47 Financial Analyst Interview Questions, From the 3 Statements to the CFO Pushback Case

Accrual accounting records revenue when it's earned and expenses when they're incurred, regardless of when cash actually moves. Cash accounting records both only when cash changes hands. GAAP requires accrual accounting for anything beyond the smallest businesses, which is exactly why net income and cash flow diverge, and why the cash flow statement exists as a separate document in the first place.

COGS captures the direct cost of producing whatever the company sells, materials, direct labor, manufacturing overhead tied to the product. Operating expenses cover everything else needed to run the business that isn't directly tied to production: sales and marketing, R&D, G&A. The line matters because gross margin (revenue minus COGS) tells you about pricing power and production efficiency, while operating margin tells you about how disciplined the whole organization is.

EBITDA adds depreciation and amortization back to EBIT, stripping out both the effect of financing decisions (interest) and accounting decisions around fixed asset write-off timing (D&A). For two companies with different debt loads but similar operations, EBITDA gives a cleaner read on core operating performance because it removes interest expense entirely, which is a financing choice, not an operating one. EBIT still reflects each company's own D&A policy, which can vary for reasons unrelated to how the business actually performs.

Similar business model and revenue drivers matter more than same industry classification on paper. Similar growth rate and margin profile. Similar size, since scale affects multiples in ways that are hard to normalize away. Similar capital structure, or at least similar enough that you're using the right metric (EV/EBITDA vs P/E) to neutralize the differences that remain. A "comp set" of five companies in the same GICS sector with wildly different growth rates isn't actually a comp set, it's a list.

VLOOKUP only searches left to right and breaks if someone inserts a column, since it references a fixed column number. INDEX/MATCH (or the newer XLOOKUP) searches in either direction and references column positions dynamically, so inserting or reordering columns doesn't silently break the formula. It's a small habit, but it's one of the fastest tells for whether someone has actually maintained a model long-term versus built one once and moved on.

Consistent formatting conventions: blue for hardcoded inputs, black for formulas, green for links to other sheets or files. Formulas that read left to right and top to bottom without jumping across the workbook unpredictably. No hardcoded numbers buried inside a formula. And a habit of checking that a model's outputs still make sense after changing one input, not just trusting that the formula ran without an error.

Organize by category rather than listing ratios randomly, it shows you understand what each one is actually testing, not just that you memorized formulas. See the table above for the four buckets: liquidity, use, profitability, and efficiency. A company can look fine on profitability and still be in real trouble on liquidity, which is exactly the kind of gap this question is designed to surface.

Assess urgency and complexity honestly rather than defaulting to whoever asked loudest or most recently. Communicate proactively with each stakeholder about realistic timelines rather than silently letting one deliverable slip. If genuine triage is needed, say so directly, "I can get you a rough estimate by Thursday, but the fully validated version needs until Monday," rather than overpromising and delivering late without warning.

"EBITDA gets misused as a proxy for cash flow in capital-intensive industries where depreciation represents real replacement costs."

That's a genuinely defensible answer, and it's the kind of specific, slightly opinionated take this question is fishing for. A manufacturer or telecom replacing physical equipment every few years has depreciation that reflects a real, recurring cash outflow eventually, treating EBITDA as if it's equivalent to free cash flow overstates how much cash the business actually generates. Have your own version of this answer ready. A vague "I think all metrics have limitations" tells the interviewer nothing about how you actually think.

Not the DCF mechanics themselves, most candidates can recite those. Across financial analyst interview questions in general, the hardest part is the follow-up: interviewers change one assumption and ask you to trace the effect through all three statements or the whole valuation, live, without a spreadsheet in front of you. That's the question that separates candidates who understand the model from candidates who memorized it.

No, not for most corporate financial analyst roles, though it can help for investment management or research-heavy positions. What actually gets candidates through the technical round is fluency with the three statements, DCF mechanics, and Excel, which a CFA curriculum covers but doesn't guarantee you can explain out loud under pressure. The 2025 CFA Institute survey found specialized skills, not credentials, are what graduates themselves rank as the biggest factor in advancement.

Less technical on average, but the gap has narrowed since 2023. Corporate FP&A and industry-side financial analyst roles focus more on budgeting, variance analysis, and business partnering than on building an LBO from scratch. That said, valuation and DCF questions increasingly show up even outside banking, especially at companies doing frequent M&A or capital allocation work, so treat the core technical sections in this guide as necessary regardless of which side of the industry you're interviewing on.

