Leveraged Buyout Model From Scratch: Step-by-Step Excel Build
A leveraged buyout model translates a private equity thesis into roughly 800 hard-coded cells, a dozen named formulas, and six linked sheets in Excel. This guide walks through the exact build order a first-year private equity analyst follows when opening a blank workbook on day one, from typing the purchase price into cell B5 of the assumptions tab all the way to the IRR table that gets printed and slid in front of an investment committee. Every formula a beginner would actually type is shown, every common error that quietly destroys returns is flagged, and a worked example follows a hypothetical $25 million EBITDA manufacturing buyout from sources and uses to exit so you can sanity-check your own model side by side. By the end you will have a working file that mirrors the structure used by Wall Street Prep, Financial Edge, A Simple Model, and Macabacus, calibrated to mid-2026 SOFR levels and the leverage multiples Pitchbook LCD is currently reporting on closed deals.
If you already understand the broader mechanics of an LBO and want a tour of every supporting schedule, our companion piece, LBO Model: Step-by-Step Guide, is the right read. For the back-of-envelope mental version asked in private equity interviews, see our Paper LBO Example Walkthrough. This piece focuses on the build sequence inside Excel itself.
What a Leveraged Buyout Model Builds
An LBO model produces three numbers an investment committee actually cares about: the internal rate of return (IRR), the multiple on invested capital (MOIC), and the absolute dollar profit at exit. Everything else in the file, the operating projections, the debt schedule, the working capital roll-forward, the sensitivity grids, exists to support those three outputs and to prove the assumptions behind them are defensible. The model is not an academic exercise. It is the centerpiece artifact that gets attached to an investment memo, shown to lenders during syndication, and dragged into a data room when the deal closes.
The mechanics rest on a single observation that defined the 1980s buyout boom and still drives the math in 2026. If a sponsor buys a company at 10x EBITDA using 60% debt, then pays down that debt over five years using the target’s free cash flow, and exits at the same 10x multiple, the sponsor’s equity stake grows in two ways: through EBITDA growth (operational alpha) and through debt repayment (deleveraging alpha). The Bain Global Private Equity Report 2026 frames this clearly in its “12 is the new 5” finding, noting that during the 2010s a typical deal needed roughly 5% annual EBITDA growth to generate a 2.5x return over five years, while today’s higher entry multiples and tighter financing markets push that requirement to 10-12% annual growth. The LBO model is the spreadsheet that proves you can hit that number.
The recent deal record gives you something concrete to compare against. The September 2025 announcement of the $55 billion Electronic Arts buyout by PIF, Silver Lake, and Affinity Partners disclosed an approximately $36 billion equity contribution and $20 billion of debt financing led by JPMorgan, a 36% debt mix that reflects the size and quality of the target. At the other end of the dial, the March 2025 Sycamore Partners acquisition of Walgreens Boots Alliance ran at roughly 83% debt according to PE Stakeholder Project analysis of the 8-K filings, with HPS Investment Partners and Ares Management among the private credit lenders. Two deals, two extremely different capital structures, and both built off an LBO that worked through the same six sheets in the same order.
The 6 Excel Sheets Every LBO Model Has
Open a blank workbook. Add six tabs in this order: Assumptions, S&U, Operating, Debt, Returns, and Sensitivity. That six-sheet structure is the shared backbone behind nearly every published tutorial, from the Wall Street Prep short-form course to the Financial Edge template to the Macabacus reference workbooks. The names vary, but the function does not. Each tab does one thing well and feeds the next.
Assumptions is the input layer. Every hard-coded number lives here: purchase multiple, transaction fees, debt tranches, interest spreads, revenue growth, margin assumptions, capital expenditures, working capital days, tax rate, exit multiple, and hold period. Nothing else in the workbook should contain a typed number. If a downstream cell needs a value, it links back to Assumptions. This discipline is what makes a leveraged buyout model auditable.
S&U is the sources and uses table. It reconciles where the money to buy the company comes from (sponsor equity, term loan, revolver, seller notes, rollover equity, excess cash) against where it goes (purchase of equity, refinancing of existing debt, transaction fees, financing fees, minimum cash). The two sides must balance to the penny. If they do not, the rest of the model is wrong.
Operating projects the target’s five-year (sometimes seven-year) income statement and the working capital and capex lines needed to compute free cash flow. This is the heart of the operational thesis. Every assumption about price, volume, margin, and reinvestment lands here.
Debt is the schedule. Tranche by tranche, year by year, it tracks beginning balance, mandatory amortization, optional sweep, interest expense, and ending balance. The cash sweep formula on this sheet is the single most error-prone calculation in the entire workbook.
