Deep Dive Brief: How to Build a DCF for Serial Acquirers Like Constellation Software - Part III
A More Fundamental Approach: The ROIC and Reinvestment Rate-Driven DCF
Expanse Stocks x Finchat (now Fiscal AI) Partnership!
🎁 Get 20% off + 2 months free on any plan! →🔗 Claim Discount
Welcome Back Explorer!
In this thirteenth edition of my Deep Dive Briefs series and the third and final installment in the Constellation Software (CSU) trilogy we’ll wrap up our research journey on this serial acquirer with a topic I know many of you will appreciate:
How to build a DCF for serial acquirers like Constellation Software.
If you've ever tried to build a Discounted Cash Flow (DCF) model for a company like Constellation, you’ve likely felt a sense of frustration. This remarkable wealth-creation machine, with its relentless pace of acquiring small, sticky software companies, seems to break the rules of traditional valuation.
Forecast 15-20% revenue growth? Sure, but where does that growth actually come from?
Assume stable margins? Maybe, but how do you account for the massive cash outlays for acquisitions that don't appear on the income statement?
The standard DCF model, the one taught in business schools and used by most analysts, is built on a foundation of revenue growth and operating margins. You project revenues, apply a margin to get your operating profit (EBIT), tax it, and then meticulously adjust for things like Capital Expenditures and Working Capital.
For a company like Coca-Cola, this works beautifully. But for a serial acquirer like Constellation, this method is not just imprecise; it's fundamentally misleading. It misses the very engine of the business. You end up modeling the result (revenue growth) without modeling the cause (the capital spent to acquire it).
Today, I want to introduce a more robust methodology which incorporate the fundamental drivers of value creation: Return on Invested Capital (ROIC) and Reinvestment Rate.
Topics We’ll Cover
🔹 The Flaw in the “Traditional” DCF
🔹 A More Fundamental Approach: The ROIC and Reinvestment Rate-Driven DCF
🔹 The Step-by-Step Mechanics for your DCF
🔹 The Key Assumptions to Get Right
🔹 When to Use This Framework
🔹 The Final Layer: Separating Organic vs. Acquired Growth
🔹 The Models for Constellation & Its Spin-Offs
🔹 The Bottom Line: Why This Method Is Superior
Let’s get to it! It’s free to read.
📰 What’s New at Expanse Stocks?
📊 Quarterly Update - Portfolio Composition (by Industry and Geography) + Valuation Metrics → 🔗 Behind-The-Scenes [Free access]
📚 Articles
🔎 Deep Dive Briefs
⛅ Cloudflare | 👷♂️ Parsons | 𓇲 MPS | 🥼 Medpace #1, Medpace #2 | 🔌 Arista | 🛒☁️ Amazon | 🤖 ASML | 🦎 Topicus | 💡 Lumine | ✈ HEICO | 🧙♂️ CSU Part #1, Part #2, Part #3 | 🤖 Intuitive | 🛒 💳 MELI Part #1, Part #2, Part #3 | 💻 Cadence
💸 General Investing
💎 Hidden Gems Series
✨ Annual Specials – Annual Letters, Investing Philosophy, Industry Write-ups & Top Picks of the Year
📚 Resources for Investors
📢 Latest Stock News → Earnings Digests & other related news
💬 Join My Chat → [🔗 Learn more]
🔐 Paid Subscriber Exclusives
💼 Portfolio Corner – Holdings, valuation models, trades, performance & more!
🤫 Exclusive Sections – From select articles, Deep Dive Briefs & Hidden Gems
👀 Recent Releases!
💸 Software Darwinism and Software Darwinism 2.0 → We talk AI, SaaS, VMS and Where the CSU family stands
💎 Hidden Gems: Sixth Edition → A Primer on TerraVest Industries
👀 Coming Soon
🔎 Deep Dive Brief: Norbit ASA
💎 Hidden Gems: Seventh Edition —> 2 Nordics Niches, early stage.
The Flaw in the "Traditional" DCF
A traditional DCF forces us into a series of clumsy workarounds when dealing with a business like Constellation. The single biggest use of cash for growth is acquisitions, which has no natural home in the standard framework.
Analysts might forecast huge top-line growth but then fail to properly account for the billions of dollars in acquisitions required to buy that growth. This creates a dangerous disconnect, making the company appear more cash-generative than it actually is.
What “Reinvestment” Means for Constellation Software
The key insight is redefining "reinvestment."
