Two kinds of startups

When you spin inside a certain field long enough, you start to forget that some concepts and terms are not obvious to people around you. That happened to me with the term 'startup'.

It always felt axiomatic to me that startups can be segmented by many different criteria, but probably the most important one is the goal of creating the startup. That is the split I want to describe here.

1. Startups that plan an exit strategy and the cadence of raising investments from the moment they are created – venture startups.

By a strange twist of history, this approach in recent years has started to be called 'startup'. Here the founders build a product that will definitely be interesting to investment funds or large players, and the entire strategy is built around becoming as visible as possible, preparing the strongest pitches, and exiting with a big upside in order to go build the next product.

2. Startups that are born because something started itching in the founders, and they realised this is exactly the area that might be interesting to others. They are not created for an exit or even for raising investment, but to solve a specific pain. This is what I call 'garage bootstrap', because in the Valley it really did once happen in garages, without outside money.

In this approach every cent is counted, because the mantra is not 'the main thing is that we are growing' but 'the main thing is that we are profitable'. The main goal of bootstrap founders is to preserve autonomy, not dilute the meaning of what they are doing, and not report to anyone for their chosen strategy and/or decisions.

Bootstrapping demands huge involvement and responsibility, and it can sometimes feel lonely and sad, because it is rarely fast. Yet this is exactly the type of startup that teaches you how to build a business, not just ride the hype.

Why did 'startup' end up meaning venture, not bootstrap?

The answer is simple: venture startups are a product of an accelerating economy. And from the early 2000s the global economy was accelerating at a crazy pace, where growth of less than 5% a year was barely considered growth at all.

In that kind of market, money is cheap, and after a solid pitch and a crooked little MVP you could raise a few hundred thousand dollars and then solve most problems with cash.

Now the economy has overheated and is slowing down, and you can see this especially clearly in the hypest markets like AI.

What happened to the AI market over the last couple of years?

Very roughly, by 2024–2025 analysts expected a market worth hundreds of billions in annual revenue, and 2–3 trillion dollars by 2030.

Any company that had 'AI' in its name or product description was priced as if its unit economics were perfect and there were simply no problems. Reality turned out differently:

1. The cost of infrastructure for AI turned out to be enormous.

2. Most small and mid-sized startups operate at a loss, using either someone else’s expensive infrastructure or someone else’s models.

3. Because of the first two points, margins turned out to be nothing like SaaS, more like a few percent.

What about bootstrap startups?

As a rule, they do not particularly care what is happening with the macroeconomy, how overheated it is, or how fast it is growing. For them, the key is profitability, which, with minimal expenses and no bloated staff, can come even from a few thousand euros in profit.

At the same time no one is pushing them, in tough economic conditions, to make bad decisions just to speed up the next funding round.

So I am almost 100% sure that over the next few years the term 'startup' will drift back toward the garage bootstrap, and that venture stories will shrink dramatically, because they simply cannot exist without wild global economic growth.