PE Modeling Test: How to Pass the 90-Minute and 3-Hour LBO
The PE modeling test is the gatekeeper between your first-round interview and a final-round invite. Unlike the paper LBO, which tests speed and conceptual fluency, the modeling test measures whether you can build a clean, functional LBO in Excel under time pressure. Get it wrong and you are out, regardless of how strong the rest of your interview was.
This guide covers what to expect, how to structure your time, the exact build order that protects you when the clock is running out, and the disqualifiers that quietly kill candidates without them knowing.
What the PE Modeling Test Actually Tests
There are two common formats. Knowing which one you are walking into changes how you prepare.
The 90-Minute (or 2-Hour) In-Person Test
You receive a one to three page prompt and a blank Excel file. Sometimes you get a CIM excerpt with operational data. The expectation is a working LBO model with sources and uses, a 5-year projection, debt schedule, and returns analysis. You will not have time to make it pretty. Speed and accuracy of the core mechanics are everything.
The Take-Home Modeling Test (3 Hours to 1 Week)
Shows up more often in growth equity, smaller funds, and remote processes. You receive a CIM and are asked to produce a full model plus a short investment memo. Time ranges from 3 hours (most common) to a full week. The bar is much higher: formatting, sensitivities, an operating model with revenue build, and a clear thesis are all expected.
Time Allocation: The 90-Minute Test
The single biggest mistake candidates make is spending 40 minutes on the operating model and running out of time before the returns calculation. Returns are the headline output — if they're missing or obviously wrong, you fail regardless of how clean the rest of the model is.
Here is the time budget that protects you:
| Section | Time | Why |
|---|---|---|
| Read prompt & sketch | 5 min | Identify what you actually need vs. fluff in the prompt |
| Assumptions tab | 10 min | All hardcoded inputs in one place — saves time later |
| Sources & Uses | 5 min | Foundation for the whole model |
| Operating model (P&L) | 15 min | Revenue, EBITDA, D&A, taxes |
| Cash flow & debt schedule | 20 min | The most error-prone section — allocate time |
| Returns (MOIC, IRR) | 10 min | Non-negotiable. Get here even if other sections are rough |
| Sanity check & cleanup | 10 min | Catch sign errors, broken links, formula typos |
| Buffer | 15 min | You will use it. Trust the buffer |
The Build Order That Saves You Under Time Pressure
Build in the order that gets you to a complete model fastest. You can always refine sections that are rough — you cannot get credit for sections that do not exist.
Every input the prompt gives you goes in one tab, color-coded blue (industry standard for hardcoded inputs). This includes purchase multiple, leverage, interest rates, growth rates, margin assumptions, capex, working capital, and exit assumptions. Doing this first means you reference the same cell everywhere — if a number changes, you change it once.
Calculate Enterprise Value, then layer in transaction fees and refinanced debt to get total Uses. On the Sources side, build your debt stack (revolver, term loans, mezzanine if applicable) based on leverage assumptions, then plug equity to balance. This is your foundation. Do not skip checking that Sources = Uses to the dollar.
Revenue, COGS or gross margin, OpEx, EBITDA, D&A, EBIT. Stop here. Do not add interest yet — you do not have the debt schedule. Use a placeholder for interest expense (zero or a hardcoded estimate) and come back to it.
Net income (with placeholder interest) + D&A - capex +/- change in working capital = FCF. This feeds the debt schedule.
For each tranche: opening balance, mandatory amortization, optional paydown (typically 100% cash sweep on term loan), interest expense, ending balance. Now circle back and link your real interest expense into the P&L. This creates a circular reference if you have a cash sweep — that is normal. Enable iterative calculations (File > Options > Formulas > Enable Iterative Calculation, max 100 iterations).
Exit EV = exit year EBITDA × exit multiple. Subtract net debt at exit to get exit equity. MOIC = exit equity / sponsor equity invested. IRR = use the IRR or XIRR formula on the cash flow series (typically -equity at year 0, then 0s, then exit equity at exit year). For management equity, layer in promote economics if specified.
If time permits, build a 2-axis sensitivity (most commonly entry multiple × exit multiple, or revenue growth × EBITDA margin) on IRR and MOIC. Use Excel's Data Table function (Data > What-If Analysis > Data Table). This is a major credibility signal.
The Take-Home Test: Different Game
For a 3-hour or longer take-home, the firm is testing judgment more than mechanics. They want to see:
- A clean operating model with a real revenue build. Not just "10% growth applied to topline." Drivers like price × volume, retention × new logos, units × ASP. Show you actually thought about the business.
