Last Updated on September 9, 2024 by
Mark Leonard is one of my all-time investing heroes.
His conglomerate Constellation Software (CSU) compounded C$25m in equity into a~C$89bn conglomerate in 29 years.
That’s a 33% CAGR 👀
Today, they do over $6 billion in revenue – an absolute juggernaut of a company.
It’s a cash flow generating machine – last year alone CSU generated $1.9Bn in free cash flow – a free cash flow margin of 21%!
He did it by rolling up tiny vertical market software (VMS) businesses.
VMS businesses are absolutely beautiful – asset light, high margin businesses with switching costs, customer lock in and recurring revenue.
Constellation is a big ‘kimono doll’. Businesses, inside businesses, inside businesses.
They have six operating groups, but within each of these groups are smaller companies.
How did he do it?
Well, when Mark started, ‘SaaS’ wasn’t even a thing.
Mark actually spent over a decade working as a VC before he realised how incredible VMS businesses were.
“The most important revelation of my venture career was that vertical market software businesses have great economics but had been poor venture capital investments because they served small markets … The rest, as they say, is history.”
PE firms hadn’t worked out that software companies were incredible. They were still seen as ‘risky’ tech assets.
So Mark went on an absolute spree rolling these things up for peanuts.
But today, the tides are different. Every Harvard MBA wants to go buy a software company.
While Constellation has gone on an incredible journey, can CSU stock keep going up?
In the words of Marc himself, the only “barrier to starting a software conglomerate is a phone and a chequebook”.
In this financial teardown, I uncover some of the ‘secret sauce’ behind CSU’s success. Strategies and tactics that you can leverage to build your own capital-compounding juggernaut.
We cover:
- The average acquisition multiple they pay
- The number of deals they do per year
- How the manage ~700+ businesses
- How they structure incentives to run a ‘decentralised’ acquisition model
- If the model will work moving forwards
Let’s begin.
Constellations Acquisition Multiples
One of the dirty secrets about Constellation is that they play hardball when acquiring businesses on price – throwing in ‘lowball’ offers in deal processes.
In fact, Constellation prefers not to get involved with processes at all, and be the sole bidder for an asset.
Constellation is also extremely discreet about the multiples they pay these businesses, probably because they tend to price pretty low…
One of the research analysts covering CSU Howard Leung quoted Constellation doing their average deal historically on an 0.8x revenue multiple.
More recently, they have done some of their bigger deals at 6x LTM EV / EBITDA, which is certainly not pricey as far as SaaS deals go, and 2x-3x LTM revenue multiples.
Part of this makes sense – PE firms pump these software companies with leverage whereas Constellation has a buy and hold model, so they can’t afford to overpay.
But still – CSU is price savvy.
Deal Volumes
Constellation also does a boatload of deals. In 2023, they acquired ~100 businesses or so.
In 2022, the number was rumoured to be 143. That equates to doing a deal roughly every 2.5 days. 😲
Their internal database is perhaps the most incredible, and what allows the growth engine to scale.
“We have a database….(with) 40,000 opportunities in it. And we’re constantly adding more…
…we look at the number of businesses that were sold in a given quarter, and for those businesses that are also in our database, we calculate how many of those transactions we were aware of. And what we’re finding is we are not aware of close to 70% of the transactions, even though the company was in our database, and we supposedly had a relationship with the company
We are focused on how we improve that coverage ratio from, let’s say, 30% to something higher. I don’t think it will ever get to 100%, but could it get to 60%? Maybe. And if it does, hopefully, we could deploy twice as much capital” – Jamal Baksh, Constellation Software CFO
Each target has a contact name next to it, with the expectation of staying in contact 3 to 4 times a year. The corporate development team have KPIs on the number of touch points.
It feels like Mark and his disciples have their tentacles on just about every vertical software company under the sun.
At some point, you’d have to think CSU is going to run out of these VMS businesses to go after.
On structuring incentives
As a business owner, one of the biggest headaches is managing people.
One on one’s, KPIs, team meetings – managing people sucks.
So how does Mark manage ~700+ companies and 50,000 employees? He doesn’t.
Mark prefers to sit in his basement playing Dungeons and Dragons while his managers do all the work..
But seriously, Constellation is, by design, extremely decentralised.
Mark always hated authority as a kid.
He loved small, high performance teams from his time in venture, and he tries to keep Constellation that way.
“Our business units rarely get large. The biggest is 307 employees, and the average business unit currently has 44 employees. Two thirds of our employees are working in business units with less than 100 employees.”
