Vertical Market Software
Vertical Market Software (VMS) holding companies have some unique characteristics that put them into a class of their own. The software products they own take the opposite approach to that of the typical large tech company. By owning an array of products that cater to a specific industry or niche, a VMS holding company can create a robust portfolio of highly profitable and diversified software products.
The companies listed below are what we'll be covering in this article:
Both Lumine Group and Topicus are spinoffs of Constellation Software.
See our VMS companies page for updated data and metrics.
Company Strategy
The basic strategy for these companies is to acquire existing software products at a bargain price. Then hold onto them permanently. This includes building, maintaining, and growing the user base. This is a proven business model which can provide amazing returns. Since going public in 2006, Constellation Software started trading for around $70 (CAD) and has now trades at over $4,353 per share. Annual compounding has been well over 30% per year.
Serial Acquirers
VMS holding companies are serial acquirers. That is, they primarily grow by acquisitions rather than from organic growth. Similar to that of Warren Buffet at Berkshire Hathaway, the success of an individual VMS holding company is based on capital allocation. Allocating capital requires discipline and having high standards in order to maximize long-term compound annual growth. Having a disciplined approach means both avoiding overpaying for acquisitions and moving aggressively in environments where deals are favorable. We will cover Constellation Software's unique acquisition strategy in this article.
Because making a large volume of acquisitions has a number of management challenges, an unconventional approach is needed to avoid the problems that plague most other large companies.
Vertical Market Software
Vertical Market Software (VMS) is software that applies to a specific sector, industry, or niche. This contrasts with Horizontal Market Software (HMS), which applies to a broad number of companies and/or users. Think Microsoft Excel as an HMS product since spreadsheets are ubiquitous and used by nearly everybody.
VMS products have an array of attractive economic characteristics. These include:
- High margins
- High growth
- Asset light
- Sticky revenue
- Recurring revenue
- Moat
- Less competition
- High barriers to entry
In addition to being highly profitable, the profits of many VMS products tend to be reliable and predictable. This is due to high barriers to entry because each VMS product often requires an extensive amount of domain knowledge. In addition, because VMS products focus on a specific niche, the largest tech companies won't launch products that compete with them. This is simply because the market for any individual product is too small.
From the customer's standpoint, once a customer starts using a product, there tends to be an unwillingness to switch. This benefits the first VMS product that captures a large market share within a certain niche. Another important feature to note is that typically the product cost is a small fraction of the customer's operating expenses. In CSI's annual letters, Mark Leonard (CEO of CSI who we'll be covering in another section) wrote that the annual cost rarely exceeds 1% of a customer’s revenues.
Company Structure
The VMS holding companies covered in this article are structured much differently than traditional conglomerates. They make many small acquisitions and remain decentralized. The traditional conglomerate, on the other hand, makes large acquisitions and then centralizes operations. Business units within the VMS holding company run autonomously. The holding company itself is not concerned with economies of scale.
The headquarters portion of VMS holding companies is intentionally kept very small and lean. Mark Leonard refers to the traditional approach of scaling headquarters in large companies as "overhead creep". One key attribute that sticks out at CSI is the fact that headquarters-related costs in relation to revenues and operating income has consistently fallen over time. In his 2017 letter, Leonard stated that head office expenses fell from 3.0% of Net Revenue in 2004 to 0.5% by 2016.
Operating Groups and Business Units
As CSI got larger, the number of business units grew well into the hundreds. To deal with this from a management standpoint, the units were placed into operating groups. Similar to the company's headquarters, operating groups have their own portfolio managers that can even make their own acquisitions. In his 2017 letter, Mark Leonard wrote "Each of the Operating Groups is the equivalent of what CSI was ten years ago (plus or minus three years)". Each operating group is structured just like the company itself.
Finding Acquisitions
Constellation has been known for creating and maintaining a database of VMS software companies they would consider acquiring. "We track many thousands of these acquisition prospects and try to regularly let their owners know that we'd love the chance to become the permanent owners of their business when the time is right for them." (CSI 2012 Letter) Over time, this has contributed to CSI's core competence. Because they have a big advantage when it comes to information and experience, they are able to find better deals and avoid mistakes.
