How long should I prep for an IB superday?

Plan for six to eight weeks of focused prep if you are starting from zero, four to six weeks if you have already worked through the 200-question deck once. The technical recall layer takes the longest to build; the behavioural layer is fast to polish if you have real stories to tell. Cramming the week before a superday rarely works because the technicals have to be cold-recall, not just recognised.

Are IB and finance still strong career paths in 2026?

Yes, with caveats. Bulge-bracket and elite-boutique analyst seats remain among the most lucrative entry points in any industry, with a clean two-year exit option to private equity, hedge funds, or growth equity. The hours and culture have not improved meaningfully despite the headlines. The class size has stabilised after the 2023 cuts. If you want optionality and you can tolerate the lifestyle, IB is still the best two-year apprenticeship in finance.

What's the difference between bulge bracket, elite boutique, and middle market?

Bulge brackets are the largest full-service banks (Goldman, Morgan Stanley, JPMorgan, BofA, Citi, Barclays). Elite boutiques are advisory-only or near-advisory-only shops with strong M&A franchises (Evercore, Lazard, Centerview, Moelis, PJT, Perella Weinberg). Middle market covers a wide band of regional and industry-focused banks (Houlihan Lokey, Jefferies, William Blair, Piper Sandler). The interview process and culture differ across the three; the technical bar is broadly similar at the top of each tier.

Do I need to know LBO modeling for a first-year analyst interview?

Yes — at least the conceptual layer. Most banks expect a first-year to be able to walk through sources & uses, debt structure, and the four IRR drivers (operational improvement, multiple expansion, deleveraging, and dividend recap). Live Excel modelling is rare in 2026 superdays, but the paper LBO question — where you do the math verbally on a notepad — is still common at PE-prep groups within elite boutiques.

Can I use an AI assistant during a live finance interview?

It is technically possible and the etiquette is contested. The honest answer: an AI assistant is most defensible as a recall fact-check on technical questions, where the bar is to know the formula cold and the assistant is essentially a faster cheat sheet. It is least defensible in behavioural and stock-pitch rounds, where the interviewer is specifically testing for personal conviction. Acedly is built to keep the choice and the disclosure with the candidate.

What's the most common question on a superday?

The three-statement walk-through. Some variant of 'walk me through what happens to the income statement, cash flow statement, and balance sheet if depreciation goes up by ten dollars' or 'walk me through the impact of a $100 inventory write-down across the three statements' is the single most-asked technical question in IB interviews and has been for over a decade. If you can do this cold in 90 seconds, you have cleared the bar; if you cannot, the round is essentially over.

Role Guide17 min read

Finance Interview Prep: Investment Banking, Hedge Funds, and Superdays (2026)

A practical guide to 2026 finance interviews — investment banking superdays, DCF/LBO/M&A technicals, behavioural fit rounds, and hedge fund/buy-side cases — with where a real-time AI assistant helps and where it doesn't.

Devon Park

Head of Research, Acedly

Published July 26, 2025

What a finance interview actually looks like in 2026

Finance interviewing has been more stable than the rest of the recruiting market. The 200-question canon that circulated in the late 2010s — Vault, Wall Street Oasis, Breaking Into Wall Street — is still the spine of investment banking technicals, and the superday format is still the gating round at every bulge bracket and elite boutique. What changed since 2020 is the funnel before the superday: the HireVue-style asynchronous video screen has become near-universal at Goldman Sachs, Morgan Stanley, and JPMorgan, and most middle-market and elite-boutique banks have followed suit.

The result is a loop that looks like this for an investment banking analyst seat:

Networking and coffee chats, weighted heavily at elite boutiques like Evercore, Lazard, Centerview, Moelis, and PJT. Less load-bearing at the bulge brackets — but still expected.
Application through a target-school recruiting portal or a referral.
HireVue or first-round phone screen, typically a recorded video answering 4–6 questions on a 30-second prep, 2-minute response cadence.
Superday, 4–6 back-to-back 30-minute rounds in one day. Traditionally on-site at the office. In 2026, hybrid is common at second-tier banks; the bulge brackets pulled back to mostly on-site for full-time analyst roles.
Offer or ding, with very little middle ground. The superday is binary.

The buy-side path runs in parallel and often through different channels — headhunters at SearchOne, Ratio, CPI, Henkel — and replaces the technical canon with a stock pitch and a markets check.

Bulge bracket vs. elite boutique vs. middle market

The three tiers of the IB recruiting market interview differently and weight things differently. A Goldman Sachs superday is not an Evercore superday and is not a Houlihan Lokey superday, and pretending they are is the most common mistake recruits make.