Returns is the equity waterfall. It calculates exit enterprise value, subtracts ending net debt, distributes proceeds across rollover equity and sponsor equity, and computes IRR and MOIC. If your model has a preferred return, management equity incentive plan (MEIP), or tiered carry, this is where the tiers live.
Sensitivity is the optionality layer. Three to five two-variable data tables flex purchase multiple against exit multiple, leverage ratio against IRR, and EBITDA growth against MOIC. Investment committees read this tab first.
Sheet 1: Assumptions Page (Purchase Price, Capital Structure, Exit)
Start with the assumptions tab because every other formula in the LBO model points here. Use three input zones, each clearly labeled. The first zone is transaction assumptions, the second is operating assumptions, and the third is exit assumptions. Color every hard-coded cell blue and every formula black, the convention used across the industry and enforced in nearly every Wall Street training program.
In the transaction zone, type these inputs in column B starting at row 5: LTM EBITDA, entry multiple, transaction fees as a percentage of enterprise value (typical range 2-3%), financing fees as a percentage of debt raised (typical range 2.5%), minimum cash on the balance sheet at close (often $5-20 million for a middle-market deal), and existing debt to refinance. Compute enterprise value in cell B12 as =B5*B6. Compute equity purchase price as enterprise value minus existing debt plus cash assumed acquired. The Pitchbook LCD all-sector LBO average for closed deals over the past two years has been 5.1x debt to EBITDA, with software running about a full turn higher at 5.6x. Use that as a sanity check on your capital structure.
The capital structure zone sits directly below. List each tranche in its own row: Revolver, Term Loan A, Term Loan B, Second Lien, Senior Notes, Subordinated Notes, Seller Note, Rollover Equity, Sponsor Equity. For each tranche record the amount (or as a multiple of EBITDA), the interest rate (SOFR plus spread for floating-rate tranches, fixed coupon for notes), the mandatory amortization percentage, and a Y/N flag for whether the cash sweep applies. SOFR on June 8, 2026 closed at 3.63% per the Federal Reserve H.15 release. Spread it appropriately: senior secured term loans for healthy middle-market borrowers are pricing roughly SOFR + 350 to SOFR + 450 in the broadly syndicated market, with the LSTA publishing weekly secondary trading levels that confirm where new issuance lands. Direct lending unitranche paper tends to run 100-200 basis points wider.
The operating assumptions zone holds revenue growth by year, EBITDA margin by year, depreciation as a percentage of revenue, capex as a percentage of revenue, change in net working capital as a percentage of revenue change, and the effective tax rate. Hold-period default is five years. Some sponsors run six or seven to model a longer hold; the Bain midyear report shows median hold periods stretched to 6.7 years for buyouts exited in 2025, longer than the historical norm of 5-6 years.
The exit zone is the smallest and arguably the most consequential. Exit EBITDA multiple is a hard-coded input. Many beginners try to make it a formula tied to entry multiple; resist that urge. The exit multiple is an assumption you have to defend, not a derivation. Sensitivity tables will flex it later.
Sheet 2: Sources and Uses Table
The sources and uses tab balances the deal. Set up two columns side by side: column B is Sources, column D is Uses. In rows you list each line. On the Sources side: Revolver Draw at Close (often $0), Term Loan A, Term Loan B, Second Lien, Senior Notes, Subordinated Notes, Seller Note, Excess Cash on Balance Sheet (cash above the minimum that the sponsor effectively borrows from), Rollover Equity, and Sponsor Equity (plug). On the Uses side: Purchase of Equity, Refinancing of Existing Debt, Transaction Fees, Financing Fees (capitalized), and Minimum Cash to Balance Sheet.
Every line on the Sources side links to a cell on the Assumptions tab. Every line on the Uses side does the same, with Purchase of Equity computed as enterprise value minus existing debt assumed refinanced. The Sponsor Equity cell at the bottom of the Sources column is the plug. Its formula is =SUM(Uses)-SUM(Sources_above_sponsor_equity). That cell is what tells you how much cash the private equity fund actually has to write. Total Sources and Total Uses both sit in row 20 or wherever your last line is; build a check cell in row 22 that says =B20-D20 and conditional-format it red if non-zero. That single check cell catches roughly half of the mistakes a beginner makes in their first week.
The deal-fee piece deserves special care because it is where errors compound. Transaction fees (advisory, legal, accounting) typically run 2.0-2.5% of enterprise value for a middle-market deal and are expensed at close for tax purposes. Financing fees (commitment, underwriting, original issue discount) run 2.0-3.0% of total debt raised and are capitalized to the balance sheet and amortized over the loan tenor under ASU 2015-03. The accounting matters because amortized financing fees flow through interest expense in your operating model and depress reported net income, which can move covenant calculations.
