Post New

I had an alert triggered today. So posting it here as seems like this is a well loved company.

AI summary of cause is
The recent drop in its stock price appears to be primarily driven by the resignation of founder and President Mark Leonard in September 2025, which raised investor uncertainty about the future leadership and strategy, especially concerning the impact of Artificial Intelligence (AI) on the VMS sector. Despite the stock pullback, the company recently reported solid Q3 2025 results with strong growth in cash flow, though revenue slightly missed analyst expectations.
-----
would like to learn more about what people llike/dislike about this firm now that founder aura is gone.

I’ll be extremely brief on this one:

No AI here!

- Quote in slow motion free fall as investors exit after CEO Leonard resigned for health reasons. In my research his departure isn’t that important because the capital allocation has been distributed years ago away from Leonard.
- Further selling pressure from paranoia that AI will replace the need for their software. This reason is very naive when you are familiar with specialized vertical software. I estimate AI hype will be a net benefit for organic growth as it will stimulate auxiliary upgrades but the systems (flight scheduling, travel booking, insurance brokerage and billing, etc) cannot be replaced.
- They are more than anything a value investing compounding machine and have compounded 30%+ the last 20 years. That wasn’t a typo. In recent years compounding around 20%.
- Their circle of competence is small/ medium specialized vertical software companies. I know this industry well and revenue is insanely sticky. It will take a longer post to explain that. So what they do is buy with a return on invested capital in the 20-30% range.
- They get such a high 20-30 percent return range precisely because they are buying firms that generally dot *not* grow revenue but just sustain it with 20-40 year life cycles - so is like buying infrastructure or property, but with much higher revenue/cost ratios. So there is really high growth by acquisition but almost no organic growth. That is deliberate. The zero organic growth *is* what allows them to acquire at such a low multiple to get the 20-30% return on invested capital.
- They also do deals that allow the seller to continue programming independently without interruption in most cases. Part of their most is the trust they earned for allowing extreme autonomy for the software firm being purchased. Berkshire shareholders who have seen interviews with owners of their acquired firms will get this.
- Unlike other growth by aqusition firms, which are centralized for capital allocation even if when allow their purchased firms to run independently - Constellation uniquely allow extreme autonomy for *both* the acquired firm, and also the capital allocation process itself. Buffett allows capital allocation autonomy with just two investors but Constellation Software have applied this autonomy in an systematic way and they have proven that it works for years, with Leonard involved in fewer and fewer deals himself but the huge returns on capital having continued.
- So they have over 500 software firms and several separate capital allocation companies between constellation and the small software firms, each with a slightly different culture; but the same extremely disciplined value investing mandates. If a candidate cannot be acquired with a ROIIC of more than 20% it is skipped instantly. Their database of candidate firms has more than 50,000 firms.
- Do not price by EPS (thus ignore PE) as they use amortization to lower the earnings (with a bonus that they pay lower tax) and hold what they buy forever -- to an even more extreme degree than Berkshire. They grow not by the software firms charging more over time, but by using all the cash flow to chase very high return on invested capital. So value them by free cash flow; both rate of cash flow growth to represent how they are growing; and combine that with price to free cash flow ratio to gage your own return from the shareholder perspective.
- Can buy today at under 20 x free cash flow; peaked at 37x, and had usually been around 25x (though lower than 20x prior to 2015 when the market was still working them out) So your investment return from here should be similar to their return on incremental invested capital (ROIIC) presuming they don't trade greatly under 20x free cash flow when you want to sell.
- This is the kind of firm you hold really long term, but in recent years it has been impossible to get it at reasonable cash flow multiple.

I would anticipate continued downward falling knife effect for a few months, but any purchase from now (around 20x free cash flow or lower) should give 15-20% returns over the next 15 years or so, with some multiple gain plus acquisitions continuing at 10-15% ROIIC for a couple of decades.

Ok I wasn’t so brief.. but just wanted to cover the crucial factors.

- Manlobbi

Many thanks for a very nice overview. I agree and initiated a small position a month ago for similar thinking/reasoning. I was just considering adding to it on its further price decline and the market overreaction to the CEO transition and AI concerns imo.

I also hear excellent things from very good sources wrt the very new CEO, Mark Miller, who has been an outstanding leader there for decades. Mr. Miller also has a lot of skin in the game, as my info reveals he owns over 1% of Constellation. The successful culture, management and allocation/ growth will continue.

I also hear excellent things from very good sources wrt the very new CEO, Mark Miller, who has been an outstanding leader there for decades. Mr. Miller also has a lot of skin in the game, as my info reveals he owns over 1% of Constellation. The successful culture, management and allocation/ growth will continue.