Bulge brackets (Goldman Sachs, Morgan Stanley, JPMorgan, Bank of America, Citi, Barclays, Deutsche) run the most structured loop. The technicals are predictable, the behavioural questions are predictable, and the superday is genuinely a check that you fit a template — high stamina, high pattern-matching, high willingness to grind. Networking matters but isn't decisive. Volume of applicants means the early rounds are mostly screening you out on basics.

Elite boutiques (Evercore, Lazard, Centerview, Moelis, PJT, Perella Weinberg, Guggenheim) weight networking very heavily. Coffee chats with associates and VPs at elite boutiques are not optional — many shops only run superdays with candidates who came in through a banker referral. The technicals are if anything harder than at the bulge brackets, because the deal teams are smaller and analysts get exposure earlier. The behavioural rounds are more conversational and more pointed about why this specific shop, not why "M&A advisory in general."

Middle market (Houlihan Lokey, Jefferies, William Blair, Piper Sandler, Stifel, Lincoln International) is the most varied. Some shops run loops that feel like bulge brackets; others feel like elite boutiques. Generally the technicals are slightly less demanding and the cultural fit weight is higher. Middle market is also where you'll most often see industry-specific groups (healthcare, restructuring, financial sponsors) hire directly, which means your industry knowledge can offset a less-polished generalist deck.

Technicals: the canonical 200 questions in 2026

The technical canon hasn't been rewritten in a decade, but the frequency distribution has shifted. Here's what's actually asked in 2026, ordered by likelihood per superday:

Three-statement walk-through. Still the single most-asked question in IB, ever. "Walk me through what happens to the income statement, cash flow statement, and balance sheet if depreciation goes up by $10." Variants on inventory write-downs, deferred revenue, capex. If you can't do this cold in 90 seconds, the round is over.
DCF — what's in it and why. UFCF projection, WACC, terminal value (Gordon growth vs. exit multiple), discount to present value, bridge from enterprise value to equity value to share price.
Trading comps and precedent transactions. Why we use them, when one is preferred, what multiples to choose by industry, the control premium gap between trading and precedents.
Accretion/dilution math. Combined EPS, the rule of thumb on cash vs. stock vs. debt-financed deals, and the "is this deal accretive" cheat: if the acquirer's P/E is higher than the target's after-tax interest yield (for cash) or P/E (for stock), it's likely accretive.
LBO basics. Sources & uses, sponsor equity, debt structure (revolver, TLA/TLB, mezz), the IRR drivers, what a paper LBO looks like.
M&A and synergies. Cost vs. revenue synergies, the "phasing" of synergies, why revenue synergies are taken less seriously than cost synergies in deal valuation.

What used to come up but doesn't anymore: live Excel modelling on a laptop ("build me a quick LBO right now") is rare in 2026. Conceptual deep-dives have replaced it almost everywhere except a few PE-prep boutiques.

DCF deep-dive

The DCF is the most heavily-asked valuation question and the one where bad answers compound fastest. The clean version, the way an interviewer wants to hear it:

Step 1. Project unlevered free cash flow for 5–10 years. UFCF = EBIT × (1 - tax rate) + D&A - capex - change in NWC. Use revenue growth and margin assumptions you can defend.

Step 2. Discount each year's UFCF at the weighted average cost of capital. WACC = (E/V × cost of equity) + (D/V × cost of debt × (1 - tax rate)). Cost of equity from CAPM: risk-free rate + beta × equity risk premium.

Step 3. Compute terminal value. Two methods: Gordon growth (TV = final-year UFCF × (1 + g) / (WACC - g)) or exit multiple (TV = final-year EBITDA × exit multiple). Discount the terminal value back to present.

Step 4. Sum the present-valued UFCFs and the present-valued terminal value to get enterprise value. Bridge from enterprise to equity by subtracting net debt and adding cash. Divide by share count to get implied share price.

The follow-ups that catch people:

"What's the discount rate for a private company?" You can't observe beta directly. The clean answer: lever up an industry beta from publicly-traded peers using their D/E, then unlever and re-lever to your target's capital structure. Then apply CAPM. Some shops also add a size premium or illiquidity discount.
"Why does a higher growth rate sometimes lower share price?" Trick question. If g > WACC in the Gordon formula, the terminal value goes negative or explodes, and most interviewers want to see you flag the constraint g < WACC as a sanity check. The deeper version: even within bounds, a higher g requires more reinvestment, which lowers near-term UFCF.

LBO deep-dive

The paper LBO is the technical question that separates analysts who passed the deck from analysts who actually understand it. The structure:

Sources & uses. A sponsor (private equity firm) buys a target for, say, $1B. The sources side: $400M sponsor equity, $400M term loan B, $200M senior unsecured notes. The uses side: $1B purchase price plus $30M of fees and expenses, with $30M of cash taken from the balance sheet to make it balance. (The ratios will differ by deal — these are illustrative.)