As CSI becomes larger, making a large volume of acquisitions has become a scalability problem. This requires delegating the acquisition process to others. In CSI's 2017 letter, Mark Leonard writes the company has "26 Operating Group and Portfolio Managers who spend >50% of their time on M&A". This was happening at a time in which CSI was trying to ramp up to 100 acquisitions per year. Amazingly, CSI has been able to maintain impressive growth rates up to this point. We will cover these potential scaling challenges again later.
Acquisition Options
Generally, there are several options when it comes to acquisitions:
- founder-led companies
- larger corporations
- private equity (PE) firms
The preferred acquisition option is a founder-led company. This has a number of advantages. The first is that they are the best cultural fit. These companies are small and scrappy and respond quickly to change. In CSI's case, much of their top talent comes from these types of acquisitions.
Larger corporations can be another source of acquisitions. The typical large company makes acquisitions poorly. Great serial acquirers like CSI are, in fact, very rare. The most common reasons for acquisition failures include overpaying for the acquisition, a lack of strategic fit, cultural mismatches, and underestimating shifts in consumer behavior or technological advancement. For example, eBay bought Skype to integrate voice communication into its platform, thinking it would add value to its auction business. The synergy that eBay was expecting never materialized and Skype eventually got sold again. This situation happens often enough that serial acquirers can find great deals.
Private equity firms are the third main source for acquisitions. Their investment strategy differs from CSI. While CSI intends to hold their acquisitions permanently, PE firms often have "exit strategies" attached where there is a requirement to cash in (after a period of say 5 years). For VMS products within a tight niche, there may be no large corporation willing to take it off the PE firm's hands.
Hurdle Rate
Hurdle rate is another important concept used by CSI. This is the minimum investment return required to make a deal acceptable. At CSI, Mark Leonard is very strict on this requirement because it enforces high returns as well as a high margin of safety. The one catch is that by setting a high hurdle rate, the number of possible deals dwindles. In a frothy investment environment, it may force the investor (serial acquirer in this case) to sit on cash until an environment with better deals arrives. As a result of this, CSI has made many great acquisitions during recessions.
The virtues of a high hurdle rate are such that Mark Leonard is unwilling to budge on this requirement. In his 2015 President's Letter, he stated, "when we dropped our hurdle rate, it dragged down the expected IRR’s (internal rate of return) for all the opportunities that we subsequently pursued, not just those at the margin" and that "hurdle rates are magnetic".
There is, however, some evidence that CSI has at least some flexibility when it comes to hurdle rates. Smaller acquisitions with less than a million dollars in revenue have a higher hurdle rate requirement than larger acquisitions (In Practice interview). Having multiple tiers based on company size makes sense because smaller acquisitions still require a fixed amount of time and energy by the acquirer and have a smaller overall impact on the parent company. Larger acquisitions, on the other hand, have more competition to buy them and have a bigger impact on the overall portfolio.
More on VMS Software
Attrition
For a specific software product, attrition rate is one of the most vital metrics. This measures the overall health of a software product. At CSI, Mark Leonard addressed this in one of his first annual letters. If attrition is high, customers may be:
- switching to a competitor
- don't believe in the value of the product
- going out of business
Attrition directly impacts organic growth. Unless the company is releasing new products that are more successful, the company's IRR will be low and profits will eventually dry up.
The macro environment is also another important thing to consider. Some sectors and industries are deeply cyclical. And the overall economic and credit cycles can't be overlooked. For any given product that's struggling, it's important to distinguish whether the attrition problem is due to the product itself or due to temporary environmental factors. Part of CSI's portfolio included software products for homebuilders which went through a crisis during the Global Financial Crisis (GFC) in 2008.
Recessions
Recessions, being part of the business cycle, have historically occurred every 7-10 years bringing major challenges and opportunities with them. Recessions tend to 1) cause struggle within the portfolio and 2) present the best acquisition opportunities.
