Skip to main content
Goldman Sachs Technology Case Study

Project Fernpath — Software Valuation & Growth Regression

A 2-hour Software / Sell-Side Valuation case study with a complete model answer

120
Minute Format
2
Deliverables
6
Concepts Tested
Advanced
Difficulty

Modeled After

Goldman Sachs

Sell-side valuation and process materials for a software company: a growth-versus-valuation regression plotting EV/Revenue and EV/EBITDA against growth, EV/EBITDA and EV/unlevered free cash flow multiples charted over time, a football field rendered as a table of methodology, implied share price and commentary, shareholder cost basis derived from quarterly VWAP, and a party-engagement scorecard tracking sponsors against strategics by stage

Structure and exhibit set are modeled after Goldman Sachs. The company, the financials and every figure in this case are entirely our own.

The Situation

Fernpath Learning Software, Inc. (NASDAQ: FRNP) is a Delaware corporation headquartered in Madison, Wisconsin.

Fernpath Learning Software, Inc.

Sector
Vertical software — K-12 curriculum, assessment and analytics sold to United States public school districts on three-year subscription contracts, alongside implementation services and printed instructional materials
Size
Geography
United States only. Delaware incorporation, Wisconsin headquarters, and district customers across the country buying on a school-year budget cycle
Ownership
Situation

The Prompt

You are the Board's financial advisor. 00 per share is adequate and what it should counter.

120 minutesTechnologyModeling

Supporting Materials

What you are handed at the start of the case, in the format a real process would use.

Requires Diamond Tier
Unlock
  • Blank model template

  • Raw data extract

    ExcelUnlock
  • Board presentation

  • Completed model

What You Have to Produce

The deliverables, in the order the committee will read them. The exercise runs 120 minutes.

Requires Diamond Tier
Unlock
  1. PART 1

    The plan and the cash it actually produces

  2. PART 2

    What the company has actually traded at

  3. PART 3

    The regression, and the residual

  4. PART 4

    The process layer

  5. PART 5

    The ladder, the field and the recommendation

Attempt It First

Blank modelling template

XLSXUnlock

The answer model with every produced cell cleared — the shell you build your attempt in. Work it in Excel against the clock, then check yourself against the model answer below.

How to Approach It

The order a strong candidate works in, and why. This is the shape of the answer — the finished answer deck and Excel model are in the solution set below.

Requires Diamond Tier
Unlock
  1. 01

    Decide what you are NOT building before you build anything

  2. 02

    Charge stock-based compensation, and watch what happens to the rule of 40

  3. 03

    Fit the line, and report the fit

  4. 04

    Read the residual, then decompose it as far as the evidence allows

  5. 05

    Screen on the fit, not on the median

  6. 06

    Separate what the company is worth from what the register will accept

Key Concepts

The ideas this case is built on. Know these cold and the case becomes a question of execution rather than knowledge.

Why software trades on a line rather than on a multiple

For a business with high gross margins and recurring revenue, the dominant question a buyer is answering is how large the revenue base will be in five years, and that is a question about the growth rate. So across any screen of subscription software companies the forward revenue multiple varies systematically with forward growth, and the relationship is measurable by ordinary least squares. That changes what a comparable companies page is for. A median multiple claims the subject deserves what the median peer gets, which is only true if it grows like the median peer. A regression measures the relationship instead of asserting a level, produces a fitted value at the subject's own growth rate, and a residual.

The residual is the analysis

The fitted value tells you what a company with the subject's growth rate typically trades at. Subtracting it from what the subject actually trades at gives the residual, and that number is the finding. Expressed as a multiple of the regression's own standard error it tells a reader whether the gap is noise or signal, and compared with every constituent's residual it tells them whether the subject is an ordinary member of the distribution or an outlier. A residual that is large and negative is a question rather than an opportunity, and the rest of the exhibit set exists to answer it.

R-squared, and why an honest one is usually mediocre

R-squared is the share of variation in the dependent variable explained by the independent one. On a real screen of a dozen software companies, forward growth typically explains a solid majority of the variation in the revenue multiple and leaves a substantial share unexplained, because growth is not the only thing being priced. An R-squared in the high nineties on a hand-built peer set is a warning sign. It usually means constituents that sat off the line were removed without a note. Report the number you actually got, and let the width of your range follow from it rather than from preference.

