Quality Of Earnings Analysis

Getting a QoE Done Right

Quality Of Earnings Analysis is one of those processes that looks straightforward on paper and falls apart the moment you open a real company's books. You sit down with a P&L, you start making adjustments for add-backs, and you think you're done. Then you find the revenue recognition issue that eats half your adjustment schedule. Or the working capital trap where the seller pulled a distribution right before close and called it normal operations. Here's how it actually works in practice.

The Core Approach

You start with the financial statements. Most of the time these are GAAP or IFRS reports, but sometimes they're management accounts prepared with a loose hand. You need to verify which before you commit to any numbers. The goal is to strip out everything that isn't representative of normalized recurring earnings and arrive at a figure a buyer would actually be willing to underwrite a debt facility around. The standard add-back categories are the ones you already know: owner compensation above market, non-recurring legal settlements, one-time consulting fees, restructuring costs, and the usual suspects. The real work happens in the details. I spent three weeks on a manufacturing deal last year where the seller had layered a holding company management fee through the operating entity. It looked like an ordinary SG&A expense on the trial balance. The fee was disclosed in the notes to the financial statements, buried under "related party transactions." If you hadn't traced it back to the parent company, you would have adjusted it as a recurring expense. That's the kind of thing that surfaces when you're actually digging instead of just scanning the P&L.

Revenue Quality Checks

This is where most QoE reports get it wrong. Everyone focuses on earnings adjustments and barely touches the top line. But revenue quality drives everything downstream. Channel stuffing, bill-and-hold arrangements, side agreements that give customers a right of return, revenue recognized before acceptance criteria are met — these are the problems that survive the initial adjustment pass and surface during due diligence months later. You need to do a cut-off test across at least two consecutive periods. Pull the last fifteen invoices in each month and the first fifteen of the following month. Verify shipment dates, delivery acknowledgments, and any acceptance clauses in the contracts. It takes about four hours for a small business with under two hundred monthly transactions. More than that and you stratify by customer and materiality threshold. Also check concentration. If one customer represents more than twenty percent of revenue and their contract has a twelve-month auto-renewal with a sixty-day cancellation window, you're carrying risk that the QoE model won't show you unless you flag it. Buy-side deals fail on concentration assumptions more often than they fail on the actual EBITDA number.

Working Capital Normalization

Working capital in these deals is almost always defined as a target level in the purchase agreement. The buyer gets a true-up at closing based on the actual working capital delivered. The quality of the QoE depends heavily on whether you set a fair target. A common mistake is averaging the trailing twelve months without looking at seasonality. If a business ships heavily in Q4 and the trailing period includes that quarter, the average inventory and receivables will be inflated relative to what operations look like for most of the year. The buyer ends up overpaying at closing and then arguing about the true-up. I've seen this cost a buyer forty thousand dollars on a two-million-dollar deal because nobody adjusted for seasonal inventory buildup ahead of a holiday rush. The fix is straightforward. Build a monthly decomposition of current assets and liabilities. Remove outliers that are clearly non-operational — tax refunds held up in processing, intercompany balances that don't belong in the working capital calculation. Set the target as a weighted average of the months that represent normal operating conditions, not a simple arithmetic mean of the entire period.

Growth and Sustainability Assessment

A QoE isn't just a backward-looking exercise. The buyer needs to know whether the earnings you're adjusting for will actually recur. This requires a market-level check against industry benchmarks and a look at the order book or contract pipeline. If the company claims a twenty percent growth rate but the broader market is flat and there's no evidence of new contracts being signed, you're adjusting for growth that probably won't continue. Conversely, a temporary dip in margins that's explained by a known one-time investment — say, a new ERP implementation that disrupted production for six weeks — should be treated differently from a structural margin decline driven by competitive pressure. The first is adjustable. The second changes how you value the entire revenue stream going forward.

Common Pitfalls

The biggest mistake I see is treating add-backs as a ceiling rather than a floor. Every adjustment needs a source document. If the seller says they spent fifty thousand on a trade show that won't recur, you want to see the invoice, the attendee list, and ideally confirmation that no similar event is planned in the next twelve months. Without that, the adjustment is just a number. Another issue is double-counting. You adjust for a one-time legal fee in the EBITDA reconciliation and then forget that the corresponding tax effect is still baked into the net income figure. Net working capital adjustments sometimes double-count items that are already reflected in the cash balance at closing. You need a single reconciliation map that tracks every line item from reported net income through to adjusted EBITDA and makes sure each adjustment appears exactly once.

What It Can't Tell You

QoE analysis has real limits. It can't predict whether a key customer will leave next year. It can't verify that the founder's relationships will transfer to new ownership. It can't tell you whether the accounting policies used are consistent with what the industry is moving toward, especially in sectors shifting to subscription or usage-based revenue models. If the company's revenue recognition is already ambiguous under current standards, the QoE will surface inconsistencies but can't fully resolve them without extending the review period significantly. In those cases, a deeper revenue assurance engagement or a specialized forensic review is more appropriate than a standard QoE. The standard process usually delivers results within two to four weeks depending on data availability. Extending it to six weeks or more for a full revenue forensic review is the right move when the revenue model is complex or the controls environment is weak.

Practical Workflow

Start with the chart of accounts and the general ledger. Pull a trial balance for the full period under review. Reconcile it to the financial statements. If the TB doesn't tie, you've already identified a problem before you do any adjustments. Next, run the revenue tests. Confirm the cut-off, check for side agreements, test a sample of contracts for performance obligations that haven't been satisfied. Move to expense analysis after you're confident about the top line. Build the add-back schedule with supporting documentation for every item. Keep a separate file for items that couldn't be verified — disclose them as unverified adjustments. Buyers and lenders will look at those first anyway, so better to surface them proactively than have them become a negotiation point at the last minute.

Finally, prepare the normalization schedule and the bridge from reported net income to adjusted EBITDA. Include a separate section for growth and sustainability observations. The financial numbers alone don't close deals. The narrative around what those numbers mean for the future is usually what determines whether the deal proceeds at the agreed multiple or gets renegotiated hard.