As mentioned earlier, attrition generally increases during recessions. If a particular product is truly great, however, it can still gain market share while attrition is rising. When the environment returns to normal, the product can be even more successful than prior to the recession. Especially if some competitors leave the market or go out of business. In CSI's 2009 letter, Mark Leonard wrote "During the recession we believe that our market share in the homebuilding software industry has grown, even while our revenues and profits have decreased."
Recessions reduce the cost of acquisitions as valuations drop. This is from less competition and less credit being available. Economic booms have the opposite effect. These periods favor the seller rather than the buyer and result in high valuations. Having a strict hurdle rate in frothy markets shrinks the list of potential deals. Some serial acquirers, like Berkshire Hathaway, often have sit on large amounts of cash late in the cycle to take advantage of opportunities when bull markets inevitably come to an end.
Intangible Assets
Being asset light, VMS products have very few tangible assets simply because the software products are the main assets. The lack of tangible assets means that the VMS holding company's balance sheet will mainly consist of intangible assets listed at amounts far below their true worth.
The approach CSI has to amortization is that these assets generally hold or increase their value. Typically, assets are thought of as things that have a useful life over a certain period of time, but gradually depreciate in value as they become "used up". However, CSI only considers these assets to be depreciating if revenue fails to keep growing organically. These assets are tested periodically for impairment.
CSI uses an "adjusted net income" (ANI) to deal with this issue. Because net income will be understated due to amortization, ANI adds amortization back in. Deferred income taxes are also part of the ANI formula but this amount tends to be small relative to net income and amortization. When analyzing financial statements, free cash flow (FCF) should be compared to these net income metrics. FCF will contain some more adjustments from foreign exchange rates and other financing activities. But it should generally move in tandem with ANI.
Constellation Software
CSI is the best case study in the VMS space. As mentioned earlier, this company is unconventional and was unique before it spun off smaller versions of itself.
Constellation's CEO
At CSI, Mark Leonard, who we've already mentioned numerous times, is not a conventional CEO. Being a former venture capitalist, Leonard has combined the best of value investing with knowledge of the tech industry. He has also learned from great capital allocators and great serial acquirers of the past. A number of which are mentioned in his annual letters.
At one point, he stopped taking a salary. Instead, his compensation is tied directly to the performance of CSI's stock through the ownership of the company's shares. "Last year I asked the board to reduce my salary to zero and to lower my bonus factor. CSI had a great year, so despite those modifications, my total compensation actually increased." (CSI 2014 letter)
He also has a preference for frugality. Despite being a billionaire, he would fly economy on business trips. "I've traditionally traveled on economy tickets and stayed at modest hotels because I wasn't happy freeloading on the CSI shareholders and I wanted to set a good example for the thousands of CSI employees who travel every month." (CSI 2014 letter) In this letter, Leonard would go on to say that he would use his own money instead of company money if he did not fly economy.
Withholding of Information
Mark Leonard consistently wrote annual letters up until 2021. Since then, he has become much more secretive. 2018 was the first letter that claimed they would be limiting the amount of information being disclosed. Specifically, this entails not releasing too much information regarding new acquisitions. Increased competition has been cited as the reason. CSI's success has attracted copycats. The increase in competition can lead to having too much cash and not being able to deploy all the available capital (similar to Berkshire Hathaway).
That being said, the annual letters up until 2021 are still worth downloading and reading (see the links section below). In the next section, we will examine some financial metrics that have been most often highlighted in the president's letters and annual reports. These add additional clarity on top of the standardized financial statements reported every quarter.
Metrics Unique to Constellation Software
Combined Ratio
At CSI, Mark Leonard pays special attention to what he calls the "Combined Ratio". This is simply the sum of ROIC and Organic Net Revenue Growth. Being a measure that combines profitability with growth, Leonard wrote in his 2015 letter: "We have touted the Combined Ratio as the best single measure of CSI’s performance."
CSI's Combined Ratio reached 40% in 2014. More impressively, CSI has consistently sustained a high Combined Ratio for many years. If that's not impressive enough, CSI achieved all of this without using an extensive amount of leverage. CSI currently has a debt to equity value of around 1 and a debt to assets value of only 0.2. This is despite the fact that most assets are intangible and are underrepresented on the balance sheet.
