Invisible Companies
colossus.com
colossus.com
Some of them are a little more visible simply because they’ve scaled and developed a moat - logistics, particularly trucking in the UK, and international cargo shipping, stand out here for me. But also social care, floristry and grocery wholesaling, weird sectors in aviation (I worked in that for a while), all sorts.
And then I look at family businesses that shouldn’t be able to exist all the time. In my nearest town centre there is a grocer, a fish shop and a butchers on the high street. Common sense tells us the extra large supermarket built on the outskirts of the town in the 1990s should have obliterated them, but no.
In the UK there is a common assumption that a lot of vape shops, barbers and even some takeaways are fronts for money laundering (and that definitely happens - there is a chicken shop in one Northern town I know that to my knowledge has no ability to actually sell you chicken, nor can the 4 others next door to it). But there are definitely some that are making good money via the legal route.
As a software guy I look for opportunities to sell to them and there’s one theme that stands out across many of them that is also an invisible sector: have you noticed how many small companies make and sell touchscreen tills/cash registers? In the UK it seems to be a thriving cottage industry.
Individual sales are in the thousands, so they operate more like high-end jewelry stores than a "normal" business. One sale a day is minimum 250K/year.
The store itself seems to be more of just a show room. There are other sales routes too. Selling into hotels would be an example. There are also maintenance services, so a high end rug in some sense can be seen as a recurring revenue source.
It's a interesting little corner.
Then I had a house full of oriental rugs and haven't needed a new one ever since.
One of the ironic reasons that Ikea/Wayfair do well is their stuff looks good but is basically garbage and can't survive a move or two so people are forced to buy from them more often than they would with better brands. Better brands are too expensive (and also a bit stodgy usually) so younger and poorer people just keep getting the same lower end stuff repeatedly.
Amazing.
[The Amidi family through] its various holdings and brands, currently owns and manages more than 1 million square feet of real estate space along with assets under management exceeding $2 billion.
This guy was really good at picking winners.
> Rahim co-founded the Amidi Group in 1980, a conglomerate of privately-held business enterprises consisting of several domestic and international entities with significant manufacturing, distribution, trading, retail, and real estate holdings
Probably 2 or 3, all based in China, that will happily print your company logo on the machines.
It’s easier to get away with large margins and not spawn competitors if nobody scrutinizes you, and it’s easier not to get scrutinized if you are small or the domain is boring.
Not a knock on the article. It’s nice to double-click on a concept and explore it with examples and from many angles. But for me, it would have been easier to start with that framing, because it took way too long to understand the purpose of the article, atleast for me.
I reread listened to the article now and looked up some research due to the things that bugged me.
The summary I made is what’s uniquely interesting about such firms. But from a rollup perspective it’s just one of potential mechanism for a firm to be on the cheap. And even if you think that is a usefull angle on the topic, it’s still covering just a the subset of such firms that are invisible AND has good margins AND the owner is willing to sell on the cheap due to ignorance of lack of buyers so they can’t get good offerings.
I could have expanded the summary with: “ and because nobody’s bidding, they’re cheap to buy up and consolidate, which is where rollups make their money.“
The problem with that is that it a claim, and it’s at best not well founded and maybe even wrong. There are many failures in the same industries the article celebrates. Loewen Group rolled up funeral homes and went bankrupt in 1999, the 1990s physician-practice rollups collapsed, Waste Management itself restated years of earnings in 1998 in one of the largest accounting scandals of its era. None of that is in the article.
And many of the article’s examples (marina software, niche aircraft parts) are markets too small to support a second firm at efficient scale. If so, nobody enters not because they didn’t look but because they looked and correctly declined.
So yea, the article is a bit scatterbrained and much more speculative than it pretends.
Dry cleaners.[1]
Parking lots.[2]
7 boring businesses.[3]
[1] https://www.youtube.com/watch?v=_B0cASirKyE
[2] https://www.youtube.com/watch?v=La4SAUvmKz4
[3] https://www.youtube.com/watch?v=ebaiF97mCtM&pp=ugUEEgJlbg%3D...
Now lets talk about factuality. A human creator can just as easily lie or not fact-check all their sources. But if they do decide to fact-check, wouldn't it be easier to just task the AI to fact check. Simply tell it, "leave out any data that has not been fact checked and mapped back to a reliable source."
Not "as easily". A human creator can make a video without fact checking anything, but people using AI can push out thousands of bullshit videos in the same amount of time.
The human creator also actually sees the content they put into their video and can catch the most obvious errors that don't even require fact checking. Humans using AI can just keep spewing out slop as fast as possible without having any idea of what any of the videos actually contain.
Usually I find out by noticing symptoms of worsening quality, costs increasing more than I might expect, ramp up in aggressive cross selling of services and subscriptions, shifting to call services that are clearly not local and know nothing of our area, etc.