A small but real point: the source line for Excess Cash on Balance Sheet often confuses first-timers. If the target has $40 million of cash at close and you require $10 million of minimum operating cash to keep the doors open, the sponsor effectively gets to use $30 million of the target’s own cash to fund the purchase price. That $30 million shows up as a Source, and the $10 million minimum cash shows up as a Use. The 10-K of the target tells you how much cash sits on the balance sheet; the operating team usually sizes the minimum.
Sheet 3: Operating Model (5-Year Revenue, EBITDA, Capex, NWC)
The operating tab is where the LBO model meets reality. Set up columns C through H as years 0 (LTM) through 5. Build the income statement top down: Revenue, COGS, Gross Profit, Operating Expenses (split into SG&A and R&D where relevant), EBITDA, Depreciation & Amortization, EBIT, Interest Expense (linked from the debt tab), Pre-Tax Income, Taxes, Net Income.
Revenue in year 1 is computed as =PriorYearRevenue*(1+Assumptions!RevGrowth_Yr1). EBITDA is computed two ways for cross-check. Method one: =Revenue*EBITDA_Margin. Method two: =GrossProfit-OpEx. The two should agree to the dollar; if they do not, your margin assumption is inconsistent with your COGS and OpEx build. Most analysts pick one approach and stick with it. The margin approach is cleaner for a model where you want to flex margin directly in sensitivity.
Depreciation can be modeled three ways in increasing order of complexity. The simplest is depreciation equals a fixed percentage of revenue, often calibrated to the historical relationship. The intermediate version splits maintenance capex (depreciated over useful life) from growth capex (depreciated over a longer schedule, often 10-15 years). The full version builds a property, plant, and equipment roll-forward with separate depreciation schedules per asset class. For an LBO build the intermediate version is usually sufficient.
Capex is modeled as a percentage of revenue, typically split between maintenance capex (sized to roughly match depreciation in steady state) and growth capex (the incremental investment to support revenue growth). The classic mistake here is letting capex drift below depreciation in every projection year, which silently inflates free cash flow. A reasonable check is to confirm that cumulative capex over the hold period covers cumulative depreciation plus enough incremental investment to support the revenue growth you are modeling.
Change in net working capital is modeled three ways too. The cleanest method for an LBO is to express working capital as days of revenue or days of COGS for each component (accounts receivable, inventory, accounts payable), then compute the change year over year. The formula in year 1 is =-(NWC_Yr1 - NWC_Yr0), with the negative sign because an increase in NWC is a cash outflow.
Free cash flow is built at the bottom of the tab: =EBITDA - Cash_Interest - Cash_Taxes - Capex - Change_in_NWC. This is the cash available to repay debt, the single most important line in the entire workbook. Some practitioners build a full three-statement model (income statement, balance sheet, cash flow statement) and pull free cash flow from the CFS; others build a direct FCF schedule. For a buyout model, the direct schedule is faster and just as accurate provided you reconcile working capital correctly.
Sheet 4: Debt Schedule and Cash Flow Sweep
The debt schedule is the single most error-prone tab in any LBO model. Build it tranche by tranche, with each tranche getting its own block of five rows: Beginning Balance, Mandatory Amortization, Optional Sweep, Interest Expense, Ending Balance.
For a Term Loan B tranche, the formulas look like this in year 1. Beginning Balance is the closing balance from the Sources tab. Mandatory Amortization is typically 1% of the original principal per year for TLB, payable quarterly with a 99% bullet at maturity, but for an annual model you can simplify to =-1%*OriginalBalance. Interest Expense is =Average_Balance * (SOFR + Spread), where Average Balance equals =(BeginningBalance + EndingBalance) / 2. This creates a circular reference that Excel resolves through iterative calculation; enable iterative calc in File > Options > Formulas with a maximum of 100 iterations and a max change of 0.001.
The optional cash sweep is where most first-time builds go off the rails. The formula needs to repay debt only if there is cash available after mandatory payments and only up to the remaining balance. The clean structure is:
Cash_Available_for_Sweep = Free_Cash_Flow - Mandatory_Amort - Min_Cash_Build
Optional_Sweep = -MIN(Cash_Available_for_Sweep * Sweep_Percentage, Remaining_Balance_After_Mandatory)
The negative sign keeps the sign convention consistent (sweeps reduce the balance). The MIN function prevents you from paying down more debt than exists. The sweep percentage is typically 50-75% in middle-market credit agreements, stepping down to 0% as the leverage ratio falls below a covenant trigger (often 4.0x).