Applying this metric to other companies should be done very carefully. The first thing to be aware of is that it may not be possible to separate organic growth from growth through acquisitions. CSI disclosed this amount in a number of their annual letters. When the typical company releases their financial statements, it may only report a single number for revenue.
This metric has some potential drawbacks despite its usefulness. It can become skewed in several different ways. First, using a lot of leverage can show both higher growth and higher returns on capital. This is because borrowing money to fund acquisitions requires 1) less capital to be deployed and 2) larger acquisitions are possible with less capital. This increases risk. And debt has to be repaid out of future cash flows. Some leverage is OK or even preferable. Used responsively, debt can be used strategically to fuel future growth. But it should not be used to the extent to which an unexpected event risks the company becoming insolvent.
Another red flag (or means in which the Combined Ratio can be "hacked") is through dividends and share buybacks. Returning money to shareholders takes capital off the balance sheet when money is deployed. Extreme cases are companies that borrow money fund the dividends and share repurchases. Less capital results in a higher ROIC. Of course, for many companies, dividends and share buybacks make sense. However, a company prioritizing growth should be able to deploy capital at high rates of return, making growth a better option than returning money to shareholders.
ROIC
Return on Invested Capital (ROIC) has long been considered one of the most important determinants of future returns. As companies get larger, a high ROIC becomes harder to maintain. Especially if the company in question already has substantial market share or is in an industry that's maturing. Companies that have a high ROIC but aren't growing much are able to do so by returning money to shareholders through dividends and share buybacks. Tobacco companies, for example, have been doing this for decades. This is because the capital base doesn't expand. In this case, a higher ROIC is easier to achieve because there's less capital available on the balance sheet.
Serial acquirers, on the other hand, are attempting to make investments that compound future growth. This means they tend to reinvest capital rather than distribute it to shareholders. Although both growth companies and cash cows can have a high ROIC, the results from long-term compounding can be much different. This doesn't mean the tobacco company investments haven't done well over the decades. They have clearly performed great. However, companies that can consistently reinvest capital for high returns will generate more compound growth.
To illustrate this point, we'll examine the 10-year ROIC and 10-year compound annual growth in free cash flow (FCF) for Altria (a tobacco company) and CSI. Notice that the ROIC is similar (around 25%) but the FCF compound annual growth rate (CAGR) is much different. CSI has compounded FCF by 23.6% per year while Altria has only compounded FCF by 7.9%.
Tobacco companies can still provide a good return despite barely growing revenues (only 1.5% CAGR per year for Altria). This is because they have consistently paid high dividends and bought back shares. In fact, it's not uncommon for a tobacco company to pay out over 60% of its FCF in dividends. Another important distinction is that Altria consistently trades at a low multiple (an EV/FCF of only 12.8). Compare this with CSI currently trading at 33.4 EV/FCF. Since CSI has reinvested capital at high rates of return rather than distributing it to shareholders, it has compounded growth in revenues, profits, and shareholder returns much faster. At 33.4 EV/FCF, the valuation is high at the moment (which we'll address shortly).
For CSI, maintaining a high ROIC also goes back having a strict hurdle rate. Buy not settling for mediocre investment opportunities, ROIC doesn't get dragged down as the company grows in size. Mark Leonard has stated that ROIC shouldn't revert to the mean if a business is truly great. In his 2012 annual letter, he wrote "If ROIC starts to erode significantly, then either we’ve damaged our existing businesses, or our new acquisitions are less attractive than those that we have made in the past." (CSI 2012 letter)
Valuation
Now that we've spent considerable time covering the importance of not overpaying for acquisitions, we must apply the same logic as investors. Of course, we want to invest in high quality, growing businesses. However, we must look at a company's valuation in order to avoid the mistake of overpaying.
There are many different metrics for assessing the intrinsic value of a company. It is important to distinguish between a company's intrinsic value and the valuation the market is putting on it. At times, these can diverge greatly. We will begin by looking at multiples of revenue. At the time of writing, CSI has an Enterprise Value to Revenue (EV/Revenue) of 7.2 (Enterprise Value is preferred over Market Cap because it takes debts and cash balances into account). According to the company's CEO, CSI is trading over the company's intrinsic value. In his 2015 letter, Leonard wrote "When really good companies start trading at 5 and 6 times revenues, it’s time to start worrying. I hope our shareholders are never in that position."