I listened to as much of Miller that I could get my hand on and he does seem excellent. He communicates in a rather technical way related to his main role as capital allocator. He was involved with Leonard in all the important deals and plays the same game. As far as I could perceive, when Miller talks, his ideas are coming straight from his mind with conviction, rather than just trying to follow Leonard's values.

Miller recently explained that you can't grow your business in a decentralized way without letting people have a lot of rope, make mistakes, and part of the built in trust over time. Leonard followed Miller's remarks by saying that by default when a firm grows, if you don't have trust then you are forced to supplement with bureaucracy, such as a guard watching the guards, and the bureaucracy grows in a geometric fashion, so you have got to have trust to avoid being eaten by that. If you *don't* trust, then you *do* need systems, and then the bureaucracy. He want on, "Does bureaucracy naturally shrink? -- Generally .. *not* - so you have to actively shrink it." Miller followed immediately, almost viscerally, "When your central headcount is growing in proportion to the business - you have a problem, so you have always got to ... be let other people figure out how to spend their money, rather than you becoming a central taxing authority downloading all those costs that you built up over time, so that's a key measurement".

- Manlobbi

Great post.

This is just an out loud ramble really from a very much non expert about the company, or industry. But hey, we are hear to learn and test our thinking.

Clearly this is a fine business with its hooks in a lot of meat. I am not in the buying mood of late but do have this on my radar, if it got very cheap, which of course may never happen.

As I am not buying and do not own it, I can comfortably attempt to throw rocks at the idea.

My take is that the expectations for software quality have been raised incredibly high, with consumers enjoying great software from companies like Amazon, Microsoft, Google and Apple. And of course, the tools that now exist, to build better and better software are more accessible. I imagine Constellation’s software is largely in the commercial world and they operate in all kinds of little niches that make it less attractive to competitors due to lack of scale.

However, the expectations of end users have and are increasing. In my limited experience, there are a lot of software solutions in the marketplace today that are pretty awful.

Paying for the new systems that are possible, is a big barrier to progress. As is risk aversion and also leaders of companies not fit for purpose and sitting in leadership positions until they get their pension. It does take a fairly wide range of qualities in a CEO to completely rebuild an IT stack (as the kids call it). But I do wonder about the build up of these poorly performing legacy systems. Over time they will surely get replaced. No doubt Constellation will be there with the modern solutions and will get the technology advances to work for them.

But there may be organisations that are so tired of being squeezed, that they just want a fresh start with a new technology provider. Maybe the transition to cloud and AI is the moment of great change. Then again, that new provider increases the risk even more. They don’t understand the business processes the way the legacy provider does. The more I try to convince myself that Constellation could loose lots of customers, the less convinced I am that that is true. I suppose the rise of highly configurable systems is a threat, as that model can work well in some circumstances. But again, most of these little niches need bespoke solutions.

Another advantage as an investment, is that the revenue stream is so diversified, that even if they lost clients to new software providers, there is no concentration risk.

I would want to understand if the subsidiaries are innovating and bringing in new customers and riding the wave of new technologies.

I do wonder why they pay low multiples for their acquisitions and why I would think it attractive to pay double or triple what they themselves are paying for these little cash cows that could eventually die.

I am also a little sceptical of the idea that the original founders of the subs will stay motivated. Will they not tire of driving their businesses forward eventually, as they get older. I imagine it takes unusual determination and focus to run a software company and burn-out is always a risk. Will a salaried CEO of a subsidiary have the all round skill set to deal with the challenges and pressures. Then again, you can argue that an established subsidiary is much easier to run, than it was during the early years.

I guess I’m saying it could be an interesting sticky idea but I would like to pay 10 or 12 times owner earnings to have a great chance of a nice return. I haven’t looked at the numbers but imagine it’s a long way from that.

I have seen organisations with crappy systems, under the constellation umbrella, get tired of the costs of maintaining legacy systems and have stopped introducing new functionality due to what can seem like poor value for money for the improvements and then eventually start again with a new provider. But that is not a statistically sound survey.

I’m maybe less concerned about the acquisitions drying up. Buy of course, it gets harder to move the needle just like at Berkshire.

Many VMS packages exist purely to satisfy deep regulatory needs (e.g., specific government or healthcare billing). This functionality cannot be replaced by a general-purpose, high-gloss SaaS application.

A very very important point made above by Manlobbi that I didn’t fully appreciate initially!

I will interject myself into this conversation with a out loud ramble in response to this out loud ramble.