Capital structure. Walk the layers from senior to junior: revolving credit facility (often undrawn at close), term loan A (rare in sponsor deals), term loan B (the workhorse, 6–7 year term, covenant-lite typical), senior unsecured notes, mezzanine debt or PIK notes, sponsor equity at the bottom of the stack. Pricing rises as you go down.

The IRR drivers. A sponsor's IRR comes from four things:

Operational improvement. EBITDA growth from cost cuts or revenue expansion during the hold.
Multiple expansion. Selling at a higher EV/EBITDA than you paid. Increasingly hard in a high-rate world; sponsors are no longer underwriting to expansion.
Deleveraging. Free cash flow paying down debt during the hold, which transfers value from creditors to equity at exit.
Dividend recap. Re-leveraging during the hold to pay a dividend to sponsor equity, locking in returns before the exit.

A worked example: buy at 8x $100M EBITDA = $800M EV, $640M debt and $160M sponsor equity. Hold for 5 years. Grow EBITDA to $140M, pay down $200M of debt, exit at 8x = $1,120M EV with $440M of debt remaining. Equity at exit = $680M. Sponsor equity went from $160M to $680M, roughly 33% IRR over five years. A real interviewer will push on every assumption — that's the whole point.

M&A and accretion-dilution

The accretion-dilution test answers the question "does this deal increase or decrease the acquirer's standalone EPS in year one?" The math:

Combined EPS = (Acquirer net income + Target net income + after-tax synergies − after-tax cost of incremental debt) / (Acquirer shares + new shares issued)

Compare combined EPS to acquirer's standalone EPS. Higher = accretive, lower = dilutive.

The cheat that interviewers respect: in an all-stock deal, if acquirer P/E > target P/E, the deal is likely accretive on a basic level (you're buying earnings cheaper than your own market values them). In an all-cash deal, if acquirer's after-tax cost of debt yield < target's earnings yield (i.e., target P/E inverse), it's likely accretive. Mixed-consideration deals require the full math; there's no shortcut.

The follow-ups: synergy phasing (cost synergies are typically realised in years 1–3; revenue synergies are taken at a heavy discount or zero in deal valuation), tax-deal structure (asset deal vs. stock deal vs. 338(h)(10) election in the US), and the goodwill / DTA / DTL bridge on the post-deal balance sheet.

Behavioural and fit rounds

The fit round is where IB recruiting most differs from tech. Tech rewards crisp STAR stories. IB rewards the Why banking? Why this firm? Why this group? triad, told with the right amount of polish and the wrong amount of obvious rehearsal.

The 90-second elevator. "Tell me about yourself." Three beats: where you come from (school, region, family if relevant), why you got interested in finance (a specific moment, not "I've always loved markets"), what you've done that proves it (a club, an internship, a project). Land at "and that's why I'm here today, talking to you about [Firm]." Ninety seconds, max.

Why investment banking? The trap is the cliché answer ("the steep learning curve," "high impact early," "exit options"). The interviewer has heard them eight times today. The strong answer is specific: a deal you followed, a banker you talked to, a problem in capital markets you found interesting. Specificity is what makes this answer survive a follow-up.

Why this firm specifically? You should know two or three named bankers in the group. You should know one or two recent deals the group worked on (DealBook, Mergermarket, Bloomberg are your sources). You should have an answer for why the culture of this shop fits you, even if it's a polite lie. "I want to work here because it's the most prestigious bank that gave me a superday" is the answer interviewers know is in your head; do not let it leave your mouth.

Team-fit on the superday. The associate and VP rounds are explicitly checking whether you'd be tolerable on a 100-hour week. They are not trying to trap you. They are trying to picture you in their group. Be likable. Be calm. Be the kind of person who can be told to redo a deck at 2am without becoming a problem.

Buy-side and hedge fund: the parallel track

The hedge fund and long-only buy-side interview process has its own rules, and the technicals look almost nothing like an IB superday.

The stock pitch is everything. The structure: thesis (what's the trade and why), catalysts (what makes the trade work in the next 6–18 months), valuation (where it should trade and why the market is wrong), risks (what would kill the thesis), position sizing (how much of the book, with what stop). You should be able to deliver this in two minutes and defend each beat under follow-up. Most candidates rehearse the thesis and forget the risks.

"Tell me about a stock you'd short." The classic separator. Long ideas are easy and consensus; short ideas show conviction, structural thinking, and the willingness to be unpopular. A good short has a thesis, a catalyst, and a borrow check. ("It's expensive" is not a short thesis.)