This is a typical problem. Investors usually know when companies are truly great and often bid the stock prices up to lofty valuations. That being said, there's no reason to get discouraged. All great companies go through rough periods when it comes to stock prices. This is why it's important to maintain a watchlist of great companies so they can be bought when valuations are more favorable. Sometimes these time windows can be very small, so it's necessary to always be paying attention.
Between February 19, 2020, and March 23, 2020, the S&P 500 dropped around 34% from its peak. The rapid decline was spurred by concerns over the economic impact of lockdowns, business closures, and widespread uncertainty. However, markets rebounded very quickly, with central banks and governments introducing stimulus measures that helped restore investor confidence and drive recovery in the following months. The key point was that the time window was very small. In typical bear markets, the time window will be much larger than this. By building a watchlist and preparing oneself to take advantage of opportunities to acquire shares in great businesses, a savvy investor will be in a position to act when great opportunities open up.
Although the COVID example isn't typical, it's important to recognize that macroeconomic events are not predictable. Which is why preparation is the key. Also, not all great opportunities develop from slumps in the overall economy. Sectors and industries go through their own trends. Many opportunities arise from companies that are performing well, but also miss investors' expectations. This can cause a selloff in the stock, especially if a stock was "priced for perfection". An overvalued company only has to disappoint (sometimes meaning earnings were barely short of expectations) to trigger a new downtrend.
At CSI, Mark Leonard prefers the stock price to be near intrinsic value. While most CEO's are trying to do whatever it takes to push the stock price higher, Mark Leonard would prefer to avoid the consequences of CSI's stock being overvalued. Based on his annual letters, he favors long-term investors over speculators. "If a stock is over-priced and sophisticated investors sell, they are generally replaced by unsophisticated investors who are ultimately disappointed." (CSI 2012 letter)
Being undervalued also has its consequences. Undervalued stocks get the attention of activist investors and attract takeover bids. From a CEO standpoint, having a stock price near the company's intrinsic value is generally best assuming the company is performing well.
Now that we've covered many of the unique features of CSI, we will now briefly cover two of its spin-offs which are also publicly traded. Since these companies are smaller and newer, they provide opportunities for investors that missed investing in CSI in the past.
Topicus.com
Topicus is a VMS holding company that is focused on the European market. It was spun off from Constellation Software in 2021. Topicus has a similar acquisition strategy to CSI. This company, along with Lumine Group, are considered to be where CSI was 10 or 15 years ago. Which means it's earlier in its growth stage with a long runway of future growth ahead.
The European market does have its challenges, however. The UK is a small geography with a fair amount of competition. Europe overall is less competitive than the US when it comes to generating a continuous supply of new tech companies. In addition, there are challenges with different cultures, languages, laws, taxes, and governments. Despite these challenges, revenue growth has been strong since the company was spun out.
From a valuation standpoint, Topicus is the cheapest of the three VMS companies. Having an EV/Revenue slightly above 5, it appears to be close to fair value. This could be, in part, due to analysts believing that growth will slow in the coming years. Free Cash Flow (FCF) margins, however, have been excellent at over 25%.
Lumine Group Inc.
Lumine Group is another VMS holding company that was spun off from CSI in 2023. Lumine's focus is on the communications and media industries. Like CSI, Lumine has a similar acquisition strategy where the intent is to buy and hold permanently. One key difference here is its focus on certain industries. CSI's acquisition strategy is much more broad.
At the moment, Lumine's valuation is quite high. Revenue growth has been very strong and is expected to continue.
Capital Allocation
The benefits of being an "asset light" company, as described earlier, show up in the company's cash flows. In fact, growth in cash flows can outpace growth in CAPEX with CAPEX already being a low percentage of revenues (around 0.5% for CSI) and operating cash flows. The chart below shows the effects of CSI's ability to compound free cash flow over time. (Topicus and Lumine don't have enough historical data at this point but results should end up being similar.)