In the field, there's a lot of concern about development velocity, moats, and what constitutes a durable competitive advantage for a software company in a world of AI enhanced products, AI enhanced development, and independent agentic processes.

If you have a legacy software vendor with a public service interface, it is easier than it's ever been to functionally reproduce that interface by a competitor.

If you have an existing product, you have to evaluate if you still have product-market-fit in a competitive landscape with other tools that integrate agentic flows. To not sound like I'm just spouting jargon, I'm talking about something like Microsoft Word and whether it's valid to assume that all customers want to use Word to author .doc files, or if they want it to accept prompts and co-author a .doc file with Copilot.

If you run a software development organization, you are wondering about how much are leveraging model based code assistants (Cursor + chat) or agentic development tools (Claude Code). You are wondering about what a target for a product developer looks like now, and if your competitors are leveraging these tools better than you organization.

If you are a software architect, you are wondering if you need to develop proprietary models and have previously deterministic components of your platform be replaced with agentic processes that interact with other components via things like MCP (model-context protocol: think having code you write interact with code someone else wrote, but instead of via remote procedure calls, doing a glorified google search). These agentic processes may replace humans in the operations of your platform or add new capabilties.

Having said all that....

Mungofitch wrote something about missing the old crappy internet that I almost replied to. I have joked about a "old web" startup that is human only. Ironically it would be a lot like AOL depending on how you look like it (a private hyperlinked document graph) and Google (pagerank, but with a penalty in detection or reporting of generative content). I do not plan on quiting my day job.

But everyone knows basic LLM generated content when they see it. There is an uncanny valley quality to it that I personally find repulsive.

There's a joke domain (lmgtfy) for "Let me google that for you" and when I get an LLM generated piece of content, my mental reaction is something like, "Well, if I wanted ChatGPT's opinion I would have just asked ChatGPT myself"

I find the impact of AI, in both transformer models and other frontier models in active research, to be disruptive of civilization as we've known it, grossly overestimated, and mostly misunderstood; all at once.

So back to Constellation.

I think that these small operators they roll up understand niches that are consumed by humans with unique and specialized needs and the good ones probably require a great deal of time to develop the requisite understanding of the history, regulatory environment, place within the broader economy, and many other factors.

A little aside about AI, and particular transformer models (the T in ChatGPT), is that context window is limiting. What I just said has a specific meaning in linear algebra, but you can think of it as "the irreducible complexity in a thought that a model can work with". Consider long, formal conversaton greeting and "hey". The former may be dozens or hundreds of words, but could be summarized as "hey, spoken in formal English". This gets into information theory/Claude Shannons work a bit, but if you think of context window as applying to the uncompressible/irredicuible complexity of a thought expressed in words, it is close enough. To my knowledge - which is not at the frontier research level, to be clear - the computational complexity associated with context window size is O(n^2) in so it gets increasingly expensive to have models interact with increasingly complex thoughts. Again, imprecise in a technical sense, but to get the point across: if it takes 100 computers to reason with thoughts expressible in 100 words, it takes 10,000 computers to reason with thoughts expressible in 200 words.

Back to Constellation again...

I bought a share. I hope these little niches are complex enough that current and near term models will not be able to disrupt them economically. It may even deepen the moat of specialized operators that are capable of taking advantage of those tools.

Thank you for the suggestion, homosapien, and follow up posts, Manlobbi and EVBigMacMeal.

First, a correction: "Thank you for the suggestion, homosapien, and for the follow up posts Manlobbi, EVBigMacMeal, and WEBspired.". Sorry about the omission, purely my error.


As a postscript to my first post, I've known about Constellation for some time, but don't follow them closely and had missed the drop since this summer. I was trying to work out the catalysts, and sharing what I've found the ADR price has dropped since roughly July 3; with a pronounced drop on August 11 and another pronounced drop on September 19. Those appear to line up, respectively, with

- August 8, 2nd quarter earnings release: https://www.csisoftware.com/category/press-release...

- September 25, they announced the Mark Leonard's resignation https://www.csisoftware.com/category/press-release...

My suspicion, with no actual evidence, is that the CEO health issues were picked up by the market prior to the official announcement.

There did not seem to be any correlation in the price with the AI call on September 16: https://www.csisoftware.com/category/press-release... (transcript: https://seekingalpha.com/article/4824855-constella...). This was an interesting and unusual call. More unusual than interesting.

Based on my read of the earnings report and some articles discussing the results (like https://dariusdark.substack.com/p/constellation-so...), it looked like the Q2 results were fine from a cash flow perspective (keeping Manlobbi's comments about earnings, amortization, and cash flow in mind), but earnings disappointed due to non-cash charges in forex (118M vs 4M yoy) and a "IRGA/TSS Membership liability" item related to a spin-out, Topicus, https://finance.yahoo.com/quote/TOI.V/ (discussed in https://thecompoundingtortoise.substack.com/p/cons...).

