Investment Banking Interview Questions

"What pulls me to banking is the transaction work. I like that there's a real event at the end, a company gets bought or raises money, and the analysis feeds straight into a decision with a deadline. Consulting interests me too, but a lot of it ends with a recommendation, and I want to be close to the numbers and the execution. A corporate finance role would teach me one company deeply, but I'd rather see many businesses and many deal types early on. I got hooked during my internship at a small advisory firm, where I helped build a buyer list and a simple comps table for a sale process. Watching that deal move from pitch to signing is what made me sure."

"I studied economics, and the course that stuck with me was corporate finance, especially the valuation module where we took a listed company apart and rebuilt its numbers. That led me to join the student investment club, where I ran the research for two stock pitches. Last summer I interned in the transaction services team at an accounting firm, doing due diligence on working capital and one-off costs for a buyout. That showed me the accounting side of deals, but I kept wanting to be on the side that decides the price and runs the process. So banking is the obvious next step: it uses the accounting I've learned, adds valuation and deal execution, and puts me on live transactions from day one."

"I looked at the deals your firm advised on over the last couple of years, and what stood out is how many are mid-sized sell-side processes. That means analysts here work on smaller teams and see more of each deal, which is what I want. I also spoke to two of your analysts, and both said juniors get real model ownership early. As for this group, I'm drawn to healthcare because the businesses are so varied, from hospitals with heavy assets to software with almost none, so you learn a lot of valuation approaches in one sector. My final-year project was on hospital operating costs, so I'd come in already knowing some of the key metrics the group looks at."

"Net income from the bottom of the income statement is the starting line of the cash flow statement. From there I add back non-cash charges like depreciation and stock-based compensation, and adjust for changes in working capital. That gives cash flow from operations. Then I add cash flow from investing, like capital spending, and from financing, like borrowing, repaying debt or paying dividends. The net change in cash, added to last period's cash, gives the cash balance on the balance sheet. Net income also flows into retained earnings in shareholders' equity, less any dividends. And the other balance sheet lines, like PP&E, debt and working capital items, change in step with what went through the cash flow statement, which is why the balance sheet still balances."

"On the income statement, operating income falls by 10. At a 25 percent tax rate, taxes fall by 2.5, so net income is down 7.5. On the cash flow statement, I start with net income down 7.5, then add back the extra 10 of depreciation because it's a non-cash charge. So cash from operations, and cash overall, goes up by 2.5. That 2.5 is really the tax saving. On the balance sheet, cash is up 2.5 and PP&E is down 10, so total assets are down 7.5. On the other side, retained earnings are down 7.5 because of the lower net income, so equity is down 7.5 and the sheet balances."

"On day one, nothing hits the income statement. On the cash flow statement, buying the equipment is minus 100 in investing and the loan is plus 100 in financing, so cash doesn't change. On the balance sheet, PP&E is up 100 and debt is up 100. After a year, the income statement takes 10 of depreciation and 5 of interest, so pre-tax income falls 15, tax falls 3.75 and net income falls 11.25. On the cash flow statement, I start with minus 11.25 and add back the 10 of depreciation, so cash is down 1.25, assuming no principal is repaid. On the balance sheet, cash is down 1.25 and PP&E is up 90, so assets are up 88.75. Debt is still up 100 and retained earnings are down 11.25, which also nets to 88.75, so it balances."

"Equity value is what the business is worth to shareholders only. For a listed company, it's the diluted share count times the share price. Enterprise value is the value of the core operating business to everyone who funds it, so both equity and debt holders. To get from equity value to enterprise value, I add debt, preferred stock and noncontrolling interest, and subtract cash, because cash isn't part of the operating business and a buyer could use it to pay down what they owe. Going the other way, from enterprise value to equity value, I do the reverse. The reason it matters is matching: revenue and EBITDA belong to all investors, so they go with enterprise value, while net income belongs to shareholders, so it goes with equity value."

"When the company issues 100 of new shares, equity value goes up by 100 because there are more shares outstanding. But cash on the balance sheet also goes up by 100, and since cash is subtracted in the bridge to enterprise value, enterprise value stays the same. That makes sense, because the operating business hasn't changed; it just has more money sitting in the bank. If it then uses the 100 to repay debt, cash falls by 100 and debt falls by 100. In the bridge, one is added and the other subtracted, so enterprise value still doesn't move, and equity value doesn't move either. In real markets the share price might react, but in the mechanics, financing moves don't change what the core business is worth."