In some cases I'll get an employee that knows the situation and let's spill the PE sale and then I need to find a new service provider.
With the expected outcome when those vampires touch anything.
He'd started in his 20s as an industrial process engineer and eventually opened a string of carpet stores, then got into restaurants, earned a commercial real estate license and started buying his own stores instead of leasing, then had the opportunity to buy this bar, a rundown, tiny, poorly lit dive in Belmont Shore, a terrible place in a great location, and it eventually became the number one seller of Jameson on the entire west coast. What made owning a bar so profitable? No food, which is low margin, expensive and difficult to get approval for, requires more skill on the part of your labor to make well and handle. Low labor costs because they earn tips and don't need a real salary. Fixed space cost since he owned the building. But more than anything, sheer luck. At some point in the late 90s, his bar acquired a reputation as an easy place to get laid and it's had lines a hundred feet out the front door every night of the week ever since. Neither he nor his staff played any role in this. It's random rumors from college students creating a trend and he's the benefactor. Ended up with a cash business that 20 years ago was earning revenue in excess of a million dollars a month.
How is a competitor supposed to enter the market and replicate that? Land is scarce, so you can't just pop up into the same location if there isn't an empty building already. You can't replicate a multi-decade earned reputation except by waiting decades and earning it yourself. You can't magically become the place young people want to be even if you serve exactly the same drinks with roughly identical looking bartenders in a roughly identical looking interior. More succinctly said, you can't copy the intangibles.
From 2008-2017, “we” (I was a contractor 08-09, and employee 15-17) basically turned every spreadsheet into a web application internally.
We added authentication and authorization, used a LOT of ETL-type processing to move data around. So many things could have been packaged and sold, but that was the “secret sauce” that kept the company so profitable.
Then the end in 2017 when a new CIO came in, killed off IT (laid off all non-managers over the course of a year), and replaced everyone with South African consultants to turn IT from a cost center to a profit center.
He lasted another couple years then left.
Our startup noticed our first major-brand customer had a high-stakes B2B process depend on a spreadsheet that looked handmade and perhaps error-prone.
There was a moment of thinking This could be one of those startup pivot points, when what you thought was the incidental to your exciting technology solution, is actually the simple CRUD app/integration SaaS that customers need even more.
Then there was a moment of thinking We don't want to be the communication middleperson on that high-stakes communication, if there was ever a dispute.
Also, either it wasn't a perceived pain point of the customer, or they intended their ongoing CRM-ish conversion to address it eventually.
So we listened to the customer on what problems they (or at least the stakeholder executives' KPIs) actually wanted us to solve, and we continued to consume that spreadsheet in a manual and "zero-impedence" way, for the little bit we could use that info.
(A management consulting firm or YOLO/fly-by-night startup might do that one, though.)
Search funds have existed for a long time now, and it's hard work.
That being said, if I've learned anything during 20 years in various industries ranging from publishing to hedge funds:
1. People have very different brains
e.g. quants can't do sales and vice versa.
2. People don't have infinite time and money
Even billionaires have items on their "want to do" list that don't get done b/c it's either too complex, would require too many people, too much time etc
3. Some people are great at systems building and other are terrible
This could really be downstream of item 1 but it's worth pointing out that building a system of people and processes is it's own specific skill. If you live in an organization where specialization is rewarded, you'll find that there are fewer and fewer people who have this skill.
4. People underestimate what it takes to do <thing>
The classic case is the HN "Oh you want to start a software consulting company b/c you like programming? That's like saying you want to start a pizza parlor b/c you like making pizza. Making the pizza is only 10% of the job."
...and stigma could be overcome for a price
Great news
Spoiler alert from the article: turns out economists are wrong. This will sound inflammatory, but has the field of Economics produced anything of value to society? It seems to be a social "science" based on incorrect generalizations about human psychology and studies that don't replicate. At the end of his long career, Alan Greenspan (their Michael Jordan) basically said "whoops, I was wrong that people would behave rationally."
What does humanity get from the field of economics that justifies the amount of attention and status we give economists?
But I think economists did pretty well: supply and demand, marginalism, opportunity cost, Coase theorem... Not physics-level well but well.
> non-obvious
I'd maybe question this a bit. I suspect that many things we consider obvious are only so because of the field of economics -- but that's admittedly only an initial hunch.
Hence economics is a probabilistic field. The things that occur have a probability behind them. Instead of a policy that needs X to happen the policy can be adapted to "If A then B, if X then Y".
The main problem I see with public economists is Goodharts law: When the predictive model becomes visible to the agents in the economy, there are stable strategies to counteract those models. So if economists want to remain good predictors of real outcomes, they have to collect massive amounts of data to feed the model but then remain mostly invisible afterward, so that the agents being analyzed don't change behavior too much.
In some sense, a known bad model that is used to set central policy may be good because the rest of the economic agents can work on a diversity of better models without correlating as much with each other and the known central model.