The standard error of the regression versus the standard error of the fitted value

These are two different statistics and a document that prints both must not confuse them. The standard error of the regression is the typical size of a residual across the sample — the spread of the cloud around the line. The standard error of the fitted value is the uncertainty in where the line itself sits at one particular x, and it is much smaller, because a line estimated from a dozen points is pinned down far better than any single point is. Residuals are measured against the first. Confidence bands around a fitted multiple are struck on the second. Using the wrong denominator turns a two-standard-error residual into an eight-standard-error one and invites a conclusion the data does not support.

The rule of 40, and which margin belongs in it

The rule of 40 adds a growth rate to a profitability margin and asks whether the sum clears forty. The trouble is that the answer depends entirely on which margin you use. Computed on adjusted EBITDA it adds back stock-based compensation, which for most subscription software is a high single-digit percentage of revenue, and it sits above capitalized software development, which is real cash paid to engineers that happens to land on the balance sheet. Computed on post-compensation unlevered free cash flow, both of those are charged. The same company can clear the bar comfortably on one definition and miss it on the other. Pick one, apply it the same way to the subject and to every peer, and disclose which one you picked.

EV / revenue and EV / unlevered free cash flow over time

A single point-in-time multiple tells you where a company trades. A three-year quarterly series tells you what happened to it, and what happened to it relative to its peers. Deriving the cash flow multiple at each date as the revenue multiple divided by the then-consensus cash margin keeps the two series consistent, and it exposes something a revenue multiple alone hides: a company whose margin expanded while its revenue multiple fell has seen its cash flow multiple compress by far more than the headline. Whether that is an opportunity or a warning about the durability of the margin is a judgment, but the arithmetic that raises the question is a five-minute exhibit.

Shareholder cost basis via quarterly VWAP

Where there is no controlling stockholder, a merger requires a majority of the outstanding shares, so what the register will accept is a genuine constraint separate from what the company is worth. Multiplying each quarter's volume weighted average price by that quarter's volume and dividing by cumulative volume gives an average price at which shares changed hands over a window. The trap is that the window decides the answer: on a de-rated stock the most recent quarters carry the lowest prices, so a short window flatters any bid. The stronger evidence is the disclosed top-ten register and the prices at which those positions were built, because the register is who holds rather than who traded. And a cost basis is never a valuation, only an accident of when somebody bought.

A football field rendered as a commentary table

A horizontal bar chart shows where ranges sit relative to each other and has nowhere to put the reason each range is what it is. On a case whose primary methodology is a fitted relationship, the reason is the argument: which specification, which percentile, applied to which year, and what is wrong with it. Rendering the field as a table with methodology, key assumption, low, high and a commentary column lets a reader with ten minutes read the commentary column alone and still come away with the case. Reference methodologies stay marked as reference and shade gray, exactly as they would on a bar chart.

Screening a peer universe when the median will not move

Screening is graded because it is where judgment lives, and the interesting screens are built so that the obvious check does not catch them. A constituent far above the line and one far below it can leave the median almost exactly where it was, and can even narrow the range a regression contributes, because their errors cancel at the point estimate. What they do instead is destroy the fit: the R-squared falls and the standard error of the regression widens. So the diagnostic for a screen on a regression case is the fit statistics, never the median, and never the width of the output range.

What Makes It Hard

The specific traps in this case — the places candidates lose the assessment without noticing.

Requires Diamond Tier
Unlock

Check Your Answer

Type the figures you produced and find out how many are right before you open the worked answer. You get a verdict and, where you are off, a pointer to the part of the build to re-check — never the number itself. Everything you type stays on this device.

How to type a figure. Digits, with an optional unit: 1,234.5, $1,234.5, 2.6x, 21.4%. For a negative use (20.0) or -20.0. Enter as many decimals as you carried — precision is never penalised.

Your Figures

  • $ in millions · graded within ±2%

  • a plain count · graded within ±0.5%

  • $ per share · graded within ±1%

  • $ in millions · graded within ±2%

  • a multiple — type 2.6 for 2.6x · graded within ±0.5%

  • percent — type 20.0 for 20% · graded within ±1%

  • percent — type 20.0 for 20% · graded within ±1%

  • a multiple — type 2.6 for 2.6x · graded within ±0.5%

  • percent — type 20.0 for 20% · graded within ±1%

  • a multiple — type 2.6 for 2.6x · graded within ±0.5%

  • percent — type 20.0 for 20% · graded within ±1%

What the Case Asked For

The arithmetic is only half of it. Tick off what you actually produced — this half is yours to score, because nothing can grade a written recommendation from a checkbox.

The Model Answer

The worked answer in full: answer deck and Excel model, built the way a banker would actually build them. It is a reference, not a submission — a strong answer under the clock is far shorter.