"First I check the options are in the money. The strike of 5 is below the share price of 10, so they count. I assume all 10 are exercised, which creates 10 new shares and brings in 10 times 5, so 50 of proceeds. The method assumes the company uses those proceeds to buy back its own stock at the current price of 10, so it repurchases 5 shares. The net new shares are 10 minus 5, which is 5. So diluted shares are 100 plus 5, or 105. If the strike were above the share price, I'd leave those options out, because nobody would exercise them. And I'd use the offer price, not the current price, if I were valuing the company in a takeover."

"The three I'd always mention are trading comparables, precedent transactions and a DCF. Comps show how the market values similar listed companies today, so they work best when there's a good set of true peers. Precedents show what buyers actually paid for similar companies, which includes a premium for control, so they matter most when the client is being sold. A DCF values the company on its own forecast cash flows, so it's most useful for stable, predictable businesses, and less useful for early-stage companies where the forecast is mostly guesswork. I'd add an LBO analysis to show what a financial buyer could afford to pay, and a sum of the parts if the company has very different divisions. In practice we show several side by side and look at where they overlap."

"EV to EBITDA compares companies regardless of how they're financed. Two businesses with the same operations but different debt levels will have very different net income, because one pays much more interest, so their P/E ratios can look very different even though the businesses are alike. EBITDA sits above interest, and enterprise value includes debt, so both sides of the multiple are measured for all investors. It also strips out depreciation, which can vary with accounting choices. P/E makes more sense for banks and insurers, because for them interest and debt are the raw material of the business, so enterprise value and EBITDA don't really mean anything. There I'd use P/E or price to book. P/E is also handy as a quick cross-check for mature companies with steady earnings."

"I start with companies that sell similar things to similar customers, because business model matters more than the industry label. Then I narrow by size, growth, margins and where they operate, aiming for maybe five to ten real peers rather than a long list. For each one I work out equity value on diluted shares and bridge to enterprise value, then pull revenue, EBITDA and earnings for the same periods, usually last twelve months and next year, and strip out one-off items so I'm comparing like with like. I look at the median and the range, not just the average, because one outlier can drag an average a long way. Finally I place the target in that range with a reason: if it grows faster or has better margins than most peers, it deserves the upper half."

"Trading comps value a small, non-controlling stake that changes hands on the market every day. In an acquisition, the buyer takes control of the whole company, so it usually has to pay a premium over the market price to get shareholders to sell. A strategic buyer can also afford to pay more because it expects synergies, like cutting duplicate costs. That's why precedents often come out higher. The weak spots are real, though. Deals can be several years old, done when interest rates, credit markets and sector sentiment were very different. There are often only a handful of truly similar deals, and for private targets the numbers disclosed can be limited. So I treat precedents as a range to sense-check a sale price, not a precise answer."

"First I forecast the company's unlevered free cash flow, usually for five to ten years. That's EBIT times one minus the tax rate, plus depreciation and amortisation, minus capital spending, minus the increase in working capital. Then I estimate a terminal value for all the years after the forecast, either with a perpetuity growth rate or an exit multiple. Next I work out the discount rate, which for unlevered cash flows is the weighted average cost of capital. I discount each year's cash flow and the terminal value back to today and add them up. That gives enterprise value. To get to equity value, I subtract debt and similar claims and add cash. Finally I divide by the diluted share count to get an implied value per share."

"WACC is the cost of equity times the equity share of capital, plus the after-tax cost of debt times the debt share, and preferred stock too if there is any. I use market values for the weights where I can, often a target capital structure from peers. For the cost of equity I use the capital asset pricing model: the risk-free rate plus beta times the equity risk premium. For the cost of debt I look at the yield on the company's bonds or the rate on its loans. That gets multiplied by one minus the tax rate, because interest is deductible, so the government effectively pays part of it. Dividends to shareholders are paid out of after-tax profit, so there's no tax shield on equity and no adjustment."

