Picks-and-Shovels

Picks-and-Shovels

Picks-and-Shovels refers to an investment strategy in which one does not invest in the winners of a boom, but in the companies that supply the tools to everyone involved. The name comes from the California Gold Rush: the ones who reliably got rich there were the traders selling picks and shovels.

During the California Gold Rush around 1850, tens of thousands of people set out to find gold. Most of them found nothing and went broke. What reliably made money, however, were the stores that sold shovels, picks, pants, and provisions. These merchants didn’t need to know who would find gold — they earned from everyone who tried. This exact pattern is what the term Picks-and-Shovels refers to, roughly meaning picks and shovels. It describes the decision not to put one’s money into the seekers, but into the outfitters.

Why investors prefer to bet on the outfitters

In every major tech boom, one thing is fairly certain and one thing is very uncertain. What’s certain is that a lot of money will be spent. What’s uncertain is which single company will end up winning. Anyone who bet on internet companies in 1999 had a good chance of picking one of the many that disappeared shortly after. Anyone who bet on networking technology and servers at least earned money for as long as building continued.

The strategy thus shifts the risk from the question “who will win?” to the question “will there be any investment at all?”. That is a considerably easier question. An outfitter also often has many customers at the same time. If one falls away, the others remain. A single gold seeker has no such safety net.

However, the approach has a clear catch worth knowing. When the boom ends, demand for tools collapses too — and often quite abruptly. Outfitters rarely sell to end customers, but to other companies. Their investment budgets can be slashed within a few months. Picks-and-Shovels is therefore safer than betting on a single winner, but by no means risk-free.

Who is selling the shovels in the AI boom

To apply the strategy, one breaks a market down into layers. At the very top are the visible applications, such as chatbots or image generators. Below that are the companies that build the actual AI models. And below that is everything these models need in order to run. Picks-and-Shovels means: you buy in the lower layers.

In artificial intelligence, the bottom layer is unusually well visible. An AI model needs special computing chips, so-called GPUs, which carry out very many calculations simultaneously. It needs data centers, meaning huge halls full of such chips. It needs electricity, cooling, and fast network cables. Each of these ingredients is supplied by companies that don’t need to understand anything about which chatbot ends up being popular.

The chain extends further than one might initially think. Behind the chipmaker stands the factory that physically manufactures the chip. Behind the factory stand the makers of the machines used to work there. Every stage is a possible Picks-and-Shovels position. The deeper you go, the less competition there often is — some of these machines are built by only a single company worldwide.

The term in market reports and analyses

In financial news, the phrase usually appears when an analyst recommends a stock that isn’t the obvious choice. Sentences like “a classic picks-and-shovels investment in the AI space” then mean: this company benefits from the trend without being in the spotlight itself. Funds also use the term to justify their selections.

One also encounters this pattern far beyond AI. In electric vehicles, the outfitters are the battery manufacturers and lithium mines. In cryptocurrencies, they were the chipmakers and the trading platforms that earned money on every purchase. Even in streaming services, one could argue that providers of server capacity earn more stably than the streaming companies themselves.

A common misconception is that Picks-and-Shovels is automatically the cautious choice. That’s only partly true. Popular outfitter stocks are often already very expensively valued during a boom, because everyone has the same idea. The share price then already reflects the expectation of years of rising demand. If that demand fails to materialize, the price falls just as hard as with any other bet.

Subscribe free. Unsubscribe the second it sucks.

High-signal news across AI, business, UX, and tech. Every morning.