Most candidates report 3 to 6 weeks for corporate roles, a bit longer for investment management or banking-adjacent positions where multiple rounds and case studies are standard. A typical loop runs a recruiter screen, one or two technical rounds covering statements and valuation, and a final round mixing case work with behavioral questions and a conversation with the hiring manager or team.

Excel is still the baseline expectation for almost every financial analyst role, full stop. SQL has become increasingly common as a secondary skill, especially at companies where financial data lives in a warehouse rather than a set of exported spreadsheets, but it rarely substitutes for Excel fluency as the core requirement. If a job posting lists SQL, expect basic query and join questions layered on top of the standard modeling and statement questions in this guide, not instead of them.

LIFO (last-in, first-out) assumes the most recently purchased inventory is sold first. FIFO (first-in, first-out) assumes the oldest inventory is sold first. During inflation, LIFO matches higher, more recent costs against revenue, so COGS is higher and net income is lower than under FIFO. Lower net income also means lower taxable income, which is why LIFO is popular in the US for tax purposes. The IRS conformity rule requires companies that use LIFO for tax filing to also use it for financial reporting, so you can't pick the flattering method for shareholders and a different one for the IRS.

FIFO shows lower COGS and higher net income during inflation because it matches the oldest, cheaper costs against current revenue, leaving newer, more expensive inventory on the balance sheet. That's part of why FIFO balance sheets show inventory values closer to current replacement cost, while a company that's held LIFO layers for decades can carry inventory on its books at a fraction of what it would cost to replace today. One wrinkle worth knowing: IFRS doesn't permit LIFO at all, only FIFO or weighted average, so when you're comparing a US company against an international peer, check the 10-K footnote for the LIFO reserve and add it back to make COGS and inventory comparable.

Net income sits at the bottom of the income statement and flows into retained earnings on the balance sheet through shareholders' equity. Depreciation and amortization get added back on the cash flow statement because they reduced net income without actually costing any cash. Capital expenditures show up in the investing section of the cash flow statement and increase property, plant, and equipment on the balance sheet. Changes in working capital, receivables, payables, inventory, reconcile net income to the actual cash a business generated in a period.

Interviewers almost always follow this with the depreciation walk-through below. Treat this answer as the setup, not the whole test.

Growth eats cash before it shows up as profit. A company can book a sale, recognize the revenue and net income immediately, and still wait 60 or 90 days to actually collect the cash from the customer. Do that at scale while also building inventory ahead of demand and spending on capex to support growth, and operating cash flow can run negative for years while net income stays positive the whole time.

This is the plain-English version of what shows up on an income statement as "profitable" and on a cash flow statement as a going-concern problem. Retailers and fast-growing SaaS companies both hit this, for almost opposite reasons.

Working capital is current assets minus current liabilities, most practically receivables plus inventory minus payables. Negative working capital sounds alarming but isn't always. Subscription businesses and some retailers collect cash from customers before they deliver the service, which shows up as deferred revenue, a current liability. Grow that business fast enough and current liabilities can outpace current assets even while the company is healthy and growing.

Deferred revenue sits on the balance sheet as a current liability when a company collects cash for a good or service it hasn't delivered yet, an annual software subscription paid upfront is the classic case. As the company delivers the service over the contract term, it recognizes a slice of that revenue on the income statement each period and reduces the deferred revenue liability by the same amount. No cash moves at recognition. It already moved when the customer paid.

The argument for adding it back: SBC is non-cash, so if you're trying to measure cash-generating ability, it doesn't belong in the calculation the same way depreciation doesn't. The argument against it: SBC is a real economic cost. It dilutes existing shareholders even though no cash leaves the company, and companies that hand out large equity grants every year are transferring value away from current holders whether or not it shows up as a cash expense.

I don't think there's a clean answer here that works for every company. A reasonable interview response says both things are true and then explains how you'd adjust your model for dilution separately instead of pretending the SBC add-back makes the issue disappear.

Goodwill is the premium a company pays over the fair value of a target's identifiable net assets in an acquisition, it captures things like brand value, customer relationships, and expected synergies that don't show up as a specific line item. Companies test goodwill for impairment at least annually, and take a write-down when the reporting unit's fair value drops below its carrying value, usually because the acquisition underperformed the assumptions used to justify the original price.