"With the perpetuity growth method, I take the final year's free cash flow, grow it one more year at a long-term rate, and divide by WACC minus that growth rate. The growth rate should be modest, no higher than the long-run growth of the economy, because nothing grows faster than the economy forever. With the exit multiple method, I apply a multiple like EV to EBITDA from comparable companies to the final year's EBITDA. Either way I discount the terminal value back to today from the end of the forecast. To cross-check, I use one method to imply the other. If my growth rate implies an exit multiple far above today's peers, or my multiple implies a growth rate higher than the economy, one of the assumptions is off. That check matters because terminal value is often most of the total."

"First I set the purchase price, usually a multiple of EBITDA, and lay out sources and uses. Uses are the purchase price plus fees and any debt being refinanced. Sources are the new debt, often in a few layers, plus the equity the sponsor puts in. Then I project the income statement and cash flows for about five years. Free cash flow after interest goes to pay down debt, so the debt balance shrinks each year. At the end I assume an exit, usually at a multiple of the final year's EBITDA, which gives an enterprise value. I subtract the remaining debt and add cash to get the equity value at exit. Comparing that to the sponsor's original equity gives the money multiple and the IRR."

"Doubling in five years works out to roughly 15 percent a year, because 1.15 to the fifth power is almost exactly two. As a reference, tripling in five years is about 25 percent, and doubling in three years is about 26. The returns come from three places. First, growing EBITDA, through revenue growth or cost cuts, raises the exit value. Second, paying down debt with the company's cash flow shifts value from lenders to the equity holders. Third, selling at a higher multiple than you paid, which is called multiple expansion. I'd say the first two are what a sponsor can actually control and underwrite. Multiple expansion depends on the market, so most sponsors assume they exit at the same multiple they paid, and treat anything above that as upside."

"The quick way is to compare what the target earns against what it costs to buy. For an all-stock deal, the cost is the buyer's earnings yield, which is one over its P/E. So if the buyer trades at 20 times earnings, its stock costs it 5 percent. If it's paying 10 times the target's earnings, the target yields 10 percent, which is more than 5, so the deal is accretive. For cash, the cost is the interest the buyer gives up on that cash, after tax. For new debt, it's the interest rate after tax. If the funding mix costs less than the target's yield, earnings per share go up. In a real model I'd also add synergies, and extra depreciation and amortisation from writing up the target's assets, and financing fees, which can change the answer."

"The one I've followed most closely is a large packaged-food company buying a fast-growing snack brand. The logic was growth: the buyer's core brands were barely growing, and the target was growing quickly with a younger customer base. It was paid mostly in cash and new debt, at a high multiple of the target's earnings. So what the target earns is less than the after-tax interest on the money used to buy it, which makes it likely dilutive to earnings per share for the first year or two. The buyer's case rests on pushing the brand through its distribution network, which is a revenue synergy and harder to prove than cost savings. My view is that it makes sense strategically, but the price leaves little room for error. The thing I'd watch is whether growth holds once the brand is in mainstream stores."

"In my internship, our team had four days to turn around a due diligence report for a buyer who had moved their deadline up. There were three of us, and we worked until about two most nights. Early on we split the report by section and agreed on one shared file for every number, so nobody was retyping figures. I took working capital and one-off costs. What kept quality up was a simple rule we set: nobody checked their own numbers, we swapped sections before each draft went to the manager. On the third night I caught a timing error in a colleague's quarterly table, and she caught a sign error in mine. We delivered on time, and the manager's only changes were wording. I still use that swap-and-check habit."

"On a valuation project during my internship, the associate wanted to include a large competitor in the comps set. I thought it didn't fit, because most of its revenue came from a different, lower-margin segment, which pulled the median multiple down. I didn't raise it in front of the whole team. I went to him with a short note showing the segment split and two versions of the table, with and without that company. He agreed it was different, but said the client would expect to see it. So we kept it in, and I added a footnote and a second median excluding it. I was fine with that, because the reader could see both. It taught me that giving a senior person an option works better than just saying they're wrong."

"On a sell-side pitch in my last role, we had three days and two analysts. The first thing I did was write a page listing every slide, who owned it, and when I wanted a first draft. I gave the stronger modeller the valuation pages and the newer analyst the company profile and market pages, with a template to follow. Then I reviewed in stages: I looked at their structure and sources on day one, rather than waiting for a finished draft on day three. For the numbers, I checked every figure on the summary page against the model myself, since that's what the client reads first. I also told the team up front which night would be late, so nobody was surprised. We sent it on time, and the MD's comments were mostly on the story, not the numbers."

