Investment Analyst Interview Questions

A DCF values a business by discounting its future free cash flows back to today. The key inputs are the revenue and margin forecasts that drive unlevered free cash flow, the weighted average cost of capital used as the discount rate, and the terminal value, which typically accounts for 60 to 80 percent of total value in a standard model. Analysts go wrong in three common places. First, terminal growth rates: assuming 3 to 4 percent perpetuity growth for a mature business overstates value materially. Second, WACC: using a market-cap-weighted cost of equity without adjusting for the company's specific risk profile. Third, circularity: forgetting to iterate the debt schedule when modelling a leveraged business. I always sanity-check a DCF output against comparables. If the implied EV/EBITDA is 20x for a business where peers trade at 8x, the model is almost certainly wrong somewhere.

I start by defining the peer set carefully, which is where most analysts cut corners. The right peers share similar business model, end markets, growth profile, and capital intensity, not just the same broad sector. I typically screen by SIC code or GICS sub-industry, then filter manually based on the above criteria. I collect trading multiples: EV/EBITDA, EV/Revenue for high-growth names, and P/E where earnings are meaningful. I normalise for one-off items and differences in depreciation policy before comparing. I present multiples on a current year and forward year basis. The output is a valuation range, not a single number. I then stress-test where my subject company sits versus peers on the key value drivers: if it trades at a premium on EV/EBITDA but lags on EBITDA margin, that premium needs to be justified by growth or quality of earnings.

A strong thesis has three parts. First, a clear variant view: you believe something the market is currently mispricing, and you can articulate specifically why the market is wrong. This might be a misunderstood cost structure, a product cycle the sell-side has underestimated, or a balance sheet catalyst no one is pricing in. Second, a path to value realisation: knowing why a stock is cheap is not enough if there is no reason for the gap to close. The thesis needs a catalyst or a timeline. Third, a defined downside: what is the bear case, what would make you wrong, and is the risk-reward skewed in your favour given current pricing. I also test every thesis against a simple question: if I am right, what does the P&L look like in two years, and is that upside compelling enough to justify the capital and the opportunity cost?

Pick a sector you know deeply. Structure your answer in three parts: first, the macro and industry dynamics (what is driving growth, what are the structural tailwinds or headwinds); second, the competitive landscape and which players are best positioned to capture value; third, a specific view on a name, including the variant thesis and a rough sense of valuation. For example, if you cover European industrials, you might highlight how the reshoring trend is creating multi-year order visibility for certain equipment makers, and then defend why one player has a durable margin advantage the market is not reflecting at current multiples. The key is to have a specific view with numbers, not a generic sector overview. Interviewers are testing whether you form opinions or just describe facts.

Early in my career I modelled a consumer discretionary business and underestimated the impact of rising input costs on margins. I had used three-year historical averages for COGS without stress-testing for commodity price movements. My margin forecast was 200bps too high, which inflated the DCF by roughly 15 percent. When the business reported, gross margins came in well below my estimate. I went back through the model, rebuilt the cost of goods sold line using spot commodity prices and management guidance on hedging ratios, and updated my target price down. I also changed my process: I now build a base, bull, and bear scenario for every margin-sensitive name, with explicit assumptions on the key cost drivers. The lesson was that averaging historical data works in stable environments, but it masks structural breaks when input costs are moving.

I had to present a bond duration analysis to a client relations team who had no fixed income background. Rather than walking through modified duration formulae, I reframed the concept: for every one percent rise in interest rates, a ten-year bond loses roughly ten percent of its market value, while a two-year bond loses only two percent. I used a simple chart showing how price sensitivity scales with maturity, and I connected it to their portfolio: we held an average duration of 7.5 years, so a 50bps rate move would cost us approximately 3.75 percent in mark-to-market value before income. I avoided jargon entirely and focused on the decision they needed to make: should we shorten duration given the rate outlook? The team followed the logic and approved the rebalancing recommendation. The key was anchoring on the business decision, not the maths.

During an earnings season at my previous firm, we had three portfolio companies reporting within 48 hours of each other. I prioritised by materiality: the largest position reported first and had the most variance risk, so I built a pre-earnings model update and a post-results commentary template in advance. For the two smaller names I prepared a one-page framework of the five metrics that mattered most for each, so I could update quickly once the numbers dropped. When the results came in, I had initial commentary ready within 30 minutes of each release. The portfolio manager told me the structured pre-work was what made the turnaround possible. I learned to treat earnings season as a project with parallel workstreams, not a sequence of individual tasks.

Standard EBITDA multiples do not work for loss-making businesses, so you switch to methodologies that reflect where value actually sits. For high-growth SaaS or tech businesses, EV/Revenue is common, sometimes adjusted for a rule-of-40 score. For pre-revenue biotech or early-stage businesses, you use a risk-adjusted NPV of the pipeline or platform. For companies with near-term path to profitability, you can do a forward DCF anchored on year 3 or 4 normalised EBITDA and discount back. I always ask two questions before choosing a methodology: what is the primary value driver for this business, and at what point in the P&L does it show up most clearly? For a marketplace with strong network effects, GMV or take-rate-adjusted revenue might be more informative than gross profit. The goal is to use the metric that best captures what the market will eventually price this business on.

Enterprise value is the total capital value of the business, including both equity and debt, less cash. Equity value is what remains for shareholders after paying off debt. The distinction matters whenever you are comparing companies with different capital structures. A company with 500 million of net debt and an equity value of 1 billion has an enterprise value of 1.5 billion. If you compare its P/E ratio to a debt-free peer, you are not comparing like for like, because the levered company's earnings are suppressed by interest expense. EV-based multiples like EV/EBITDA remove this distortion. Where equity value matters most is in M&A: the acquirer pays the equity value directly, but also takes on the debt, so the total cost is the enterprise value. In LBO modelling you always work at the enterprise value level to size the debt package and calculate returns on equity.

Working capital is often where the real quality of a business shows up. A company that can collect receivables in 30 days but pays suppliers in 60 is effectively getting an interest-free loan from its supply chain. I model working capital as days of receivables, inventory, and payables relative to revenue or COGS, rather than as fixed percentages, because it gives you sensitivity to growth. If revenue grows 20 percent and DSO stays flat, receivables grow in line with revenue. If DSO lengthens, receivables grow faster and free cash flow lags earnings. I always check working capital movement separately from the income statement in a cash flow model, and I flag any business where working capital is a meaningful drag: it often signals either rapid growth or deteriorating supplier terms. For capital-intensive businesses in industrials or retail, working capital can easily account for a 30 to 50 percent swing in free cash flow relative to EBITDA.