Watch for receivables growing meaningfully faster than revenue over several quarters, that's a sign a company is booking sales it hasn't collected yet, sometimes to customers who may never pay. Check whether revenue recognition timing shifted after a change in accounting policy or estimate. Compare deferred revenue trends against the sales narrative in earnings calls. And read the revenue recognition footnote itself, not just the MD&A summary, since that's where judgment calls about bundled contracts and multi-year deals actually get disclosed.

WACC weights the cost of equity (usually via CAPM: risk-free rate plus beta times equity risk premium) and after-tax cost of debt by their respective proportions in the capital structure. A 1-point increase in WACC lowers every discounted cash flow and the discounted terminal value, since you're dividing by a larger denominator at every stage. The effect compounds more heavily on the terminal value than on the near-term cash flows, since terminal value usually represents 60 to 80 percent of total enterprise value in a standard DCF, so a small WACC change moves the total valuation more than intuition suggests.

Perpetuity growth: terminal value equals the final year's free cash flow, grown one more year at a long-term growth rate (usually 2 to 3 percent, roughly GDP growth), divided by WACC minus that growth rate. Exit multiple: apply a comparable company's EV/EBITDA multiple to the final projected year's EBITDA. Good practice is to run both and check that the implied growth rate from the exit multiple method and the implied exit multiple from the perpetuity method land in a sensible range, if they don't agree, one of your assumptions is off.

Unlevered free cash flow (before interest and debt paydown) discounted at WACC gets you enterprise value, which is capital-structure neutral, useful when comparing companies with different debt levels or valuing a business independent of how it's financed. Levered free cash flow (after interest and mandatory debt paydown) discounted at cost of equity gets you equity value directly, more common in LBO models where the debt structure itself is a key part of the analysis, not something you want to abstract away.

Enterprise value represents the value of the whole business, available to both debt and equity holders. Equity value is what's left for shareholders after debt claims are settled. The bridge from one to the other:

• Enterprise value plus cash and equivalents
• Minus total debt
• Minus preferred stock and minority interest

The distinction matters most when comparing companies with very different use. Two businesses with identical operations but different debt loads will have different equity values even at the same enterprise value, so using equity-value multiples like P/E to compare them across capital structures produces a misleading answer.

P/E is affected by capital structure (interest expense reduces net income) and by tax rate differences across jurisdictions. EV/EBITDA strips out both financing decisions and taxes, leaving a cleaner comparison of core operating performance. That's why bankers default to EV/EBITDA for cross-company comps and save P/E mostly for comparing companies within the same capital structure and tax environment, banks and REITs being the obvious exceptions.

Financial services and real estate are the two big exceptions. Banks don't have a meaningful "EBITDA" in the traditional sense, their core business is managing a spread on assets and liabilities, so analysts use P/E, price-to-book, or price-to-tangible-book instead. REITs use FFO (funds from operations) since depreciation on real estate doesn't behave like depreciation on equipment, the underlying asset often appreciates rather than wears out, which makes standard D&A add-backs misleading.

Data tables in Excel (or a two-way sensitivity grid) are the standard approach: pick the two assumptions that swing the valuation most, usually WACC and terminal growth rate or exit multiple, and vary each across a realistic range on a grid, with implied share price or enterprise value in each cell. The point isn't the single "right" number, it's showing the interviewer you understand which assumptions the whole valuation actually hinges on, and which ones barely move the needle.

First, check your own assumptions before assuming the comps are wrong. Terminal growth rate too high, WACC too low, margins expanding beyond what the company has ever achieved historically, these are the usual culprits. Second, ask whether there's a real reason the business deserves a premium, faster growth, structurally higher margins, a moat the comps don't have. If neither explains the gap, the honest answer in an interview is that you'd flag the discrepancy rather than quietly present the higher number as if it settles the question.

Start with NPV: project the incremental cash flows the capex generates, discount them at the company's cost of capital, and subtract the initial outlay. A positive NPV means the project creates value. IRR is the discount rate at which NPV equals zero, useful for comparing projects of different sizes since it's expressed as a percentage rather than a dollar figure. Layer in sensitivity analysis on the key assumptions (utilization, pricing, useful life), then do a strategic gut-check separate from the math: does this project fit where the business is actually trying to go, independent of whether the spreadsheet says yes.