Revolver mechanics get their own treatment. The revolver is the plug for cash shortfalls. If free cash flow minus mandatory amortization is negative, the revolver gets drawn; if positive and there is a balance, it gets repaid first. The formula for the revolver ending balance is =BeginningBalance + Cash_Shortfall_Or_(-)Surplus, capped at the revolver commitment amount. Revolver interest is calculated on average balance just like term debt, plus an unused commitment fee on the undrawn portion (typically 50 bps).
The order of operations for the cash sweep matters. The standard waterfall is: pay revolver first, then term loans in order of seniority (TLA, then TLB), then second lien, then any junior tranches. Subordinated notes and seller notes generally do not get swept; their balances accrete or stay flat until maturity.
Build a check at the bottom: Total Debt Outstanding at end of each year should match the Balance Sheet check (if you are running a three-statement model) and should decline monotonically if the cash sweep is working. If debt is going up in a year where free cash flow is positive, you have a sign error in your sweep formula. Pitchbook LCD’s Q1 2026 coverage shows the typical sponsor target is to reduce net debt by roughly one turn of EBITDA per year of hold, so calibrate your free cash flow assumption against that benchmark.
Sheet 5: Equity Waterfall and Exit Returns
The Returns tab converts your operating story into the IRR and MOIC that get presented at investment committee. Start with Exit Enterprise Value at the top: =Year5_EBITDA * Exit_Multiple. Subtract Year 5 Ending Net Debt to get Exit Equity Value. Net debt equals total debt minus cash on the balance sheet at the exit date.
Then build the equity waterfall. In a clean middle-market deal with no MEIP and no preferred return, the waterfall is trivial: Sponsor Equity gets its pro-rata share of Exit Equity Value, and Rollover Equity (management or seller) gets its pro-rata share. Most deals have at least a management equity incentive plan, typically 8-12% of the equity pool, vesting over the hold period, with strike pricing tied to the sponsor’s entry equity value. Some have a tiered carry where the sponsor takes a higher percentage above certain IRR or MOIC hurdles, but that is a fund-level construct, not a deal-level one.
Compute MOIC first: =Sponsor_Equity_Proceeds_At_Exit / Sponsor_Equity_Invested_At_Close. The institutional benchmark for a successful LBO is 2.5x to 3.0x MOIC over a five-year hold.
Compute IRR using Excel’s IRR function. Build a small cash flow strip: in year 0, sponsor equity invested (a negative number); years 1 through 4, zero (unless you are modeling a dividend recap, which we will get to); year 5, sponsor equity proceeds at exit (a positive number). Then =IRR(CashFlowStrip). The institutional benchmark for a successful LBO is a 20-25% IRR. For middle-market deals in current conditions, sponsors increasingly target high-teens to low-20s given higher entry multiples and cost of debt.
If the deal includes a dividend recap in year 3 or 4, add that cash flow back as a positive in the relevant year. A dividend recap typically requires the credit agreement to permit restricted payments and is sized to roughly one turn of EBITDA, taking the company back to its closing leverage and returning the cash to sponsor equity.
Build an IRR attribution at the bottom of the tab to show where return comes from. The three drivers are EBITDA growth, multiple expansion (or contraction), and debt paydown. The formula for each is straightforward but the math is satisfying: EBITDA growth contribution equals (Exit EBITDA – Entry EBITDA) times Entry Multiple, divided by Entry Equity. Multiple expansion contribution equals Exit EBITDA times (Exit Multiple – Entry Multiple), divided by Entry Equity. Debt paydown contribution equals (Entry Net Debt – Exit Net Debt), divided by Entry Equity. The three drivers plus the time-value factor should reconcile to total MOIC.
Sheet 6: Sensitivity and IRR Tables
The sensitivity tab houses the data tables that investment committees actually look at. Build three at minimum, each as a two-variable Excel data table (Data > What-If Analysis > Data Table).
Table one: IRR by Entry Multiple and Exit Multiple. Row variable is exit multiple stepped from 8.0x to 12.0x in 0.5x increments. Column variable is entry multiple stepped the same way. The diagonal is the multiple-neutral case. Cells above the diagonal show multiple expansion; cells below show contraction. Most committees draw a thick black box around the cluster of cells that meet the fund’s hurdle (typically 20% IRR) to visualize how much room the deal has.
Table two: IRR by Year 5 Revenue Growth and Year 5 EBITDA Margin. This tests the operational thesis. Row variable is revenue CAGR from 0% to 15% in 2.5% steps. Column variable is exit-year EBITDA margin in 100 bp steps centered on the base case.
Table three: MOIC by Leverage at Entry and Exit Multiple. Row variable is net debt to EBITDA at entry from 4.0x to 7.0x. Column variable is exit multiple. This shows how levered you need to get to clear the fund hurdle at various exit assumptions, useful for the credit committee at the lender side as much as the equity committee at the sponsor side.
