Handling What-If Follow-Ups and Stress-Testing Questions

Introduction

The initial "walk me through a DCF" or "walk me through an LBO" question is the invitation. The follow-ups are the actual test.

Every interviewer knows that a scripted walkthrough can be memorized. What they cannot fake their way through are the what-if questions: "What happens if WACC increases by 100 basis points?" or "What if the exit multiple contracts by 2 turns?" These questions reveal whether a candidate genuinely understands the mechanics or has rehearsed a surface answer.

The good news: what-if questions follow a predictable structure. Once you understand the underlying logic, you can answer almost any variation without having memorized the specific question.

Every what-if question in valuation can be answered in four steps:

1. State the direction of the effect. "A higher WACC reduces the DCF value." One clear sentence. Do not hedge or equivocate at this step.

2. Explain the mechanism. "WACC is the discount rate. A higher discount rate reduces the present value of every future cash flow, because each dollar of future cash flow is worth less in today's terms."

3. Quantify if possible. "A 100 basis point increase in WACC might reduce the implied enterprise value by 10-15%, though the exact magnitude depends on the projection period and the weight of terminal value." If you cannot quantify precisely, give a directional range.

4. Flag secondary consequences. "Higher WACC also makes the exit multiple more critical, because a smaller portion of value comes from the near-term cash flows that are less sensitive to the discount rate."

This four-step format works for every what-if question, whether it is about WACC, growth rates, margins, multiples, leverage, or interest rates. Practice the format until it is instinctive.

Directional Analysis (What-If Context)

The ability to immediately identify whether a change in an input variable increases or decreases the valuation output, without needing to calculate the exact magnitude. In interviews, directional analysis is the most tested skill because it reveals whether the candidate understands the causal chain between inputs and outputs. For example: "higher WACC → lower present value → lower enterprise value" is a three-link causal chain that must be stated immediately. Candidates who can deliver the direction in 2 seconds and then add mechanism and magnitude demonstrate mastery; candidates who hesitate on the direction reveal a lack of genuine understanding.

These are the most common what-if questions after a DCF walkthrough, with model answers:

"What happens if the risk-free rate increases by 100 basis points?"

The risk-free rate flows directly into CAPM, raising the cost of equity, which raises WACC. A higher WACC reduces the present value of all projected cash flows and the terminal value. The effect is amplified by terminal value dominance: if 70% of value is in the terminal value, a higher discount rate has an outsized negative impact. The sensitivity table in most DCF models shows roughly a 10-15% reduction in enterprise value per 100 basis point increase in WACC, though this varies with the growth rate.

"What if you assume a 1% higher terminal growth rate?"

Using the Gordon Growth Model, terminal value = FCF x (1 + g) / (WACC minus g). A 1% increase in g both increases the numerator and shrinks the denominator, creating a compounding positive effect. This is why terminal value is so sensitive to the growth rate assumption: the function is convex, not linear. In practice, a 1% increase in terminal growth rate might increase enterprise value by 15-25%, depending on the gap between WACC and g. This is also why assuming a growth rate close to WACC is dangerous: the denominator approaches zero and terminal value approaches infinity.

"What if operating margins compress by 200 basis points?"

This reduces EBITDA and UFCF in every projection year, lowering both the explicit period cash flows and the final-year EBITDA used in the exit multiple method for terminal value. The effect is proportional: if margins compress by 200 basis points on a 20% EBITDA margin business, EBITDA declines roughly 10%. All else equal, the DCF value falls by roughly the same proportion.

"What happens if leverage increases by 1 turn?"

Additional leverage reduces the equity check for the same purchase price, increasing MOIC and IRR if everything else holds. However, higher leverage increases interest expense, which reduces FCF available for debt repayment and amplifies downside if EBITDA disappoints. The interviewer is testing whether you understand the leverage trade-off: leverage amplifies returns symmetrically, meaning it also amplifies losses.

"What if the exit multiple compresses by 2 turns?"

Exit multiple compression directly reduces the exit enterprise value and therefore exit equity. In a 5x entry / $100M EBITDA deal with 5x leverage, compressing the exit from 8x to 6x reduces exit EV by $200M (assuming constant EBITDA). That $200M flows directly to equity holders, reducing the MOIC and IRR materially. This is why PE firms target multiple expansion or at minimum multiple preservation: multiple compression is one of the hardest risks to underwrite. Refer to the three value creation levers framework in LBO value creation.

"What if the company's EBITDA is flat rather than growing?"

Flat EBITDA means all returns must come from debt paydown and (if any) multiple expansion. At a 5x entry EBITDA multiple with $100M EBITDA flat for five years, exit EV equals $500M x exit multiple. If the exit multiple also stays flat, exit equity is exit EV minus remaining debt. The return is entirely a function of how much debt was repaid. This scenario tests whether the company has sufficient FCF conversion to service and repay debt even without EBITDA growth.

After the sensitivity questions, interviewers often push to a full downside scenario: "Talk me through what happens if the company misses its EBITDA by 20% and the exit multiple compresses."

Stress Test vs. Sensitivity Analysis (Interview Context)

In interviews, a sensitivity question changes one variable while holding others constant ("what if WACC increases by 100 bps?"). A stress test question changes multiple variables simultaneously to simulate a realistic adverse scenario ("what if EBITDA misses by 20% AND the exit multiple compresses by 2 turns AND rates rise?"). Sensitivities test mechanical understanding of individual variables. Stress tests evaluate whether the candidate can think through how adverse conditions compound, which is what actually happens in a downturn (EBITDA declines, credit tightens, multiples compress, and interest costs rise simultaneously).

This is a stress test, not a sensitivity. The answer requires combining multiple negative assumptions simultaneously:

Walk through this systematically: "If a company purchased at 8x with $400M EBITDA misses by 20%, exit EBITDA is $320M. At an 8x exit multiple, exit EV is $2.56 billion. If entry was at $3.2 billion with $2 billion in debt and $200M was repaid over five years, remaining debt is $1.8 billion. Exit equity would be roughly $760M, compared to an equity investment of $1.2 billion: a loss. This is what covenant breaches and forced recapitalizations look like."

Sometimes an interviewer asks a follow-up you have not specifically prepared. The correct response is not silence or "I don't know." It is structured reasoning:

"Let me think through the mechanics here. [Restate the variable changing.] In a DCF / LBO, this variable affects [X part of the model] through [Y mechanism]. If [it increases / decreases], the effect on value is [direction] because [reason]. Secondarily, [flag any non-obvious consequence]."

This approach demonstrates that you are not retrieving a memorized answer but actually reasoning through the model. That is a stronger signal to an experienced interviewer than a polished scripted response. The ability to decompose an unfamiliar question into its logical components is exactly the skill that separates first-year analyst candidates from candidates who have genuinely internalized the frameworks.