What the Solution Covers

  • Growth versus valuation regression
  • EV/Revenue and EV/unlevered FCF over time
  • Football field rendered as a commentary table
  • Shareholder cost basis via quarterly VWAP
  • Party engagement scorecard by stage
  • Analysis at various prices

Answer Deck

Full model answer, banker-formatted

Upgrade to Diamond

Sign up and upgrade to Diamond to unlock the answer deck, the Excel model and the audio walkthrough.

Get Started

The Excel Model

The model is linked and tied out end to end — every schedule, every formula and every check, in the file itself.

Downloads are available to Diamond members

Excel Model and PowerPoint Deck and Answer Deck (PDF) — yours to open, edit and rebuild

Upgrade

Walkthrough

A conversational walkthrough of how to approach the case under time pressure — where to start, what to cut, and how the recommendation gets defended.

Audio Walkthrough

60s Free Preview
Requires Diamond Tier

How to approach Project Fernpath — Software Valuation & Growth Regression

60-second preview — upgrade to Diamond for the full walkthrough

Unlock

Frequently Asked Questions

Why is there no DCF in a software valuation case?

Because on a five-year horizon most of the value would sit in a terminal multiple, and the terminal multiple would be struck on the same peer set the regression already prices. Choosing it by hand imports the answer into the exhibit that is supposed to check the answer, and dressing it as a discounted cash flow makes the assumption harder to see rather than easier. The clock is also two hours, and a DCF would displace the analysis that actually decides the case. The one-period version of the same idea — apply a fitted multiple to next year's revenue, get a share price a year forward, discount it back at a cost of equity — is cheap, legible and answers the question a Board actually asks, which is what happens if we do not sell.

How do I actually build the regression in Excel?

It is four functions and about six rows. SLOPE and INTERCEPT give you the line; RSQ gives you the fit; STEYX gives you the standard error of the regression. All four take the dependent range first and the independent range second, and the order catches people out. Then take COUNT, AVERAGE of the x range and DEVSQ of the x range, and you can build the fitted value at any growth rate and the standard error of that fitted value from first principles. What takes the time is not the formulas, it is deciding which rows go into the ranges — which is the point of the exercise.

What counts as a legitimate ground for excluding a peer?

A difference in the economics being priced, not a difference in the multiple. A hardware business sold into the same customer budget is not a software comparable however similar the customer is, because its gross margin and its recurring revenue share are different in kind. A consumer business selling to households rather than to institutions has a different retention mechanic and a different budget cycle. A company under an announced all-cash acquisition trades at the deal price, so its multiple is a takeout multiple and putting it in a set of trading multiples buries a control premium inside the fitted line. What is never a ground is that a constituent is inconvenient — a screen that removes only the observations hurting your case is not a screen.

Why report the residual in standard errors rather than just in multiple terms?

Because a reader cannot tell from a raw residual whether it is large. Half a turn of revenue is enormous on a tight relationship and unremarkable on a loose one, and the standard error of the regression is exactly the statistic that tells the difference. Expressing the residual as a multiple of it converts a number into a judgment a reader can check, and comparing it against every constituent's residual tells them whether the subject is an ordinary member of the distribution or the largest deviation in it. Both statements belong on the page, and both are two lines of arithmetic.

Is a low R-squared a reason to abandon the methodology?

No. A relationship that explains a solid majority of the variation is real evidence, and it is the best evidence available for pricing a software company against its peers. What a moderate R-squared changes is the honest width of the range you draw from it and the weight you put on a single fitted point. It is also the reason the second specification exists: adding the cash margin to the axis should raise the fit, and if it does, that improvement is itself a finding about what the market is paying for. Abandoning the regression because it is not tight would leave you with a median, which claims far more and measures far less.

Should the shareholder cost basis set a floor on the price?

No. It is a process input, not a valuation. A holder's entry price is an accident of when it bought, and a Board that priced a transaction to make its register whole would be underwriting somebody else's timing rather than valuing its own business. What the analysis does is tell the Board what it will have to argue at a vote it cannot take for granted, and it is only safe to print beside a valuation range at all when the register's basis sits inside that range. If it sat above everything the evidence supports, the honest answer would be to say so and let the vote fail rather than to raise the ask to meet it.

About This Software / Sell-Side Valuation Case Study

Software / Sell-Side Valuation case study for investment banking interviews. 120-minute format covering growth versus valuation regression, ev/revenue and ev/unlevered fcf over time, football field rendered as a commentary table. Includes the full prompt, a model answer deck, a tied-out Excel model and an audio walkthrough.

This case study sits in Investment Banking, under Technology. Every case ships with the full prompt, the supporting materials, a complete model answer and an audio walkthrough of the judgment behind the recommendation.