Every input, growth rate, price, unit volume, ASP, lives in its own labeled cell, usually on a dedicated assumptions tab, colored blue by convention so anyone opening the file instantly knows what's an input versus a formula. Every calculation references those cells rather than typing numbers directly into formulas. Change the growth rate assumption once, and the entire model, income statement through valuation output, updates automatically. Hardcoded numbers buried inside formulas are the single most common thing that breaks a model when someone other than the original builder has to update it six months later.

Rows: department or cost center. Columns: budget, actual, and variance (actual minus budget, with a percentage variance column alongside the dollar figure). Values: the underlying expense or revenue amounts, summed. Sort by absolute variance to surface the biggest gaps first, since a $2,000 miss on a $10,000 line matters more than a $50,000 miss on a $5 million line, and a pivot sorted purely by dollar variance without the percentage column hides that.

A dropdown cell (data validation list) drives a scenario selector, and every key assumption references a lookup table with a value for each scenario, pulled in with an INDEX/MATCH or CHOOSE function keyed to the dropdown. Toggle the dropdown and every downstream calculation updates to reflect that scenario's assumptions. The alternative, three separate model copies for three scenarios, is a maintenance trap. Update one assumption and you have to remember to update it in two other files too, and eventually someone won't.

SUMIFS is faster to write, easier for someone else to audit later, and handles most conditional summing needs (multiple criteria across multiple columns) without needing to understand array formula syntax. Array formulas (or newer dynamic array functions) handle cases SUMIFS can't, conditional logic that depends on a calculation rather than a simple match, or returning multiple results at once. Default to SUMIFS unless you have a specific reason not to. A model full of unnecessary array formulas is harder for the next person to maintain, and "the next person" is often you, six months from now.

Trace key outputs backward through the formula chain to confirm each input flows where it should. Check for hardcoded numbers hiding inside formulas, they're the most common source of a model quietly disagreeing with its own assumptions tab. Stress-test with an extreme input, set growth to zero or negative, and confirm the model doesn't break or produce a nonsensical result. And check that formulas are consistent across a row, one different formula hiding in the middle of an otherwise uniform row is one of the most common ways an error survives multiple reviews.

The strongest answers show a clear separation between providing rigorous analysis and controlling the outcome. Present the recommendation your data actually supported, acknowledge the decision-maker weighed factors beyond what was in your model (market timing, relationships, risk tolerance you didn't have full visibility into), and describe how you executed on the final call professionally rather than passive-aggressively. Interviewers are checking whether you can be right and still be a good teammate about it.

Lead with the headline number, then two or three specific causes with clear attribution, then what it implies going forward. No jargon, no walking someone through the variance formula before telling them what actually happened. "We came in $340,000 over budget in Q1, driven mostly by a vendor price increase we didn't anticipate and higher-than-planned overtime in the warehouse. Both are one-time in nature, and here's what we're doing about each" lands better than a spreadsheet screen-share with no framing.

This question rewards specificity and honesty over a polished non-answer. Name the actual error, a broken formula reference, a wrong tax rate assumption, a sign error in a variance calc, how you caught it (a sanity check against last year's actuals, a colleague's review, a number that just looked wrong on instinct), and what you changed in your process afterward so the same mistake doesn't repeat. "I don't make mistakes" reads as either dishonest or inexperienced. Neither is the impression you want to leave.

Start with the actual cash balance, then project weekly cash inflows (collections from receivables, based on realistic days-sales-outstanding assumptions, not just revenue divided by 52) and outflows (payroll, AP payments, debt service, capex) for each of the next 13 weeks. Roll the model forward each week, dropping the oldest week and adding a new one at the end, and reconcile actuals against last week's forecast to calibrate your assumptions over time. This tool matters most for companies managing liquidity tightly, it is not really about long-term valuation. It is about not missing payroll.

Income statement: pre-tax income drops by $10. Assuming a 30 percent tax rate, net income drops by $7 (10 times 1 minus 0.3). Cash flow statement: net income is down $7, but you add back the full $10 of depreciation since it's non-cash, so cash from operations is actually up $3. Balance sheet: PP&E falls by $10 from the depreciation itself, cash rises by $3 from the CFS, so total assets are down $7 net. On the other side, retained earnings falls by $7 as net income flows through, and the two sides balance.

1. Tax-affect the income statement hit first, don't skip this step.
2. Add the full depreciation back on the cash flow statement, not the after-tax figure.
3. Net the PP&E decrease against the cash increase to get the balance sheet change.