"In my last role I sent a manager a comps table on a Friday, and on Saturday morning I noticed one company's EBITDA was for a different fiscal year from the rest, which made its multiple look too low. I messaged him straight away, before he could use it, and said what was wrong, which numbers it affected, and that the median multiple moved up slightly. I sent a corrected version within the hour with the change highlighted. He said it was fine because he hadn't shared it yet, but he appreciated knowing. After that I added a column to every comps table showing each company's fiscal year end and the period I'd used, so a mismatch jumps out before it goes anywhere."

"At my last company, the operations team wanted to buy a second delivery van because the first was always busy. My manager asked me to check the numbers before it went to the owner. I pulled a year of delivery logs and found the van was busy mostly on two afternoons a week, and nearly idle in the mornings. I built a small comparison: the full cost of a second van over three years against renting one on those two afternoons, plus moving some deliveries to mornings. Renting came out far cheaper over the three years, and it kept the option to buy later if volumes grew. The owner went with renting and shifting the schedule. What I took from it is that the useful part was asking when the van was busy, not building a bigger model."

"First I'd spend five minutes working out how long each task really takes and what the minimum good version of each would be. Then I'd raise it straight away, not at midnight. I'd tell both VPs, ideally together or on the same message, that I have both requests, what each involves, and that I can't do both fully by the morning. I'd suggest options: maybe one is due to a client and the other is internal, or one could take a simpler first version. If they can't agree, I'd ask the associate or staffer to decide, since it's not my call to rank two VPs. What I wouldn't do is quietly half-finish both and hope, because then neither VP can plan around it."

"First I'd work out how big it is. If it changes a number on a page the client will see, it matters, whatever the size. I'd fix the model, then trace every page in the book that uses that number and update them, keeping a list of each change. Then I'd send the VP and the associate a short note right away, saying what the error was, what I changed, which pages moved, and that the corrected version is saved with the old one kept alongside. I'd flag it as needing their sign-off first thing, and I'd call or message again early in the morning so it's seen well before nine. I wouldn't leave it for the morning, and I wouldn't quietly swap the numbers without telling anyone."

"I'd start by asking what's behind it, because there's often a fair reason. Maybe the MD knows buyers in this sector are paying more than our precedents show, or thinks our peer set is too conservative. Then I'd look for a defensible way to get there: a different set of comps, a scenario that adds synergies a strategic buyer would see, or a wider range with the high end clearly labelled. I'd show the MD what each change implies, say the exit multiple or growth rate, so it's visible. If the only way to hit the number is an assumption I couldn't explain to the client, I'd say so directly and calmly. The pitch is the bank's name on the line, and a range we can't defend hurts us when the client pushes back."

"I'd tell them I can't talk about any company I might be working with, not even to say yes or no, and change the subject. I'd keep it friendly, because they may not realise what they're asking. But I'd be careful not to hint either way, because even a denial or a pause can tip someone off. If they pushed, or if it sounded like they actually planned to trade, I'd tell our compliance team the next day, because they need to know and they'll tell me what to do next. In this job, keeping deal information to yourself has to be automatic, whether it's a client, a friend or family. Trading on it, or helping someone else do so, is illegal in most places, and it would end the career of everyone involved."

"I've spoken to enough analysts to know the hours are real, so I'm not expecting a normal week. What's worked for me through exams and my internship is protecting a few small things rather than big plans. I try to get some exercise in most mornings, even twenty minutes, because it helps me focus later. I keep my phone and messages organised so I'm not missing requests, which saves a lot of stress. And I keep one fixed call a week with family. The warning sign for me is when I start rereading the same cell three times. When that happens, I take a ten-minute walk, and if it keeps happening I talk to my staffer early rather than letting quality drop."

"Heavy markups are normal, especially early on, so I wouldn't take it personally. The VP has seen hundreds of these books and knows what the MD and the client will look for. First I'd make every change carefully and check that nothing else in the book was affected, like a number that appears on two pages. If any comment isn't clear, I'd ask one short question rather than guess. The part I care most about is not getting the same comment twice. I keep a running list of the things each VP picks up, like how they want footnotes or which way charts should be labelled, and I check against it before I send the next draft. Over time the red pen should shift from formatting to the actual thinking."