120-Minute Format

The time limit a real assessment would give you

Answer Deck

Included in the model answer

Excel Model

Included in the model answer

Audio Walkthrough

How to approach the case under time pressure

200K+ students have used IB Vine to help land offers at top firms.

Moelis & Co.

umich.edu
Diamond Tier

I loved IB Vine. I’m going into Private Capital Advisory (PCA); you have the lessons on there for PCA and that was actually a game changer. I went into those interviews and knew about the core PCA concepts like net asset value, primary fundraising vs. secondary advisory, etc. I swear I was on IB Vine like a few hours a day. The opportunity to have that learning section and go through the questions people submit and everything — there’s no better tool out there. I loved the guides/lessons. The user interface is amazing. The way you’ve simplified it is so spectacular — it’s just so much easier to digest. It’s fun to use too; I’d rather use IB Vine than scroll through a guide. To be honest, I think the product is perfect. I genuinely owe you a big thanks.

Moelis & Co.

smu.edu
Diamond Tier

IB Vine is a tool we really love to use in the club I’m a part of, and there’s really no other resource like it. You guys do a phenomenal job with the question bank. I recruited specifically for Energy banking, and IB Vine was my most used resource for generalist questions (which were about 50% of my interview questions; the rest being Energy-specific); the majority of such questions I saw in interviews were at least similar (if not the same) to the ones on IB Vine.

Perella Weinberg

umich.edu
Diamond Tier

Once I read through the BIWS learning guides, I really didn’t refer to them again. I didn’t even really run through the 400 question guide once I found IB Vine, which I heard about through one of my classmates. We even get a free subscription (like most business schools) to Wall Street Prep, and if I’m being completely honest, I never even logged in to WSP. IB Vine is pretty much the only tool I used (along with our club question bank & mock interviews with peers) and it was invaluable for recruiting.

Cantor Fitzgerald

babson.edu
Diamond Tier

What a platform, made such a huge difference. I did superdays at Evercore, PJT, M. Klein and Barclays among others from a non-target school and did not miss a single technical in any interview process through prepping with IB Vine.

Barclays

ufl.edu
Pro Tier

Very accurate questions and all of the solutions are easy to follow. At least 10 of the questions I studied through this platform appeared in my Round 1 or Superday interviews.

TD Securities

umich.edu
Diamond Tier

THANK YOU SO MUCH IB VINE, I COULD NOT HAVE DONE THIS WITHOUT YOU, SERIOUSLY!!! IB Vine was the best website ever. I spent at least two hours on this daily (seriously) from October through I get my offer in February.

Citadel

uchicago.edu
Pro Tier

I’m doing public equities this summer and next. I know the name is “IB Vine” but at the undergrad level a lot of the technicals across public equities are the same as investment banking. I recommend your software to all my friends!

Piper Sandler

cmc.edu
Pro Tier

I loved this site! 1000% this is the best resource I used in the process.

Lazard

amherst.edu
Diamond Tier

This was the greatest tool ever. I genuinely enjoyed running through the technicals/behaviorals and it was very helpful!

Houlihan Lokey

wustl.edu
Diamond Tier

IB Vine was the most helpful resource I had during recruiting. I will continue to promote it to other students at WashU and elsewhere.

Jefferies

georgetown.edu
Diamond Tier

Awesome product, helped me crush my technicals in my interviews and land a great role. Thank you, seriously was a huge help.

Morgan Stanley

wharton.upenn.edu
Diamond Tier

Very helpful to get real-life questions unlike the 400 guide, especially for merger math. Built deeper understanding of key concepts.

Houlihan Lokey

princeton.edu
Diamond Tier

Extremely helpful study tool that carried me through the recruitment process from start to end.

Evercore

uchicago.edu
Diamond Tier

You guys are doing great work over there with IB Vine. Absolute staple for interview prep.

Rothschild

colorado.edu
Diamond Tier

IB Vine was incredibly helpful and I am forever thankful for all the help.

RBC

oberlin.edu
Diamond Tier

IB Mock was amazing - I used it for multiple hours. Also the flashcards and the lessons features on IB Vine were fantastic. Thank you!

Dragoneer

queensu.ca
Diamond Tier

IB Vine is such a great platform, really impressive. There is so much value in this. The audio podcasts / mock interview library are pretty incredible.

Case Study Preparation

Explore All Case Studies

100+ case studies, each with the full prompt, supporting materials and an audio walkthrough; most also ship a model answer deck and a tied-out Excel model.

Every case has a public page like this one. The member library is the signed-in index members work through.