The biggest weaknesses I can see are that the company is already modestly generously valued, complicated, and follows a private equity playbook (minimizing product growth and investment, maximizing cash flow) that relies on the availability of acquisition targets.

I think of two very unrelated things looking at Constellation - niche data and backtesting companies that pop up in Mechanical Investing and Palantir. Some of the MI board companies seem to charge $100 a year and look like they haven't updated their websites since 1998. Palantir has an AI reputation but has also rebranded professional services as forward deployed engineering - it is deeply integrated with customers and supposedly exceedingly sticky as a result (in a NDR sense).

Without Manlobbi's pitch I may have passed, but will keep digging. The share price drop preceded the CEO departure and cumulatively represents about $30B USD in lost market cap, so I'm having a hard time digesting there isn't some fundamental issue. But if there is I haven't found it.

Picked up some a few days ago. Seems like a good thing for a long hold. Planning to keep it until I need sell some stuff to pay for my Panda Express.

Been watching this stock since Manlobbi mentioned it. Heavy on the downside today but I haven't seen any news. Any owners / followers of the stock have knowledge that attributes to today's heavy fall?

jg

I haven't seen firm specific news.

It appears CSU is falling in line with several SaaS stocks, likely due to continuing perceived AI threats (and especially agentic AI developments, such as Claude's recent release).

It seems a baby out with the bathwater kind of situation to me, but I too would be interested in others' opinions.

Initiated a position today. If the "Canadian Berkshire Hathaway of software" is good enough for Manlobbi at 20x FCF, then it's good enough for me at 16x FCF.

This is my first new single-stock position in a number of years and I hope to own it for decades.

Down 6.5% as I type. This stock is truly a falling knife. Haven't taken a stake, but I've been watching it since Manlobbi mentioned it a couple months ago.

jg

I bought more CNSWF today @ 1,663

I wasn’t so lucky. I had a limit order to buy at $1600. Shortly after opening it dipped to $1620 then reversed sharply upward and now is above $1800.

It appears CSU is falling in line with several SaaS stocks, likely due to continuing perceived AI threats (and especially agentic AI developments, such as Claude's recent release).

Here is an extremely thoughtful article about the possible implications of various coding tools for software verticals, with specific mention of how Constellation has made the economics work. Spotted at the "Science, Technology, Engineering and Mathematics" board, thanks to weatherman
https://findthemoat.com/2026/01/27/re-pricing-the-...

To summarize it beyond the point of utility: for an investor, the main risk isn't that the clients will build their own tools. It's more that the barriers to competitive entry are lowered, so at the margin there will be more competitors. Ongoing spend to keep market share might hurt the long term economics of bond-like software products in non-growing niche markets, as they have traditionally relied on the fact that it just wasn't worth someone else's time (money) to build a new competitor in a small market.

It's very hard to say how much this going to be a problem for the economics of Constallation's past and future acquisitions. I can't see it killing them, but it has the ring of plausibility as a risk to consider. I can imagine this could cause their long their cash flow growth curve to flatten.

===========================

On an entirely different front, there is another small concern I have with Constellation. They don't really focus on dividends, tiny and flat for ages, they reinvest in the business. Fair enough.

But, as their individual products tend not to grow much, the lion's share of their growth in cash flow generation comes from that reinvestment. So cash flow per share rises nicely over time. Again, fair enough.

However a given year's cash flow can only be used once: it can be used for fast growth of more cash flow, or as a non-growing cash cow (usually to dividends), but not both. You can consider it a firm with fast rising earnings that can never be paid out, or a cash cow firm without growing earnings that could pay it out. But you can't consider it as a firm with fast growing "owner earnings", because it isn't. A bit of the same issue I have with Davita but without their buybacks.

The question then arises: if you can't GET those earnings unless they stop the breeder reactor business model, should it be valued as a cash cow? i.e., maybe 9-10 times whatever the owner earnings would be now in a steady state? In other words, (1) if they never did another acquisition, what would they be worth today as a non-really-growing profitable firm? And (2) if they do more acquisitions in future (with earnings that could otherwise be sent out to you as dividends, say) then what is the explicit reasoning that the firm is worth more than in situation (1)?

The really short version: what is infinitely delayed "jam tomorrow" worth?

I'm not saying it's a terrible pick at these levels, and in fact I do own some. Just always trying to look for flaws in any investment thesis.