What’s even better is that CSU’s head office only has 14 people!
14 people running an $88 billion dollar company.
That gives off major Berkshire vibes.
So how does this actually work? In 2005, Constellation was doing ~$200m in revenue and Mark was overseeing 30 business units.
Quickly he realised he couldn’t possibly keep his eyes and ears across everything, so he started delegating acquisition responsibility to the CEOs of his portfolio companies.
Now, the business units operate as independent M&A engines that can do acquisitions themselves without board approval. There is also a clear progression within Constellation that gears employees to become acquirers themselves.
The hierarchy
There are different ‘tiers’ of employees:
- Analysts
- Business Unit Managers
- Players/Coaches
- Portfolio Managers
- Craftsman
Unlike most companies, where you work your way up ‘within’ the existing business, at Constellation, you work your way up to a Portfolio Manager.
“A career path for an ambitious employee joining Constellation might be something like this: Immerse yourself in learning about the peculiarities of VMS economics. At some point, transition from analyst or knowledge worker into a leader of people.”
….
“For those whose ambition exceeds their good sense, we have a role that we call a Player/Coach. A Player/Coach continues to run their BU, but ambition drives them to acquire a sizable business, usually in another geography or another vertical.”
…
“Should the Player/Coach find a second or third stand-alone business to acquire, they eventually have to give up the day to day responsibilities for running their original BU and become a full-time Portfolio Manager (“PM”)
…
“If the PM is good at finding acquisitions, and helping them learn relevant best practices, and continues to deploy at least the FCF produced by their portfolio, then we refer to them as a Compounder.” – 2017 Presidents Letter
The most interesting thing is how well Constellation performed after delegating control to their employees. This is something unique about Constellation.
Today, all of management’s annual bonus packages are determined by:
Base salary x Company performance factor x Individual factor
Where many businesses use Adjusted EBITDA as their north star metric, Mark prides himself on having super super clean numbers. He has actually conjured up a few of his own metrics from the insurance world – tracking ROIC + Organic revenue growth.
“We still believe that the sum of ROIC and Organic Net Revenue Growth is the best single metric for measuring the short-term performance of our low asset intensity software businesses.”
The company performance factor is determined by net revenue growth and ROIC for the operating business. Essentially, CSU’s managers are disciplined about targeting ROIC as their “north star”, which is a relatively good metric and difficult to gamify – unlike other earnings measures like Community Adjusted EBITDA…
“There are three hurdle rates. Less than a million in revenue is 30% IRR; above four million, you can drop to 20%; and 25% is for everything in between. You can go to 20% because those bigger businesses are more competitive, but everything which falls in the middle at 25% is 90% of deals. There is another hurdle rate of 15%, but that is on deals so large they’re not in CSI’s wheelhouse. Above 50 million, you can go to 15% but they might only do one of those per year. Dropping those thresholds by 2.5% won’t have much impact.” – Former Constellation Software M&A Executive (Topicus Group) – Source
On top of this, for senior executives, 75% of their bonus is paid in cash which is required to purchase CSU stock, and escrowed for up to 4 years.
This is a genius move for two reasons:
- The entire company is banking on CSU stock to go up over time, creating alignment of incentives across the group.
- The shares are never diluted – CSU hasn’t raised a cent of equity capital since it started!
The result? A very wealthy group of shareholders and employees.
“To date there are over 100 CSI employee/shareholder millionaires. Ten years from now, my hope is that there will be five times as many” – Mark Leonard
Deal Size
Historically, Constellation has bought really small companies. Back in the 2000s and 2010s, they were rolling up companies for $1-$2m. But the reality is that for an CAD$88bn company with billions of dollars in free cash flow, it gets pretty tough to deploy all your capital buying tiny businesses.
Warren Buffett said it best: big isn’t always best when it comes to investing.
“Anyone who says that size does not hurt investment performance is selling. It’s a huge structural advantage not to have a lot of money.”
In 2023, CSU’s average acquisition size had risen to $28m, and they’ve publicly stated that they are willing to use leverage to compete with PE on larger deals.
Source – CIBC Equity Research
Given that Constellation literally has all of the VMS businesses in the world on their spreadsheet, it’s inevitable that they are going to have to go upmarket.
Interestingly, they’ve been structuring some of their recent deals as spinouts which has allowed them to pay higher multiples than a competitor could.