One implementation detail. Excel data tables can be slow on a complex workbook because they recalculate the entire model for every cell. If your model has more than about 10,000 formula cells, change Calculation Options to Automatic Except for Data Tables, then press F9 to recalculate when you need fresh sensitivity output. This is the difference between a model that responds in under a second and a model that hangs for thirty.
The Three Key Formula Building Blocks
The entire LBO model rests on three formula patterns that you will type dozens of times across the six sheets. Master these three and you can build any leveraged buyout model from scratch.
Block one: the IF-error-check formula. Used to flag balancing errors, sign errors, and missing inputs. The pattern is =IF(ROUND(CheckCell,0)=0, "OK", "BREAK"). Wrap every major check in this formula and conditional-format the cell red for BREAK. The most important ones: Sources equals Uses, Free Cash Flow Year 0 reconciles to LTM cash generation, Ending Debt Balance on the debt tab matches Total Debt on the balance sheet.
Block two: the MIN/MAX cap-and-floor. Used everywhere cash flows cannot go negative or balances cannot go below zero. Mandatory amortization in the final year of a tranche needs to cap at the remaining balance: =-MIN(Scheduled_Amort, Prior_Year_Ending_Balance). Optional sweep needs to cap at the available cash: =-MIN(Cash_Available, Remaining_Balance). Revolver draw needs to cap at the commitment amount: =MIN(Shortfall, Revolver_Commitment - Drawn_Balance). These three are the spine of the debt tab.
Block three: the average-balance formula. Interest expense on floating-rate debt is calculated on the average of beginning and ending balance. The formula is =(Beginning_Balance + Ending_Balance) / 2 * (SOFR + Spread). Because ending balance depends on interest expense (through the cash sweep) and interest expense depends on ending balance, you get a circular reference. Enable iterative calculation. Some senior modelers prefer to use beginning balance only to avoid the circularity; the trade-off is a small understatement of interest expense during years of heavy paydown.
Worked Example: A $25M EBITDA Manufacturing LBO Built From Scratch
Walk through this with a blank workbook open. The target is a hypothetical specialty industrial manufacturer with $150 million of revenue, $25 million of LTM EBITDA, $5 million of maintenance capex, and a working capital cycle of 60 days receivables, 75 days inventory, and 45 days payables.
On Assumptions, enter LTM EBITDA = $25 million in cell B5, entry multiple = 9.5x in B6 (mid-market industrial is currently transacting in the 8-11x range per Capstone Partners’ Q1 2026 Middle-Market Leveraged Finance Update), enterprise value = $237.5 million in B12, transaction fees at 2.5% = $5.94 million, financing fees at 2.5% of debt = approximately $3.72 million, minimum cash = $5 million.
Capital structure: Term Loan B at 5.0x EBITDA = $125 million, priced at SOFR + 425, with 1% mandatory amortization. Second Lien at 1.0x EBITDA = $25 million, priced at SOFR + 750, with no mandatory amortization. Rollover Equity from management at $7.5 million. Total debt is 6.0x EBITDA, which puts this slightly above the Pitchbook LCD all-sector average of 5.1x but well below the 8x-plus levels typical for software buyouts.
Sources and Uses computes Sponsor Equity as the plug. Sources line up at: $125M TLB + $25M Second Lien + $7.5M Rollover + $5M Excess Cash (assume $10M on balance sheet, $5M minimum) + Sponsor Equity. Uses: $237.5M Purchase + $0 refinance (assume the target has no existing debt) + $5.94M Transaction Fees + $3.72M Financing Fees + $5M Minimum Cash = $252.16M. Sponsor Equity = $252.16M – $162.5M = $89.66M.
Operating projections: revenue grows 6% in year 1, 5% in years 2-3, 4% in years 4-5. EBITDA margin starts at 16.7% and expands 50 bps per year to 18.7% in year 5, reflecting operating leverage and a $3M cost-out program in year 1. Capex runs at 3.3% of revenue. NWC stays flat as a percentage of revenue. Tax rate is 25%.
That gets you to Year 5 EBITDA of approximately $36 million and cumulative free cash flow over the hold period of roughly $90 million, of which $60 million flows through the cash sweep to debt paydown. Ending net debt at the start of year 5 is approximately $90 million, or about 2.5x exit EBITDA, well below entry leverage and comfortably inside any realistic covenant.