The tax-affecting step is where this one falls apart. Candidates who forget it either say cash is flat or say assets drop by the full $10, and both answers are wrong in a way that's obvious to anyone who's actually built the schedule.

Before ASC 842, operating leases lived entirely off-balance-sheet, disclosed only in footnotes. Under ASC 842, both operating and finance leases now show up as a right-of-use asset and a corresponding lease liability on the balance sheet. The difference between the two lease types today is mostly in the income statement treatment: finance leases split the expense into interest and amortization (front-loaded, higher expense early), while operating leases record a single straight-line lease expense over the term.

The practical reason this comes up in interviews: companies that leaned on operating leases to keep debt-like obligations off their balance sheet lost that advantage in 2019. use ratios calculated the old way and the new way can look meaningfully different for retailers and airlines with heavy real estate or aircraft lease books.

Project unlevered free cash flow for 5 to 10 years: revenue, margins, taxes, capex, and working capital changes, all built off explicit assumptions you can defend. Discount each year's cash flow back to present value using WACC. Calculate terminal value at the end of the explicit forecast period, either with the perpetuity growth method or an exit multiple, and discount that back too. Sum the discounted cash flows and the discounted terminal value to get enterprise value, then bridge to equity value by subtracting net debt.

Finally, sanity-check the result against public comps and precedent transactions. A DCF that spits out a number 60 percent above every comp isn't automatically wrong, but you'd better have a specific reason ready, not just "the model says so."

The revenue build. Everything else, COGS as a percent of revenue, opex ratios, working capital days, cash flow, ultimately depends on the revenue line, so getting that assumption structure right first prevents rebuilding the rest of the model later. After revenue, most experienced modelers build the income statement down through EBIT, then the balance sheet's non-cash items (PP&E schedule, working capital), and leave cash and the debt schedule for last since those are the plug that reconciles everything else.

A DCF asks what a business is intrinsically worth based on its own projected cash flows. An LBO asks a different question entirely: given a purchase price, a debt structure, and an exit multiple in 3 to 7 years, what return does the equity check generate? PE firms run LBOs because their actual decision is whether a deal clears their required IRR, typically 20 percent-plus, given realistic use and exit assumptions, not whether the business is "worth" some intrinsic number in isolation.

In practice, a PE analyst often runs both: the DCF as a sanity check on entry price, the LBO as the actual deal model that determines whether the transaction gets approved by the investment committee.

The circularity: interest expense depends on the debt balance, which depends on the cash sweep, which depends on cash flow, which depends on net income, which depends on interest expense. The standard fix is a circularity breaker switch, a cell that, when toggled to zero, forces interest expense to reference the prior period's ending balance instead of the current period's average, breaking the loop so the model can calculate without Excel's iterative calculation setting causing errors or crashes.

Alternatively, use average debt balance with a manual iteration toggle in Excel's calculation options. Either way, know why the circularity exists in the first place. Interviewers ask this specifically to filter out people who've copied a model template without understanding its mechanics.

Present the numbers accurately, with your assumptions clearly stated, and flag which specific assumptions are most sensitive to the gap between what the model shows and what the CFO expected. Don't adjust inputs to manufacture a number that fits the narrative, and don't ambush the CFO in a room full of other people either. The strongest version of this answer describes giving the CFO a heads-up before the broader meeting, so they're not caught off guard in front of their own leadership team.

The Altman Z-score combines five financial ratios into a single number that estimates the likelihood a company files for bankruptcy within two years. For public manufacturing companies the formula is Z = 1.2(working capital/total assets) + 1.4(retained earnings/total assets) + 3.3(EBIT/total assets) + 0.6(market value of equity/total liabilities) + 1.0(sales/total assets). A score above 2.99 is generally considered safe, below 1.81 signals distress, and the range in between is a gray zone where you'd dig into qualitative factors like covenant headroom or customer concentration before drawing a conclusion.

The limitations matter more than the formula. It was built in the 1960s on a small sample of manufacturing firms, so running it directly on a bank, a SaaS company with no meaningful inventory, or an early-stage company with negative retained earnings will produce a misleadingly low score even if the business is fundamentally healthy. Altman later published separate versions, a Z' score for private companies and a Z'' score for non-manufacturers, that drop the market value of equity term since private companies don't have a market cap. In practice I treat the Z-score as a screening tool that flags names for closer review, not a standalone credit decision, the same way you wouldn't approve a loan based on a FICO score alone.