Jim

To summarize it beyond the point of utility: for an investor, the main risk isn't that the clients will build their own tools. It's more that the barriers to competitive entry are lowered, so at the margin there will be more competitors. Ongoing spend to keep market share might hurt the long term economics of bond-like software products in non-growing niche markets, as they have traditionally relied on the fact that it just wasn't worth someone else's time (money) to build a new competitor in a small market.

It's very hard to say how much this going to be a problem for the economics of Constallation's past and future acquisitions. I can't see it killing them, but it has the ring of plausibility as a risk to consider. I can imagine this could cause their long their cash flow growth curve to flatten.

...

I'm not saying it's a terrible pick at these levels, and in fact I do own some. Just always trying to look for flaws in any investment thesis.


Another thought to consider comes from my brother, with whom I was talking about Constellation yesterday. He figures that in the software niche that Constellation likes, customer acquisition cost is typically the equivalent of about 2-3 years of revenue, but then you have a very long tail of revenue with very low costs and customer churn that is low, typically <10%, so when Constellation acquires these companies, they cut software development costs to the bone and just milk it for the long tail. Now with new AI tools , a certain percentage of customers will use agents to get the job done without the software, and churn rates are increasing. They don't have to go up a lot for the economics of this business to really fall apart in a serious way. To take a simple example, if you have $100,000 in revenues from 10 customers paying you $10,000/year, and one of them has to be replaced for $25,000 (2 1/2 years of revenues), you make $75,000. But if you now have to replace 2 customers a year instead of one, then your net income drops from $75,000 to $50,000.

If this is true, Constellation's model was heavily based on low churn, and AI will increase that both because competitors find it easier to write software, but mostly because it only takes a small percentage of customers that can use AI agents to get the job done to put a big hole in profits.

Thanks for the link and thoughts -- I mostly lurk but always like your posts.

As for lower barriers to entry, Sven Carlin asked Gemini to do a "deep dive" on the best bear case (https://www.youtube.com/watch?v=frh_-XyzhHU), and it came up with the following scenario: An AI-savvy startup creates a free AI-based "vampire" interface for the software in question, which can make the user's job easier (e.g. answer phone calls) while quietly subsuming the underlying database. Eventually it can do the same job for no monthly fee, collecting only a small cut of transaction fees, at which point the user is happy to make the switch.

As a recent buyer of CSU, this does worry me a little. But not much. For example, what's my motivation to create this brilliant vampire app, if I'm making much less than CSU was from it, and moreover if someone *else* can come along and disrupt *me* just as easily? It's not like I can look forward to billions in revenue if I succeed. This scenario also seems to ignore some fundamental issues such as whether AI API calls will remain as cheap as they are now (the hardware/energy costs aren't insignificant even if the trained models get cheaper, and they're currently heavily subsidized by investors). And LLMs are still probabilistic and thus subject to bugs which can kill a business and require ongoing maintenance in any case. Maybe most importantly, this scenario will still take years to play out, during which CSU is assumed to sit around and let it happen.

Valuing CSU does seem tricky too, as you say. Are they a Berkshire, but focused on cigar butts and without the free float? Viewed that way, their value depends mostly on their investing acumen (and opportunities), I guess. Or are they an Amazon, which famously had essentially no earnings for most of their growth period, but is now worth $2T or so? I'm not an analyst, so I rely a lot on other investors I trust (Manlobbi and Sequoia in this case). I highly doubt they're the next BRK or AMZN, but I'll hold until I see concrete evidence that AI competitors are at least *trying*, let alone succeeding.

- Further selling pressure from paranoia that AI will replace the need for their software. This reason is very naive when you are familiar with specialized vertical software. I estimate AI hype will be a net benefit for organic growth as it will stimulate auxiliary upgrades but the systems (flight scheduling, travel booking, insurance brokerage and billing, etc) cannot be replaced.
- They are more than anything a value investing compounding machine and have compounded 30%+ the last 20 years. That wasn’t a typo. In recent years compounding around 20%.
- Their circle of competence is small/ medium specialized vertical software companies. I know this industry well and revenue is insanely sticky. It will take a longer post to explain that. So what they do is buy with a return on invested capital in the 20-30% range.

Slip of the finger, pressed send instead of preview.

Anyways, Manlobbi, I was quoting the above as I'd be interested in your expanded thoughts on the italicized if you have that possible longer post in you.

Apologies for my question being via two posts and not one.

---

For a bullish take on why CSU may come out the other side in fine shape: https://tidefall.substack.com/p/breaking-down-rbcs...

Post New

US Policy FAQ

---

Contact Shrewd'm
Contact the developer of these message boards.