Exit at 9.5x (multiple-neutral) on $36 million EBITDA = $342 million enterprise value. Less $90 million ending net debt = $252 million equity value. Sponsor share (after MEIP and rollover) is approximately $230 million on $89.66 million invested, or 2.57x MOIC and a 20.8% IRR over five years. Multiple expansion of half a turn (exit at 10.0x) pushes IRR to roughly 23%; multiple contraction to 9.0x drops it to about 18.5%. That sensitivity is what gets shown to the committee.
Common Leveraged Buyout Model Errors That Crush IRR
Across the dozens of training programs and template libraries reviewed for this guide, the same errors recur in beginner LBO builds. Six of them appear in nearly every first draft.
Error one: sign convention drift. Some cells treat outflows as negative, others as positive, and the model balances by accident. Pick one convention (every cash flow item is a positive number; only check formulas use signs) and enforce it. Color-code if needed.
Error two: capex below depreciation in steady state. Capex equal to about 3% of revenue while depreciation runs 5% silently boosts free cash flow because depreciation is non-cash. This pulls IRR up but is operationally implausible. Confirm that cumulative capex matches cumulative depreciation plus enough to support growth.
Error three: interest expense on closing balance only. Computing interest as =Closing_Balance * Rate understates first-year interest because the average balance was higher. The average-balance formula is the correct approach, even though it requires iterative calculation.
Error four: forgetting the 163(j) interest deduction cap. Under the Internal Revenue Code, business interest expense is limited to 30% of adjusted taxable income (ATI), with the 2025 OBBBA returning ATI to an EBITDA-based calculation per Grant Thornton’s OBBBA analysis. At 6.0x leverage and SOFR + 425 pricing on the TLB tranche, interest expense can exceed 30% of ATI in the early years of the hold and create disallowed interest that must be carried forward. Build a small 163(j) check below the income statement: =MAX(0, Interest_Expense - 30%*ATI). If non-zero, your tax shield is overstated.
Error five: skipping Section 382 NOL haircuts. If the target has net operating loss carryforwards, IRC Section 382 limits the post-change NOL usage to roughly the long-term tax-exempt rate times the equity value at the change date. Most middle-market targets have no NOLs, but if yours does, applying the full pre-change balance against post-change income is wrong. The limitation typically reduces usable NOLs by 60-80%.
Error six: hardcoding the exit multiple to equal the entry multiple. The model balances and the IRR looks fine, but you have eliminated the most important sensitivity in the entire deal. Always input exit multiple as an independent assumption and flex it.
SOFR + Spread Pricing: How to Source 2026 Benchmark Rates
Every floating-rate tranche in your LBO model needs a reference rate plus a spread, and getting both right is what separates a model that lenders take seriously from one they laugh at.
The reference rate is the Secured Overnight Financing Rate, published daily by the New York Fed at approximately 8:00 a.m. ET. SOFR closed at 3.63% on June 8, 2026 per the most recent Federal Reserve H.15 Selected Interest Rates release. Most credit agreements price off 1-month or 3-month Term SOFR, which run slightly above overnight SOFR; you can pull both from the CME Group Term SOFR reference rates page. For a base case, plug overnight SOFR; for a stress case, layer on 50-100 bps to reflect the curve.
The spread depends on the tranche and the borrower. The LSTA, the trade association for the leveraged loan market, publishes secondary market trading levels weekly through its data partnership with Pitchbook LCD. Typical 2026 pricing benchmarks for a healthy middle-market borrower: revolver at SOFR + 275 to 325, term loan A at SOFR + 300 to 350, term loan B at SOFR + 350 to 450, second lien at SOFR + 700 to 850, unitranche direct lending paper at SOFR + 500 to 650. The all-in cost rises if leverage is above 6.0x, the sector is out of favor (energy, retail, traditional media), or the borrower lacks scale.
Build your spreads as a static input on the Assumptions tab, then layer in a stress case that adds 100-150 bps across the structure. The Pitchbook LCD weekly credit wrap reported in the September 2025 Credit Markets Weekly Wrap showed average new-issue TLB spreads moving 25-50 bps wider during periods of market dislocation. Lenders and investment committees both want to see a model that holds up at a stress-priced cost of debt.
Add-On Acquisition Modeling Inside an LBO
Roughly two-thirds of middle-market private equity deals today are platform-and-bolt-on strategies, where the initial buyout is followed by a series of tuck-in acquisitions. The Bain 2026 report notes that multiple expansion through buy-and-build remains the single largest source of return for mid-cap funds. Your LBO model needs to handle add-ons cleanly.
The simplest treatment is a separate add-on schedule that runs parallel to the operating tab. For each add-on, model: closing year, EBITDA acquired, purchase multiple, financing mix (typically additional TLB debt plus revolver draw, sometimes a small equity check), synergies layered into the operating tab as a step function in year 1 post-close, and integration costs (one-time) charged through EBITDA.
