Do things that don't scale, and then don't scale
157 comments
·August 16, 2025kylecazar
Building things for yourself is fun -- I do it. But the original article was written for startup founders, building companies.
salomonk_mur
Building a particular type of company - a tech startup.
I'd argue that we need to apply this to companies too. Stop building startups (which start losing money and may never ever make a profit in the process of scaling) and start creating companies (which make a profit ASAP and may never scale).
timr
"Tech startups" that make enough money to be interesting [1], but do not have the property of being a winner-take-all market, are the truly magical unicorns of software. If you have one, you should covet it and guard it viciously, because otherwise someone with more money is coming to eat your lunch [2][3].
OP wasn't talking about this kind of thing, of course. The phenomenon of hobby projects being "startups" has always been a weird fit.
[1] I define this at least as "supports more than one person's salary in perpetuity," but would probably add some things like margin requirements if I thought about it more. Obviously you can exist as a consultant charging by the hour, and that's not what either of us is talking about.
[2] The only examples of this I've personally ever seen have been dominant players in niche markets with high bars to entry. So they weren't hyper-growth unicorns, but weren't really your definition of "companies", either. You don't get to these kinds of businesses by aiming to be one - you get there by trying for something bigger and topping out the market.
[3] Consider the phenomenon of the small vet clinic, which is rapidly ceasing to exist. Private equity has been steadily hoovering up these relatively high-margin, low-regulation businesses for the same economic reasons that software companies get big. If you try to create a vet clinic, you will be assimilated or crushed by a bigger player with a brand and the ability to undercut your prices.
dghlsakjg
Is the vet clinic thing a lasting phenomenon? Are independent vet clinics not economically viable or is there just a VC/PE market imbalance where capital is overbidding for a business that doesn't really have a great moat. I would be perfectly willing to believe that if you knock on the door of a perfectly viable business and offer them more than it's worth as well as a salaried position at the new business, (almost) no one says no.
I think the cash merry go round will stop eventually because the moat, in as much as there is one, is the vet themselves, not the business. If my vet leaves and opens a new practice, I will follow the person. Vet services don't really scale either, you can hit efficiencies with admin, but I'm sure a software company is willing to eat the PE firm's lunch there. With software, costs are heavily weighted to the fixed side. With vets, seeing twice the number of dogs takes twice the number of vet hours, vet tech hours, etc.
shooker435
There's probably a middle ground here. Making a profit ASAP is only really possible if you're selling expertise, such as consulting. I really like the emerging term, "seed strapping" which aims to address this conundrum in the startup world: raise just enough to build the first iteration, then act like a bootstrapped business.
derwiki
Yep, totally — the OG advice was founder-focused. I just couldn’t resist twisting it a bit, because the line itself is too good not to repurpose.
netcan
Yeah... i think that's pretty clear right from your title.
j45
It's also possible to build things to solve your own problems, that might be problems other people have too, which they also try to solve but fail at, which are problems that are painful enough that they would spend on.
They're harder to find in B2C than B2B. Individual problems can sometimes map to B2B.
chrisweekly
"Startup" is often defined as "small company designed to grow quickly".
zahlman
> There’s an old startup mantra — probably from the early Airbnb or Y Combinator days — that goes: “Do things that don’t scale.”
Yes, it was posted again just the other day: https://news.ycombinator.com/item?id=44913359
derwiki
Haha funny timing! I wrote this post last weekend and haven’t checked HN much this week
quuxplusone
I agree it would be nice to put the source of the quote in your own post. It's so easy to Google before quoting.
On another note: did you just tell a bunch of strangers on the internet how to snail-mail arbitrary photos to your mom? Or does that email thing only activate if the email comes from your own address? (Guess you'll find out! :))
derwiki
The postcard app is gated by a verification queue. FWIW, so far no one has tried to sign up. Also there’s an address book so it’s not just for mom :-)
tlhunter
It's definitely the first result on Google and sort of ruins any credibility of the article.
mekoka
Anyone who was part of the HN/YCombinator scene at the time PG wrote that piece (and many years later) would know what the title alludes to. Why would it ruin this article's credibility?
loumf
Because when you quote someone, you should credit them. It's not believable that they don't know the source of the quote because it's trivial to find.
pimlottc
I agree that not everything has to go huge. I don’t see how ChatGPT has anything to do with this though.
bravesoul2
It's like we have a 3D printer where before there was only injection molding.
topaz0
This was my first thought as well -- even some of the examples in the post are pre-llm.
j3th9n
Than you probably never used ChatGPT for coding.
aetherson
I use Claude Code to hack together a little webapp that allows me to make hex-maps for use in roleplaying games.
There are a lot of sites on the web that let you make a hex map. A lot of them are even free. Many of them have features that my little webapp doesn't have.
But mine works the way I want it. I wanted rivers and forest to be modifiers on top of the base terrain of the tile. I wanted to have a few different settlement icons. I wanted to have more variations of hills and mountains than most of the other sites do.
And if I'm ever missing a feature on this thing, I can just add it, rather than just sort of saying, "Oh well."
Because it's an app that's just for me, I don't need to worry about scale or security or monetization or anything.
It took me about an hour to two hours of my attention (spread out over two calendar days) to have the AI code it.
lionkor
Is it easy to modify, or do you think it will eventually become unmaintainable? What level of experience/software development seniority do you have?
TomWhitwell
An app can be a home‑cooked meal, by Robin Sloan: https://www.robinsloan.com/notes/home-cooked-app/
jpgleeson
Robin Sloan wrote about something similar a few years back. I think this is one of the most positive things to come out of the current cycle - even if things turn out less revolutionary than sold, enabling regular people to produce something small that sparks joy without having to know a language
pcald
Reminds me of Maciej Ceglowski’s seminal “Barely Succeed: It’s Easier”
nickserv
I feel like the main takeaway here is that if you want your personal project to last, you should run it on your own server.
sdenton4
It'll last until you lose interest or die. Shrug emoji.
j45
Also making it easy to host and maintain. The more custom or esoteric libraries, the better.
If you do build it with boring tech in a simple way, you can just rely on it.
tomasphan
This is absolutely true and the reason I left a job in software consulting. The future is asking an LLM to write anything for you. It will figure out the tech stack, hosting, integrations etc. “I am looking for an alternative to Discord” is now “make me a Discord clone for my friends and I”.
Quality of the code aside because now it doesn’t have to support millions of users anymore.
101008
I disagree. This existed way before LLM. Open source alternatives to most products are already available. And install them and deploy them is much easier than do it with LLMs, and you get updates, etc.
People don't want the responsability to keep them updated, secured, deployed, etc. Paying a small amount will always be more convenient than to maintain it yourself. The issue was never coding it.
iancmceachern
Check out the book "Small Giants, companies that choose to be great not big"
ChrisMarshallNY
We have published an iOS app that is a utility for a relatively small demographic. It's been on the App Store since January, and we have a bit over 1,000 users. I don't really expect it to get more than a couple of thousand.
I did test it with 12,000 users (fake ones), so it should handle small scales, but it will definitely have to be rewritten, for much larger scales. It would not be as usable, in that case. At this scale, it works very well, indeed.
That's fine. It works great, and we vet every signup, so we're not interested at all in scaling.
Genuinely love this. I've sort of done this by hand before the advent of good coding agents [1]. But now, it is even more enjoyavle, as development time is even less an issue.
I'd love to see more people realize this and use that new power to build things that don't necessarily scale on their own, but might trigger changes for sizeable groups, either socially, or politically.
[1] https://mariozechner.at/posts/2024-07-15-two-years-in-review...