacy of your allowance. If your reserve is consistently too low, you're going to have income statement surprises at write-off time. If it's consistently too high, you're being overly conservative in ways that can affect reported earnings.
For a deeper look at how bad debt expense sits on the balance sheet, including how auditors evaluate it, that resource covers the mechanics clearly. And if you want to understand the balance sheet approach specifically, this piece on the balance sheet approach to bad debt walks through the logic.
Here's the thing most AR teams miss: your allowance for doubtful accounts is a lagging indicator. By the time you're reserving against a customer, the credit decision has already been made. The real leverage point is upstream, at the moment you extend credit.
Strong credit management practices, setting appropriate limits, running regular credit reviews, flagging customers whose risk profile is changing, directly reduce the size of your allowance over time. If you're extending credit to customers who can't pay, no amount of reserve methodology will fix that. You'll just be writing off more accurately.
This is why the most effective AR teams treat credit risk management and collections as parts of the same system. They're watching the same customers, working from the same data, and adjusting both credit exposure and collection urgency based on real-time signals.
Platforms like Quick Receivable build both functions into a single workflow. AI-powered credit risk scoring runs continuously against your customer portfolio, so the team knows which accounts are deteriorating before they hit the 90-day aging bucket.
Manual allowance calculations have a ceiling. You can only be as accurate as your data, and you can only process as much data as your team has time for. That's been the fundamental constraint for most mid-market AR teams.
Automation changes that equation.
When your AR platform can continuously analyze payment behavior, flag changes in customer risk profiles, and update your aging data in real time, your allowance estimate gets sharper. Not because the methodology changed, but because the data inputs improved. You're not estimating based on a monthly export anymore. You're working with something that reflects what happened yesterday.
There's a meaningful difference between an AR team doing this with spreadsheets and one using purpose-built accounts receivable automation tools. The latter can handle a portfolio of thousands of accounts with the same accuracy that used to require a team twice the size.
Consider an equipment rental company running $200M in annual AR across 800 contractor accounts. Their collections team used to spend two days every month pulling aging, building reserve schedules, and running the numbers through Excel. Now that process takes a few hours, it's more accurate, and it surfaces the accounts that need immediate attention rather than requiring the team to find them manually. That's a real productivity shift with direct financial impact.
If you want to see the ROI of this kind of change for your own numbers, the AR automation ROI calculator is a useful starting point.
You don't need to overhaul everything at once. A few targeted improvements can meaningfully improve how accurately you're estimating your doubtful accounts.
Start with your historical write-off data. Pull three to five years of actual bad debt write-offs by customer segment, industry, and aging bucket. You'll probably find that the loss rates vary significantly across those buckets. Applying a single blended rate across your whole portfolio is almost certainly leaving your estimate less accurate than it could be.
Review your reserve rates at least quarterly. Markets change, customer industries cycle, and a rate that was appropriate in 2022 may not be right today. Build a calendar trigger for this.
Build segment-specific rates where you have enough data. A construction company with 300 accounts spanning general contractors, subcontractors, and specialty trades will see very different payment patterns in each group. Treat them differently in your model.
Flag high-risk individual accounts for specific identification reserves rather than letting them sit in the general pool. If you have one customer that represents 8% of your AR and they're showing distress signals, that's not a statistical event. That's a specific risk that deserves a specific reserve.
Finally, connect your collections and credit teams more tightly. The people calling customers every day are picking up on signals that don't show up in aging reports yet. If a controller is noticing that a customer is suddenly disputing everything, or that their payments have slowed from 35 to 60 days, that intelligence belongs in your reserve model.
For the broader picture of how all these elements fit together, this overview of accounts receivable management covers the full operational cycle.
Your external auditors pay attention to your allowance. It's a significant judgment area, and they're going to probe the assumptions behind your estimate. Being able to show documented methodology, historical validation (how your estimates compared to actual write-offs), and regular review processes will make those conversations much easier.
Companies that struggle in this area are usually the ones whose allowance process lives in a single person's head or a spreadsheet with no audit trail. If your CFO asked you today to walk through exactly how you arrived at your current reserve, could you do it in 15 minutes with documented support? If not, that's worth fixing before your next audit cycle.
The allowance for doubtful accounts is one of those areas where the accounting is straightforward but the execution is genuinely hard at scale. Getting it right requires clean data, good historical analysis, strong credit management upstream, and the discipline to revisit your assumptions regularly.
Most companies are doing this manually, which means they're doing it slowly and with more error than necessary. The upside of getting sharper here isn't just cleaner financials. It's better credit decisions, fewer surprises, and a collections team that's focused where the risk actually is.
If your current process is held together with spreadsheets and quarterly guesswork, it's worth seeing what modern tools can do. Schedule a conversation with the Quick Receivable team to see how AI-powered AR management looks in practice: book a time here.
40%
DSO reduction70%
less manual work95%
cash match accuracyFind out exactly how much time and money your AR team can save with Quick Receivable. No commitment, no setup required.
Managing 175,000 invoices per month in production