For a normal company, Net Reinvestment is about building factories or buying equipment or R&D spend to fuel future growth. The result is a Net Reinvestment that shows as follows in the models:
Net Reinvestment = CapEx – Depreciation+ Change in NWC (+ Capitalized Intangibles if any)
For Constellation Software, the primary form of "growth CapEx" is buying other companies. Therefore, we must include acquisitions in our definition:
Net Reinvestment = (CapEx - Depreciation) + Change in NWC + Cash Spent on Acquisitions
Suddenly, the entire model clicks into place. Constellation's high growth is a direct and logical result of:
A high Reinvestment Rate which results from the net reinvestment of generated profits: CSU consistently deploys a huge portion of their cash flow to buy new vertical market software (VMS) businesses.
A high and stable ROIC: They are brilliant at buying these businesses at prices that generate excellent, durable returns on capital.
A More Fundamental Approach: The ROIC and Reinvestment Rate-Driven DCF
Instead of guessing at revenue, let's ask a more intelligent question: How efficiently does the company generate growth from the capital it invests?
This brings us to the core equation that should drive any good valuation:
Growth (g) = Return on Invested Capital (ROIC) x Reinvestment Rate (RR)
This makes intuitive sense. If a company reinvests 80% of its profits and generates 20% returns on those reinvestments, its intrinsic value should grow 16% annually. Everything else is just accounting noise.
Let’s quickly define these terms:
🔹 Reinvestment Rate (RR) = Net Reinvestment / NOPAT
The math is clean and logical. It answers the question, "What percentage of the company’s after-tax operating profit (NOPAT) is plowed back into the business to generate future growth?" This is the fuel.
🔹 Return on Invested Capital (ROIC) = Growth (g) x NOPAT / Net Reinvestment
Here, it answers the question of “How effectively does the company grows profits from that reinvested capital?” This is the engine's efficiency.
By forecasting these two variables, ROIC and RR, you are directly modeling the company's capital allocation skill. Growth is no longer a magic number you plug into a spreadsheet; it becomes an output of the model, grounded in the economic reality of the business. Instead of just assuming Constellation will grow revenues 15% annually because it has historically, you have to justify that growth by understanding how much capital they're reinvesting and what returns they're generating on it.
The Free Cash Flow to Firm (FCFF) you get in your DCF then becomes elegantly simple: FCFF = NOPAT x (1 - Reinvestment Rate)
The Step-by-Step Mechanics for your DCF
Let me show you how this works with a practical example. Say we're modeling a serial acquirer with these characteristics:
Starting NOPAT at t0: $100 million
ROIC: 20% (consistent with historical performance)
Reinvestment Rate (RR): 80% (including both CapEx and acquisitions)
Here's how the math unfolds:
Growth Rate (g): RR x ROIC = 80% × 20% = 16%
Note: For these 3 years project I am leaving RR and ROIC unchanged, but of course these may be adjusted YoY.
Year 1
NOPAT(t1): NOPAT(t0) x (1 + g) = $100M × (1 + 0.16) = $116M
Net Reinvestment: NOPAT(t1) x RR = $116M × 80% = $92.8M
Free Cash Flow(t1): NOPAT(t1) - Net Reinvestment = $116M - $92.8M = $23.2M
Year 2
NOPAT(t2): NOPAT(t1) x (1 + g) = $116M × 1.16 = $134.6M
Net Reinvestment: $134.6M × 80% = $107.7M
Free Cash Flow(t2): $134.6M - $107.7M = $26.9M
Year 3
NOPAT(t3): NOPAT(t2) x (1 + g) = $134.6M × 1.16 = $156.1M
Net Reinvestment: $156.1M × 80% = $124.8M
Free Cash Flow(t2): $156.1M - $124.8M = $31.2M
You get the pattern. Each year, NOPAT grows by the product of reinvestment rate and ROIC, while free cash flow equals NOPAT minus the dollars reinvested.
The Key Assumptions to Get Right
The success of this approach hinges on getting two critical assumptions right:
Sustainable ROIC: This is where most models break down. You can't just extrapolate historical ROICs forever. For mature serial acquirers, you should model gradual compression as scale makes it harder to deploy capital at historical returns.
For businesses like Constellation this ROIC compression may appear non-linear as history shows. ROIC has shown a step-down trend (47% average between 2015-2019 → 38% average 2019-2023 → ~31% in 2024).
Constellation’s acquisition machine will inevitably “dilute” ROIC as it scales (larger deals, harder to find ultra-high-return niches). So, you’d expect a sharper drop in the first 5-10 years of your DCF, then a much slower decline as it stabilizes closer to the cost of capital + modest premium.