“In ring-fenced acquisitions, we’ll obviously use leverage. And we use it as a competitive tool if our competitors for those acquisitions are using leverage tied to that particular acquisition. And we ought to be competitive. We have no choice, but to do the same” – Mark Leonard, 2020 AGM
Organic Revenue Growth
The one real kicker of this model historically was organic revenue growth.
The great thing about VMS businesses is that they operate in niche markets that are too small to attract competition.
But that also means they’re super hard to grow.
Some of these markets are just so small. Once a business saturates its TAM, there’s not much else it can do to grow.
CSU’s appears to have been doing okay recently, but I suspect most of this is inflationary driven (sustaining >3% YoY growth since 2021). Mark used to track towards a double digit organic growth target in the 2000s, but it has become difficult to hit these numbers at scale.
Source – CSU Company Filings
The Future of Constellation
I know the question you’re dying for me to answer. Should I go long constellation?
Well, it’s not that simple.
Looking forward, a lot depends on the growth rate that CSU can sustain.
In 2013, Mark actually paid a bunch of consultants to work out what the business was worth. His conclusion was that the growth rate of the business has a massive impact on the stock price.
“Varying the organic growth assumption has a tremendous impact on the intrinsic value of a CSI share. Add in another 2.5% organic growth to the baseline assumption and you get more than double the intrinsic value. Subtract 2.5% from the baseline organic growth assumption and you lose almost half the intrinsic value of the stock. You can see why so many software company CEO’s are growth junkies.” – Mark Leonard
There’s already a lot of growth priced into CSU stock.
Constellation trades on a 22x NTM EV / EBITDA multiple and a 36x NTM P/E ratio, with a big premium baked, given Constellation’s insane track record.
What’s more – there are way more players in the space today than there were 10 years ago.
This will probably drive up acquisition prices and compress IRR’s.
“Externally, competition to buy vertical market software (“VMS”) businesses is intense. Vista Equity Partners and Thoma Bravo are two of the most prominent private equity (“PE”) firms that concentrate on software acquisitions. Roper Industries is a large publicly traded industrial conglomerate that we included in our HPC study and that also actively competes for VMS acquisitions. Vista currently manages approximately $28 billion of capital and Thoma Bravo is managing approximately $16 billion…
In the last 9 years, Roper Industries has invested five times as much capital in the VMS sector as CSI has since its inception, 22 years ago” – Mark Leonard, 2016
There is also some degree to which the Conglomerates model doesn’t scale. Capital becomes more difficult to deploy at high rates of return as the pile gets larger. Deals get more competitive, multiple arbitrages are harder to find.
This isn’t rocket science – it’s what happens to all of the large private equity funds as they continue to raise more capital.
Source – Exploring Context Substack
So what will Constellation do?
I think there are a few cards up Mark’s sleeve:
- Delve into verticals outside of VMS
- Use more leverage
- Start returning more capital to investors
- Build out their public company portfolio
Mark has hinted before that he is interested in acquiring businesses outside of the VMS space if they are attractive.
This is definitely possible to do for larger deals at the corporate level, but changing the company’s entire acquisition model will be difficult given Constellation built their model off the back of software.
Leverage is also another tool in the toolbox.
Constellation’s businesses are comically resilient, recording revenue growth of 49% during 2008 when the rest of the economy was tanking.
There is also a world where Constellation starts returning more capital to investors in the form of dividends.
Mark has outlined before that if he can’t meet the hurdle rates he desires for investors, he will start returning more capital from Constellation.
In 2019, Mark issued a special dividend to CSU investors. When asked about it, he responded:
“The dividend is clearly a signal about capital deployment. If we thought that we could deploy this capital at high rates of return, we would have kept it”
Finally, CSU also has the option of moving more of its capital into the public markets.
“…we are re-starting our public company investing efforts. During the period from 1995 to 2011, we made sixteen public company investments in the software sector. If you viewed our public company investments as a single portfolio, the internal rate of return (”IRR”) for that portfolio far exceeded our hurdle rat.
The average hold period was shorter than we would have liked, and most of the investments ended in the companies being acquired by third parties rather than CSI
We hope to find some attractive public software company investments in the coming year or two. At present, the pickings are slim due to generally high valuations” – Mark Leonard, 2015
All in all, Constellation software is an incredible business. Mark is one of the great capital allocators of the past decade, and has built a business which reflects his curiosity and love for software businesses. If you haven’t read Mark’s Shareholder letters, they are pretty incredible, and you should check them out here.
Sources & Other Great Reads:
Junto Investments on Constellation Software
