Why is it better to be a seller of picks and shovels vs being a gold miner
Infrastructure tools. Application layer. Relevance during AI gold rush.
Hey reader, welcome to the 💥 free edition 💥 of my weekly newsletter. I write about building startups and careers in AI. You can learn more about me here. Feel free to send me your questions and I’m happy to offer my thoughts. Subscribe to this newsletter to receive it in your inbox every week.
In one my previous posts, I had written about how VCs do diligence on AI startups. In that post, one of the sections was about how it's better to be a seller of picks and shovels vs being a gold miner. People sent me questions on why specifically is that path better.
There are a few points I'd like to highlight, so I thought I should expand it into a full post. In the AI gold rush, here are 5 reasons why being a seller of picks and shovels is better than being a gold miner:
1. Everyone needs infrastructure
People who're looking for gold may or may not get it. But every single one of them will need picks and shovels to look for gold.
Selling picks and shovels refers to selling infrastructure tools to people who're building AI applications.
That's why during a gold rush, it's better to be a seller of picks and shovels. It's a good business to be in.
Example: DataRobot provides AI infrastructure tools to data scientists. They use DataRobot to build and test a variety of models. Those models may or may not work, but DataRobot will keep on going.
Here are a few analogies across other sectors:
Cloud hosting infrastructure: Customers may love or hate a SaaS product, but AWS always gets paid.
Food delivery infrastructure: Customers may love or hate a restaurant, but DoorDash always gets paid.
Telecom infrastructure: Customers may call and yell at a service provider, but AT&T always gets paid.
Payments infrastructure: Customers may buy their groceries at a different store, but Visa always gets paid when they swipe their credit cards.
2. There's an AI gold rush happening right now
Selling picks and shovels is only relevant during a gold rush. If there’s no gold rush, nobody will need them. If you’re providing infrastructure tools to build Blackberry applications, you won’t be able to survive because nobody is building those applications. You need to make sure there's an actual gold rush happening.
There's an AI gold rush happening right now where people are building AI applications across every sector. If you provide infrastructure tools to them, you'll have a higher chance of success.
Example: Many companies are collecting all types of data. They need to label this data to build models on top of it. Companies like Scale AI that provide the labeling infrastructure are well positioned to succeed here.
3. You'll have downside protection
An application may be great today, but it can go down in popularity tomorrow. It happens all the time with video games. It could be glamorous if they surge in popularity.
The picks and shovels business might be less glamorous than mining gold. But you have a higher chance of winning.
Here are a few examples:
Games may come and go, but the Unity game engine keeps on going. So does Roblox. So does Unreal.
SaaS products may come and go, but AWS keeps on going. It's an outrageously successful picks-and-shovels business. If Amazon decides to spin off AWS as a separate business, it could be a trillion dollar company on its own. And it will still grow from there. Azure and GCP are building their cloud services as well.
An influencer may be popular today and they may not be so popular tomorrow. Their earnings might go down. But Instagram and TikTok keep on going since they provide the infrastructure on top of which influencers build their audience.
4. Opportunity to become a toll booth
Apple built the iOS app store to let developers build apps for mobile users. Google did the same with Play Store for Android. They get a percentage fee of all revenues generated by all developers. Few apps on these platforms become successful and many apps will perish due to lack of customers. But Apple and Google always gets paid. They have become a toll booth for mobile apps.
Toll booths own the distribution mechanism through which the user accesses an application. The app developer pays a toll every time the user uses it.
Apple and Google built the infrastructure to build mobile apps. The app developers are the gold miners. Some of them will definitely strike gold. If nobody does, then the app developers will go build on some other platform. It's important for Apple and Google that people strike gold every now and then.
To make sure it happens, they provide services and expertise to their top developers. They meet with them and make sure they gain users. They provide tools to help them increase their reach and provide data analytics to help them get better at their craft. They create communities to help them learn from each other. This will help create loyalty among the base of app developers.
5. You're insulated from trends and fads
By providing the infrastructure layer, you don't have to worry about the latest trends and fads. Google and Meta have built enormous businesses by building ad marketplaces. They connect people who have products to people who need that product. Every company buys ads in some shape or form. A company could be hot one year and they could go down next year. But Google and Meta will always get paid for ads.
If you’re insulated from trends and fads, your revenues will grow more consistently. And the growth compounds.
This is applicable to AI applications as well. The AI field is growing at a very rapid pace. No matter what application you build, someone might build something cooler tomorrow. But both of them will need infrastructure to build their applications.
Where to go from here
Building infrastructure tools is not the only way to succeed. Many successful companies started off by building great applications. But they eventually built an infrastructure product to enable others to build on top of them. Here are a few companies that did it: Amazon, Google, Apple, Microsoft, Salesforce, Snowflake, Twilio, and many more. These companies have enabled other companies to build entire businesses on top of their infrastructure.