Because CSI is a serial acquirer, the best leading indicator of revenue growth is how much was spent on acquisitions. The chart below shows that each time there's a spike in acquisitions, revenue growth will also spike in the following several quarters.
The relationship between acquisitions and free cash flow (FCF) also exists but has a slightly longer lag. And FCF Growth is much more volatile. Because of this, we zoom out to view this on an annual basis in the chart below.
Notice also that in the charts above, the amounts spent on acquisitions have grown exponentially over time. This is crucial to take into account because, as CSI continues to grow, larger (and more) acquisitions will be necessary for the company to maintain its growth rate. This can happen by increasing the volume of companies they acquire each year or by acquiring larger companies. Since this could become difficult at some point, we'll next explore some of the arguments against investing in VMS serial acquirers.
Arguments Against Investing in these Companies
Limited Ability to Scale?
A number of investors believe that once a VMS holding company, specifically CSI, reaches a certain size, it will run into a wall and not be able to continue scaling. It's worth noting that with CSI this has been a concern for quite a long time. However, Constellation has continually been able to grow efficiently and prove the doubters wrong.
The spinning out of Lumine and Topicus gives the investor multiple options to invest in VMS holding companies. Since these spinoffs are much smaller that CSI, it can be argued they are both where CSI was many years ago. This means they may have a long runway of future growth ahead of them.
Delegating M&A Responsibility
In recent years, CSI has not only acquired more companies but has also increased the volume of acquisitions over time as well. Increasing acquisition volume leads to more M&A responsibilities that have to be delegated. The argument here is that capital allocation quality may drop. "The further you push that responsibility down, the caliber of M&A execution gets worse." (In Practice Interview)
As mentioned earlier, CSI has an information advantage when it comes to making acquisitions. This is because of both their experience and the database of companies they track. Also, by having strict discipline when it comes to making the acquisitions (like the hurdle rate), they are better able to scale the M&A process.
Increasing Competition
Growing competition is another concern with investing in this space. Private equity and hedge funds will be the largest source of competition. More competition will increase purchase price multiples for future acquisitions. This can lead to a situation where these companies have to pass on opportunities if potential returns fall below the hurdle rate. This can mean competitors including private equity and hedge funds are willing to accept lower returns. This situation would limit the acquirer's ability to grow.
In frothy markets, it's common to see new competitors enter the market and bid up prices. This can continue until there's a downturn in the industry or the overall economy. Companies that overpaid for making acquisitions (thus providing poor returns) will eventually leave this market. This is why downturns have been historically great times for companies like CSI. There are more deals available that meet their strict standards.
AI Disruption Credible Threat?
I don't believe the risk is as high as some popular narratives claim. One such popular narrative as of this writing is that AI reduces the cost of building software (which is generally true). However, the narrative continues with claiming that more clients will simply build their own software products.
The problem with this notion is that just because software development productivity increased because of AI, it is still not sufficient enough to meet the needs provided by VMS products. This is especially true if the software product requires any degree of complexity. Not to mention that AI will be very mistake prone and create a messy code base that will actually result in more time and energy fixing the problems. VMS product customers will not switch to a chatbot-like product that doesn't consistently deliver accurate results. Instead, the VMS providers themselves will eventually adopt AI features into their already existing products. A niche software product is also a reflection of domain-specific knowledge that may not be available in the broad datasets that are fed into general AI products.
It's also important to look at the typical customer of a VMS product. Many of these customers are running small to medium sized businesses. Their time and energy is already stretched thin and the software product typically has a low cost (compared to total operating expenses). The idea of switching isn't going to be worth the risk and effort.
Wrapping It All Up
VMS holding companies have proven to be great investments. If led by a great capital allocator, these companies can build a portfolio of many profitable VMS products which together results in long-term compounding.
We've created a VMS Companies page that provides updated data and metrics. Beyond that, reading CSI's annual president's letters is another great resource. The Serial Acquirer Primer by Gustaf Hakansson is also quite helpful and is available as a free download. The links to these resources are listed below.
References
Constellation Software President's Letters
Serial Acquirer Primer. Book by Gustaf Hakansson. Free download.