Doing a startup? Don’t learn to code.

I was recently chatting with a friend of mine who was learning to code. A psychology researcher at a major university, he wanted to create an app that would make it easier for researchers to collect patient data.

Ultimately, he wanted to turn it into a software product and sell it to researchers at other universities. However, since he didn’t know any programmers and couldn’t afford to hire someone to build it out, he thought his next best move was to learn HTML and Javascript.

This is a fundamental mistake I see non-technical entrepreneurs make when getting into the startup world, and it’s understandable.

Everywhere you look nowadays, people are talking about how “software is the future,” “programming is like literacy,” and “everyone should learn to code.” And, while programming is a useful skill, there are higher ROI things you can be doing as a non-technical founder.

If you’re trying to start a company, you shouldn’t start by learning how to code. You should start by learning product design.

When you’re exploring a startup idea, there are two main types of risk: technical risk and design risk. You can think of the distinction like this:

• Technical risk — can this be built?
• Design risk — if we build it, will anyone really want it?

Most product ideas are not that technically complex. This means that most of the risk is design risk, and that if you can lower the design risk, it’ll relatively simple to build the product you’ve designed.

How does one lower design risk?

The main way to do this is talking to potential customers to get feedback on your ideas. Luckily, there is an entire field of study around this process, called “product design” or “design thinking.”

My two recommended starting points for product design are this awesome Coursera course from UPenn and this list of resources from Stanford’s design school.

Now, you might think you need to actually build something to get feedback on your idea, but by using prototyping software, like InVision, you can get feedback on a high-fidelity prototype without having to write any code. These prototyping tools are pretty robust, and the learning curve is much shorter than learning web development from scratch.

Once you’ve gotten your product in front of potential customers, you’ll either have learned that people want it, or that you need to go back to the drawing board and incorporate some of the feedback you received.

If you need to do another iteration, that’s great — you just saved yourself a bunch of time by figuring that out now instead of after you’ve built the whole product.

If, on the other hand, people clearly want what you’re building, that’s also great. Having talked to users and verified your product assumptions will have put you in the 99th percentile of people looking for technical cofounders.

If you have people who want to give you money, or even better, have pre-ordered your product, that’s even better.

At this point, you can try and find someone technical who is looking to get involved with a startup or a project outside of work. Chances are, once you’ve found someone, they’ll be able to implement in under a month what would have taken you significantly longer to build.

Since you’re not spending your time learning to code, you can continue to iterate on the design, or you can start reaching out to other potential customers so that way you’ll have revenue from day 1 once you launch the product.

At the same time, all the above doesn’t mean you shouldn’t necessarily learn to code.

If you’re looking to transition into a career in software development, for example, then by all means, learn to code. Software developers are well-paid, tend to have good work / life balance, and get to do creative & engaging work, so if you’re looking for a new vocation, software is pretty sweet.

If you’re starting a company, you still might want to learn a bit about how the underlying technology works.

Also, there are probably code-related things you can learn to help automate tasks in your daily life, for example learning how to do web-scraping, more complicated mail-merges, or data manipulation. Often, the best place to start is by automating tasks you do regularly.

Overall, though, you should focus on how code can make you more effective at the job you have or want.

And, if you’re doing a startup, focus on the area where you’ve got the most risk: design!