A deferred tax liability arises when a company recognizes an expense or income item for book purposes before it recognizes it for tax purposes, the classic example being accelerated tax depreciation against straight-line book depreciation, so the company pays less tax now and more later. A deferred tax asset is the reverse: items like net operating loss carryforwards, warranty reserves, or bad debt allowances that get expensed for book purposes before they're deductible for tax, meaning the company will realize a tax benefit in a future period.

The valuation allowance question is where this gets interesting. Under ASC 740, a company has to assess whether it's more likely than not that it will generate enough future taxable income to actually use its deferred tax assets, particularly NOLs. If a company has a history of losses, or its NOLs are set to expire before it turns profitable, management has to record a valuation allowance that reduces the DTA's carrying value, and that reduction runs through the income statement as additional tax expense. That's exactly the kind of adjustment that can turn an otherwise-profitable quarter into a reported net loss, or the reverse: a company releasing a valuation allowance because it's finally confident in future profitability can show a one-time earnings pop that has nothing to do with operating performance. When I'm looking at a company carrying a large DTA, the first thing I check is the effective tax rate reconciliation in the 10-K to see whether a valuation allowance change is distorting the reported numbers.

A maintenance covenant requires the borrower to test a financial ratio, say a maximum use ratio or minimum interest coverage ratio, on a regular schedule, usually quarterly, regardless of what the company does. Breach it and the lender can call a default even if the company hasn't taken any new action. An incurrence covenant only gets tested when the company takes a specific action, like issuing new debt, paying a dividend, or making an acquisition. If the company just sits still and its metrics deteriorate, there's no automatic trigger.

Covenant-lite loans, which became the norm in the used loan market through the 2010s, strip out the maintenance tests and rely only on incurrence covenants. That shifts real risk onto lenders: a borrower's credit can deteriorate significantly, sometimes well into real distress, before any covenant actually trips and gives lenders use to force a restructuring conversation or extract better terms. As a credit analyst this changes how you monitor a covenant-lite credit. You can't wait for a covenant breach as your early warning signal, so you build your own independent use and coverage tracking off the borrower's financials and compare it against the documented incurrence baskets, the capacity a borrower has to add debt or pay dividends before even an incurrence test would apply. This matters even more with floating-rate borrowers, whose interest coverage can deteriorate quickly in a rising-rate environment with no maintenance test to catch it early.

Income statement: the write-down hits COGS or a separate impairment line, reducing pre-tax income by $100 and, at a 30 percent tax rate, net income by $70. Cash flow statement: net income is down $70, but the write-down is added back as a non-cash item, so operating cash flow is actually up $30. Balance sheet: inventory drops by $100, cash rises by $30, so assets fall by $70 net, matched by a $70 drop in retained earnings.

Notice this is structurally identical to the depreciation walk-through above, just with a different asset and a different non-cash charge. Once you understand the pattern, you can answer this question for almost any non-cash write-off an interviewer throws at you.

Accretion/dilution analysis answers one question: does the combined company's pro forma EPS go up (accretive) or down (dilutive) relative to the acquirer's standalone EPS after the deal closes. Start by combining net income from both companies, then adjust for the financing structure. If new debt funds the deal, subtract after-tax interest expense on that debt. If acquirer stock funds it, add the new shares issued to the diluted share count. If it's cash off the balance sheet, subtract the after-tax opportunity cost of that cash, meaning the interest income the company gives up. Then layer in purchase accounting adjustments, incremental depreciation and amortization from stepping up the target's fixed assets and intangibles, plus any cost synergies the deal team is willing to underwrite in the model.

The quick shortcut bankers use is that if the acquirer's P/E multiple is higher than the target's in an all-stock deal, the deal is accretive, because you're buying cheaper earnings with more expensive stock. That shortcut breaks down fast once you add a cash-and-debt mix, different tax rates between the two companies, or meaningful synergies, so I build the full bridge rather than leaning on the multiple comparison alone. One thing junior analysts miss: year-one accretion doesn't prove the deal creates value. A deal can be EPS-accretive and still destroy value if the acquirer overpays relative to the target's actual cash flow generation, which is why accretion/dilution is a financing sanity check, not a substitute for a DCF or a return-on-invested-capital analysis on the deal itself.