Build Versus Buy: You Are Always Building, Always Buying

TL;DR:

  • “Build-vs-Buy” is a false dichotomy, you are always building, always buying
  • “This is not your ‘Core Competency’” is a faux trump-card, everything is a spectrum
  • Past experience is overrated, what matters for hiring is your ability to learn

Understanding The Trade-Off

Fully understanding the build-versus-buy trade-off requires a bunch of different context to really get right. At first glance you might think that you only need to really understand:

  • The problem you are trying to solve
  • How the thing you are looking to buy solves the problem

But in reality it is much more nuanced than that. Most of the reading out there about the build-versus-buy trade-off is written with “enterprise” software in mind, and evaluate things on a very first-order level. Concerns like:

  • How well can this software integrate with other things?
  • How much does this software cost?
  • How much time will it take to integrate the software and make it meet your organizations’ needs?

These are all important questions to ask, but there are higher-order concerns that take some time to fully appreciate like:

  • In the long run, can your developers build a solution that provides more long-term value to your organization? (even if in the short-term it costs more and takes longer to build?)
  • How will your (in-house or external) users “feel” when they notice that your solution is based on an off-the-shelf (bought) product?
  • Can you account for the in-house skills and expertise that would be simultaneously developed (and required) if the solution was developed in-house?
  • Will there be any developer or user attrition if you built versus if you bought?

These are not as easy to answer, but the answers to them may be even more significant than the $$$ figures your vendor may come up with.

I’ve never been really satisfied with the “build-versus-buy” trade-off in general, mostly because throughout my entire career, it has never been a dichotomy: I was always building, and always buying.

Understanding the False Dichotomy

The false-dichotomy comes from the fact that everything “bought” requires integration, and it is building of that integration that actually builds the value to your organization.

At the same time, we are always buying. Even when it seems like we are choosing the “Build” option, are are still buying, it just the we are buying libraries, frameworks, and other components to build the solution. Things are not built in a vacuum.

It is all building, and all buying. There is no such thing as a turn-key solution, it is just a spectrum between simpler and more complex products.

It is in the building that you provide value to your organization and to your users. It is in the buying that you “stand on the shoulders of giants” and become a more powerful engineer by building on top of existing engineering.

“Core Competency”: The Ultimate Faux Trump-Card

Sometimes it is “simple” to describe some build-versus-buy decisions by framing them in terms of what the organization’s “core competency” is. Sometimes it is so simple as “if it is within our core competency, then we build, otherwise we buy”.

What does “core competency” really mean? Before software was eating the world, “core competency” could have meant something very different. But now software is seen as a universal multiplier for every business, regardless of what they do. What does software have to do with Nike’s core competency? Are computer servers Amazon’s core competency? What would the core competency be of any retail organization? How about a game design company? What about a high-frequency trading firm?

The answer is blurry in this world, because software in so fundamental to the world we live in. Software is a multiplier for all levels of any business, from HR to Product to Manufacturing to Retail, it knows no bounds and powers all the things. Discarding an opportunity to build something that enhances your business is a mistake.

Case Study: Game Design and Looking at Game Engines

Game design, like any other software engineering pursuit, also involves many build-versus-buy decisions. From assets, to engines, to voice acting, to motion capture, there are many many micro build-versus-buy decisions to make when developing a game. In some sense, every one of these decisions fits into their “core competency”. In another sense, building your own game engine for example, sounds like a decision that 90% of game studios should choose “buy” over build. Is developing your own game engine in-house included in their “Core Competency”? Who knows.

The real question is: will buying an existing Game Engine provide more value to your game than what your team could bring to the table, or would it hold your game back from what you are trying to achieve?

If the year is 1993 and you are building Doom, then no, buying an engine will not help you achieve your goals. Building an engine will help you build what would make your game unique.

If the year is 2017 and you are building Fortnite, then yes, buying an engine will help you build your game faster by focusing on what makes it unique.

This doesn’t imply that we should not still be building new game engines in 2020. Just that fewer people should.

On Personal Engineering Capital

As an engineer, how should one look at the build-vs-buy trade-off in terms of building personal engineering capital? That is, how does career advancement come into play when deciding when to build versus when to buy?

I see this as a trick question. With two years of building on a codebase written in an obscure language, there is still room to gain new skills, and level-up the state of the environment you are in, leaving it better than when you got there.

On the other hand, does buying an existing software package mean you won’t be creating any “personal engineering capital” for yourself?

What does this really mean? Is this just a fancy way of saying “I want to work on something that will look good on my resume.“? Or is it even simpler than that: “I want to work on something that is different (or in a different language) than what the company needs.“?

Although this comes from a position of privilege, I don’t think that it is my current employer’s job to set me up for my next job. Not only that, I see the both the opportunity to buy and the opportunity to build both as ways to better my skills and be a useful engineer.

Buying an off-the-shelf container orchestration system allows me to build amazing things on top of it. Things that would be impossible to build without help from such a product. Likewise, building a company-specific ETL platform allows me to understand a problem more deeply, but I would still need to “buy” other software to make it work.

I’m personally not in it for the glamour, I’m in it for the problem-solving and the tech.

I can also say, coming from a person who interviews a lot of candidates for my own job, past experience has much less of an impact as you might think. In a world where Javascript frameworks come in and out of vogue over the course of months, your number one hireable quality is the ability to learn new things.

Conclusion

I don’t think the key to a fulfilling software engineering career involves building everything out of Rust or whatever is the hottest language out there right now. Likewise, if you feel like your job is “just” gluing pre-built things together, well guess what, the value comes from that glue, and the thing you are building is an even bigger platform on top.