The markets check. Where's the 10-year? Where's the dollar? What's the Fed doing? Who's the chair of the BOJ? Why does this matter for [thesis you just pitched]? The fund is checking whether you read a paper or just memorised a deck.

Quant funds are different again. Two Sigma, Citadel Securities, Jane Street, Hudson River, DRW — these run technical screens that look like software-engineering interviews with a probability slant. Mental math, brain teasers, basic SQL, sometimes Python. The interview part is closer to the tech industry than to traditional finance.

Where a real-time AI assistant helps and where it doesn't

Finance interviewers are unusually good at detecting rehearsed answers. The follow-up question is the diagnostic — a banker who suspects you're parroting a script will push twice on the same point and watch your eyes. This is the context to think about real-time AI honestly.

Where a real-time AI assistant fits across the finance interview surface

Feature	Technical (DCF/LBO/M&A)	Markets / live ticker	Behavioural / fit	Stock pitch (buy-side)	Superday round-robin
AI help quality	High — recall-heavy	Low — needs your own conviction	Low — must be your stories	Medium — fact-check thesis math	High during, low between
Latency requirement	Sub-200 ms	Sub-200 ms	N/A	Sub-200 ms	Sub-200 ms
Stealth requirement	High (most rounds video)	High	Highest	High	Highest
Ethical comfort	Moderate — recall, not judgment	Low — interviewers test conviction	Low — fit must be authentic	Low — pitch must be yours	Moderate
Recommended use mode	Thinking aid	Not advised	Not advised	Fact-check between rounds	Thinking aid only

The honest take: an AI assistant is genuinely useful as a recall fact-check during a technical question. If a banker asks "what's the formula for unlevered free cash flow," you're either supposed to know it cold or you're not getting the offer; the assistant doesn't change that bar. Where it helps is the second-tier follow-up — "what would happen to your DCF if rates went up 100 bps next year" — where having the formula visible while you reason through the answer keeps you honest.

It is not useful as a script for behavioural rounds, and it is actively dangerous in markets and stock-pitch rounds where the interviewer is specifically testing for personal conviction. A candidate reading an AI-generated pitch on Apple loses the round on the first follow-up.

Acedly during a live finance round

For the rounds where a copilot is appropriate, Acedly is built for the latency and stealth requirements of a high-stakes finance interview:

Eight verified meeting platforms. Zoom is the dominant IB platform; Microsoft Teams covers most large buy-side institutions; Webex is still in compliance-heavy banks; Google Meet, Lark, Chime, Coderpad, and HackerRank round out the list. Verification is rerun on a recurring cadence and published live.
~98 ms median end-to-end latency. Measured from the moment the interviewer's question ends to the first token of an answer rendering on screen. Below the threshold where conversational rhythm breaks down.
Multi-model routing. Claude for technical reasoning (DCF math, LBO drivers, accretion logic), GPT-class models for behavioural and conversational fluency, with the routing decided by question type rather than locked to one provider.
30+ spoken languages, single accuracy bar. Hong Kong superdays often run a Mandarin or Cantonese segment; European loops will switch between English, French, German, and Italian in the same superday. The same accuracy bar applies across all supported languages.

A six-week IB / finance prep plan

If you are eight weeks out from the start of recruiting season, this is the plan that has worked for analysts at every tier:

Week 1 — The 200 questions. Pick one of Vault, Wall Street Oasis, or Breaking Into Wall Street and grind the deck cover to cover. Your goal is to be able to answer any of the 200 cold, in 60 seconds, without notes.
Week 2 — DCF, LBO, and M&A practice models. Build a paper DCF and a paper LBO from scratch on a blank Excel sheet, twice. Walk through the accretion-dilution math on three real announced deals. The goal is muscle memory.
Week 3 — Behavioural story-banking and firm research. Write out 8–10 STAR stories that cover leadership, conflict, failure, ambiguity, and impact. Build a one-page brief on each of the 6–10 firms you're targeting: recent deals, group leadership, league tables, recent moves.
Week 4 — Mock superdays with peers. Run two full superdays, 4 rounds each, with classmates. Rotate behavioural, technical, and fit. Record yourself; watch the playback.
Week 5 — HireVue and async practice. Use a HireVue prep platform or just a webcam and a question deck to drill the recorded-video format. Most candidates underestimate how unsettling the camera-only format feels.
Week 6 — Firm-specific drills. Pick the three firms you most want and drill their published or known questions. Most major banks have a recognisable house style of question; you can prep to it.

The candidates who get offers are not the ones who studied the most. They are the ones who studied the right things, in the right order, and stayed calm in the room.