The cleanest approach for a beginner is to assume the platform funds add-ons through a delayed-draw term loan (DDTL) facility that the credit agreement already sized at close. A typical structure raises 1.0-1.5x of EBITDA in DDTL capacity, drawable over 24 months at the same spread as the TLB. This avoids the complexity of modeling separate financing rounds for each add-on. Pull the DDTL capacity onto the Sources tab as a separate line with a $0 closing balance, then flow draws into the debt schedule as add-ons close.
The KKR portfolio history offers a clean reference for how this plays out at scale. The firm’s Internet Brands, BMC Software, and Heartland Dental investments all relied heavily on serial add-ons funded through committed debt facilities, with detailed disclosures in subsequent 8-K and 10-K filings whenever the platforms refinanced. Pull a couple of these filings to see how the disclosures phrase the add-on facility mechanics.
Building a Paper LBO Mentally vs in Excel
The paper LBO is the back-of-envelope version of the model that gets asked in private equity interviews. The interviewer gives you a few numbers (purchase price, EBITDA, leverage, exit multiple, growth) and asks you to compute IRR mentally in three to four minutes. Our Paper LBO Example Walkthrough covers the full mental routine, but a quick comparison is useful here.
The paper LBO assumes zero cash sweep (or full cash sweep), no working capital change, and depreciation equals capex. The Excel version models all three explicitly. The paper LBO uses straight-line revenue growth and constant margin; the Excel model lets margin expand or contract by year. The paper LBO assumes exit equals entry leverage net of cumulative free cash flow; the Excel model tracks each tranche separately.
The output gap matters. A paper LBO calibrated correctly usually produces an IRR within 200-300 bps of the full Excel model. If they diverge by more than that, one of them is wrong, and it is usually the Excel model (because of a sign error, a missing tax effect, or an interest expense that is too low). Use the paper LBO as a sanity check on every full model.
For interview prep, build the paper LBO in your head first, then open Excel and replicate it in 15 minutes. The discipline of doing both forces you to internalize which assumptions actually move the IRR and which are decoration. Wharton’s finance program structures its private equity case work this way, and most bulge-bracket private equity recruiting programs do the same.
How CT Acquisitions Uses LBO Models in Sell-Side Mandates
On the sell-side, CT Acquisitions builds an LBO model for nearly every mandate before launch. The purpose is not to predict the price; it is to triangulate the price. By modeling what a financial sponsor could pay at various leverage levels and still clear a 20-22% IRR, we can back into the strike price a serious private equity bidder will offer, then position the management materials and process to extract the maximum from a strategic bidder who has to clear that financial floor.
The mechanics on the sell-side are slightly different. The buy-side analyst building an LBO has access to the data room and can model working capital, capex, and margin to the dollar. The sell-side advisor builds the model from the outside, using publicly available comparables, management projections shared in the confidential information memorandum, and informed estimates for everything else. The output is an LBO floor, a range of bids a financial sponsor could rationally offer, and an indication of where the buyer has room to stretch.
This matters for the seller because the LBO floor is often higher than the seller’s reservation price but lower than a strategic bidder’s synergy-adjusted price. The art of running a sell-side process is using the LBO floor to qualify financial bidders, then using a strategic bidder to push above that floor by 10-20%. Our team builds out the buyer mapping using the same diligence that informs our database of active private equity buyers, and the financial framework that runs through the valuation methodology investment bankers use. The LBO model is also central when the seller is considering a management buyout or a partial liquidity event structured as a recapitalization, because the math behind both structures is the same as a sponsor LBO with the sponsor seat replaced by management.
If you want to see how the enterprise value the LBO model produces ties back to the underlying business mechanics, our piece on the enterprise value equation walks through the bridge from equity value to enterprise value and how transaction-related items get treated. And if you are building LBO models because you are pursuing a private equity career, our private equity analyst career guide covers the recruiting timelines, modeling tests, and case studies that today’s PE associate hiring pipeline runs through.
Leveraged Buyout Model: Frequently Asked Questions
How long should an LBO model take to build from scratch?
A first-time analyst working from a blank workbook typically needs 8 to 12 hours to produce a working LBO model with all six sheets, error checks, and one round of sensitivities. With practice that drops to 3 to 4 hours for a standard middle-market deal, and bulge-bracket private equity associates produce full models in under 2 hours when working from a templated firm shell. The modeling tests used in PE interviews compress this to 60-90 minutes for a stripped-down version that omits sensitivity tables and the full 163(j) check.
What is the typical leverage ratio used in a 2026 LBO model?