Realistic Reinvestment Rate: This should reflect management's actual capital allocation behavior, not their stated intentions. Some companies increase their reinvestment rate as they grow (more opportunities), while others decrease it (fewer attractive deals).
For Constellation, Reinvestment is bumpy, but structurally high (80–100%), even going above 100% over 2021-2024 as they used debt to make acquisitions, because the strategy is “all-in” on acquisitions.
As long as they can find targets, they may keep reinvesting at 80–100%. At some point, deal size/market maturity forces them to slow down and reinvestment fades more sharply closer to maturity.
When to Use This Framework
This approach works best for companies where:
Acquisitions drive the majority of growth
Management has a consistent M&A playbook
Historical ROIC data is reliable and meaningful
The business model is relatively stable
It's particularly powerful for companies like Constellation, Roper Technologies, or other disciplined capital allocators where the quality of capital deployment matters more than the specific deals themselves.
The Final Layer: Separating Organic vs. Acquired Growth
We can take this one step further to build a truly superior model. The blended growth rate we just calculated is powerful, but it lumps two very different types of growth together:
Organic Growth: The slow, steady growth from Constellation's existing portfolio of hundreds of software businesses (price increases, up-sells). This growth is stable and requires very little new capital, typically around the +2% mark YoY with some bumpiness although it appears they’ve working hard to improve this taking best practices from Topicus. (Last Q was at +4% organic growth).
Inorganic Growth: The lumpier growth that comes from future acquisitions. This growth is the direct result of massive capital deployment.
By modeling these separately, we gain immense clarity. Here’s the logic for a more advanced DCF:
Step 1: Model the Base “Organic” Business. Treat Constellation's existing NOPAT as a mature entity. Assume it will grow at a modest organic rate (e.g., inflation + 1-2%) each year.
Step 2: Model the Acquisition Engine. This is where the ROIC framework shines. We forecast how much capital Constellation will deploy for acquisitions each year and what ROIC they will earn on it.
Step 3: Combine and Project.
g = (ROIC × Reinvestment Rate) + g organic
So, your total NOPAT for next year is the sum of your existing NOPAT plus the organic growth on that base, plus the new NOPAT generated from acquisitions. Your Free Cash Flow is then simply this new NOPAT minus the total reinvestment (which is dominated by the cash spent on acquisitions).
Why This Matters When Modelling Constellation
If you don’t add organic growth, your DCF will undervalue Constellation in maturity, because you assume growth stops once M&A slows.
If you do add it, you can justify a terminal growth rate closer to 3% with only modest reinvestment.
The Models for Constellation & Its Spin-Offs
If you want to dig deeper into the numbers, I’ve built a detailed 40-year DCF model for both Constellation Software and its spin-offs, Topicus and Lumine Group.
Paid subscribers can access and download these models (CSV format) in the Portfolio Corner. Here’s a preview of what’s inside if we take CSU example:
If you’re interested in exploring further, here’s the deal:
Unlock Premium Content — for just $0.39/day ($12/month) or $0.27/day ($100/year)
🔗 Portfolio Corner - 🗓️ Monthly Updates (Last Update: 3-May-2026)
The Bottom Line: Why This Method Is Superior
Traditional DCF modeling for serial acquirers is a nightmare. You're trying to forecast individual deals, integration impact, and the timing of acquisitions that haven't even been identified yet. It's an exercise in futility.
This framework sidesteps all that complexity by focusing on the aggregate outcome. You're essentially saying: "I don't know which companies they'll acquire, but I know they'll reinvest 80% of their cash flows and generate 20% returns on those reinvestments."
Building a DCF this way forces you to think like Constellation's management. It shifts the key valuation questions from the naive, "What will revenue be?" to the critical, "How good are they at allocating capital?"
It forces you to grapple with the real drivers of the business:
How much capital can they continue to deploy effectively? (The Reinvestment question)
Will they be able to maintain their high returns on acquisitions as they get bigger? (The ROIC question)
What is the underlying stability of the existing portfolio? (The Organic Growth question)
This framework allows you to model scenarios far more realistically. You can model a future where Constellation's acquisition pace slows, and its growth gracefully decelerates to its stable, organic rate. This is the key to a credible long-term forecast.
So, the next time you value a serial acquirer, I encourage you to put away the simple revenue-and-margins spreadsheet. Start with the fundamentals of capital allocation, and you'll build a model that not only gives you a better answer but a much deeper understanding of the business itself.
Thanks for following along,
—Nikotes
Growth (g) = Return on Invested Capital (ROIC) x Reinvestment Rate (RR) × Actual Conversion Efficiency