- A defensible base case with downside and upside scenarios. Toggle them with a scenario switch (CHOOSE or INDEX function).
- Sensitivity tables. At minimum entry/exit multiple. Bonus points for revenue growth and margin.
- An investment memo or thesis page. 1-2 pages. Why the deal works, key risks, return drivers (multiple expansion vs. EBITDA growth vs. debt paydown), what diligence questions you would prioritize.
- Formatting that does not embarrass you. Color coding (blue for inputs, black for formulas, green for cross-sheet links), no hardcoded numbers in formulas, freeze panes on time-series sheets.
The Quiet Disqualifiers
If your Year 3 revenue formula reads =B5*1.10 instead of =B5*(1+Assumptions!$B$12), you fail the formatting test even if the math is right. Reviewers can spot this in five seconds.
Off by one dollar from a fee or rounding. Always include a "Check" cell that displays Sources - Uses and turns red if it isn't zero.
If you build a full BS and it doesn't balance, this is an immediate flag. Always add a balance check row.
Interest expense should reduce net income. If you have it adding to net income because of a sign error, your returns will be wildly wrong and the reviewer will know instantly.
Exit equity = Exit EV - net debt at exit, not Exit EV alone. This is one of the most common errors and it inflates returns by a huge margin.
The prompt usually includes them. Skipping them changes equity check by 1-3%, which throws off returns.
Even a basic 2x3 sensitivity on IRR shows you think about scenarios. A model with no sensitivities looks junior.
Excel Skills That Actually Matter
You do not need to be a keyboard wizard, but the following will pay back hours of practice:
- Anchoring (F4) — second nature. Mixed references (locking row vs. column) for sensitivity tables.
- Paste Special (Alt+E+S+V on Windows, Cmd+Ctrl+V on Mac) — values, formats, formulas only.
- INDEX/MATCH and XLOOKUP — cleaner than VLOOKUP for scenario switches.
- SUMIF and SUMIFS — for segment-level revenue builds.
- IRR, XIRR, NPV — know the syntax cold.
- Data Tables — for sensitivities. Practice this until it is muscle memory.
- Iterative calculations — how to enable, what circular references are normal in an LBO (cash sweep + interest).
- Group/ungroup rows (Shift+Alt+Right Arrow) — collapse detail in your operating model for cleaner output.
How to Practice (Realistically)
Most candidates undervalue practice reps. The single best thing you can do is build full LBOs end-to-end, timed, from blank Excel files. Not watching videos. Not reading guides. Building.
A reasonable practice schedule for the four weeks before recruiting:
- Week 1: Build 2-3 LBOs with no time limit, working from a guided template. Get the structure into muscle memory.
- Week 2: Build 3-4 LBOs from blank Excel files, untimed. Focus on speed of layout.
- Week 3: Build 3 timed 90-minute LBOs. Identify where you slow down.
- Week 4: Build 1-2 take-home style models with full memos. Get someone to review them.
If you have access to a senior associate or mentor in PE, ask them to review one model before recruiting. A 30-minute review will catch formatting habits that would otherwise cost you offers.
What Reviewers Actually Look At First
When a PE associate reviews your model, the order of operations is almost always the same:
- Returns page. What MOIC and IRR did you get? Are they in a reasonable range (typically 20-25% gross IRR for a buyout modeling test, 25%+ for growth)?
- Sources & Uses. Does it tie? Is the leverage realistic for the business?
- Debt schedule. Is interest reducing net income correctly? Is paydown reasonable?
- Operating model. Are growth and margin assumptions defensible vs. the prompt?
- Formatting. Color coding, no hardcoded values in formulas, easy to follow.
Notice they look at returns first. If your returns are obviously wrong (negative IRR, 100x MOIC), they often stop reading. Build to make sure returns work, even if other sections are rough.
Related Guides
The modeling test rewards preparation more than talent. Candidates who have built 10+ LBOs walk in with the structure as muscle memory and spend their cognitive energy on the prompt-specific details. Candidates who have built 1-2 spend half their time figuring out where to put the debt schedule. Build the reps.
Practice LBOs and Sharpen Your Excel Workflow
PEPath includes 10 paper LBO practice prompts with full step-by-step solutions, plus 185+ interview questions covering modeling test concepts, accounting, and deal mechanics.
Download PEPath Free