Pitchbook LCD data through early 2026 shows the all-sector average debt-to-EBITDA ratio at closing is approximately 5.1x for buyouts financed in the broadly syndicated market, with software running about a full turn higher at 5.6x and traditional industrial sectors generally landing at 4.5x to 5.5x. Megadeals can stretch significantly higher (the Walgreens transaction was financed at 83% debt per the March 2025 8-K disclosures), and smaller transactions financed through direct lending often run 5.5x to 6.5x with covenant flexibility that the syndicated market does not offer.
Should I use Term SOFR or overnight SOFR in my model?
Use 3-month Term SOFR for most credit agreements, which reset quarterly. Overnight SOFR is what you see quoted on the New York Fed website daily; 1-month and 3-month Term SOFR are published by CME Group and typically run 5-15 bps above the overnight rate. For an annual model, plugging Term SOFR is closer to the actual interest the borrower will pay. For a quarterly model, match the actual reset frequency in the credit agreement.
How do you model a dividend recap inside an LBO?
A dividend recap is treated as a new debt issuance whose proceeds are distributed to equity holders. On the debt tab, add a new tranche (usually a TLB add-on or a senior unsecured note) in the recap year with the appropriate amount, spread, and amortization. On the returns tab, add a positive cash flow to sponsor equity in the recap year equal to the sponsor’s pro-rata share of the recap proceeds. Most recaps are sized to roughly one turn of EBITDA, restoring closing leverage. The 163(j) check needs to be rerun because the additional interest can push you over the 30% ATI cap.
What is the right way to model the management equity incentive plan?
The cleanest approach is to set MEIP as a fixed percentage of pro-forma fully diluted equity (typically 8-12%), vesting linearly over the hold period. On the returns tab, compute Pre-MEIP Equity Value at exit, then subtract MEIP proceeds (computed as MEIP percentage times Equity Value above the MEIP strike, where the strike equals the closing equity value). The remainder goes to sponsor and rollover holders pro rata. Some deals layer in a performance kicker that doubles MEIP at certain IRR or MOIC hurdles; model these as IF statements tied to the exit MOIC.
How do you handle existing debt at the target?
List existing debt as a separate line on the Sources side of the Sources and Uses table with a negative sign (because it gets refinanced and removed) and on the Uses side as Refinancing of Existing Debt with a positive sign. The two cancel out at the deal level but the disclosure matters for the sponsor’s understanding of the true capital structure they are stepping into. Some deals assume part of the existing debt rather than refinance it; in that case it stays on the closing balance sheet and the spread reflects the existing terms.
What discount rate do you use to evaluate the LBO?
The fund’s cost of capital, which for most middle-market and upper-middle-market PE funds is approximately 12-15% post-tax. The IRR your LBO model produces is compared against this hurdle. Funds typically require a base-case IRR at least 5-8 points above the hurdle (so a 20-22% target IRR for a fund with a 14% cost of capital) to leave room for downside scenarios.
How does the OBBBA change my LBO interest deduction?
The One Big Beautiful Bill Act, enacted in 2025, restored the EBITDA-based calculation of adjusted taxable income for purposes of the IRC Section 163(j) business interest deduction limit. From 2022 through 2024, ATI was calculated on an EBIT basis, which materially limited interest deductibility for leveraged borrowers. Starting in tax year 2025, depreciation and amortization are added back to taxable income before applying the 30% limit, restoring a meaningful tax shield for buyout sponsors. Models for transactions closing after 2025 should use the EBITDA-based ATI calculation; Grant Thornton’s OBBBA tax alert covers the mechanics in detail.
What is the most common error in a junior analyst’s first LBO model?
The single most common error across hundreds of training-program submissions is the cash sweep formula paying down more debt than exists. The fix is wrapping the sweep formula in a MIN function: =-MIN(Cash_Available_for_Sweep, Remaining_Balance_After_Mandatory). The second most common is interest expense computed on closing balance instead of average balance, which understates first-year interest by 5-10% in years of heavy paydown. The third is treating the exit multiple as a formula tied to the entry multiple rather than as an independent input that needs to be flexed.
How does this compare to a paper LBO interview question?
The paper LBO is a stripped version of the full Excel model that assumes no working capital change, depreciation equals capex, and either a full cash sweep or zero sweep depending on the interviewer’s setup. It produces an IRR within 2-3 points of the full model when calibrated correctly. The full Excel model adds back working capital, splits capex from depreciation, models each debt tranche separately, applies the 163(j) cap, runs the equity waterfall, and produces sensitivity tables. Our paper LBO walkthrough covers the mental routine; this guide covers the Excel build. Both versions sit alongside our companion LBO Model Step-by-Step Guide, which goes deeper on the supporting schedules and analytics that wrap around the build.