I was talking to an old friend who was leaving tech journalism behind for a startup tech firm. Yay him! So, I asked what he’d be doing there first.
“Building a custom content management system (CMS),” he told me.
I then blurted out the first question that sprang to mind: “Why!?”
I mean, seriously, do you how many CMSes there are? By my count, there are well over 100. They come in all varieties: open-source, proprietary, and Software-as-a-Service (SaaS). And those are just the ones that are publicly available. If you count those that are custom-built, there are probably at least as many more.
This is insane. There is no reason on Earth for anyone to ever make another CMS from scratch. Sure, occasionally, there’s a real change, like the rise of headless CMSes, such as Strapi.
Before you ask, a headless CMS doesn’t come with pre-built web interface themes or the tools to make them. Instead, unlike conventional CMSs, they don’t worry about your website’s front-end. Instead, they focus on the back-end content repository, which is used for storing and delivering structured content. This content is then made available for display via a RESTful API, typically using JSON or XML.
Got it? Good.
Now, maybe some companies could put Strapi to good use. (I should mention there are many others, including Ghost, Contentful, and Prismic.) So, you don’t need to build your own headless CMS. In fact, for most of us, old-school CMSes, such as WordPress, Joomla!, and Drupal, do just fine. I mean, heck, WordPress runs 34% of all websites in the world.
In other words, if you’re building or maintaining your own homebrew CMS, you’re wasting valuable time and money.
It’s not just CMSes, of course. There are many such examples. I’ve seen more bad homemade vertical programs for real estate, small banks, and retailers than I ever want to see. I mean I do think the lone custom real estate program I saw running on an Amiga 1000 from 1985 was cool. But let’s get real: sooner or later, that 36-year-old hardware is going to die.
Now, if your business involves making vertical software, good for you. Keep going. There will also always be room for value-added resellers (VARs) or software integrators. But if your business is putting words and images in front of people, managing property, or selling geegaws, you don’t need to be a software developer, too.
Entrepreneur Marc Andreessen was right when he explained in Why Software Is Eating the World that big companies had to become software businesses. But I’m not writing to Fortune 500 CEOs, I’m writing to small and mid-sized business owners. You will not be hiring dozens of programmers to perfect your company’s software. Even if you did, you’re unlikely to be able to show any improvement to your bottom line by developing your own programs.
Instead, you should look for best-of-breed programs that fit your needs. If you can’t do that, talk to VARs, system integrators, or managed service providers (MSP). Finding good hardware and programs—and making them work for you—is their job, not yours.
Don’t think I’m talking only to those of you who aren’t in the tech business. Tech firms make the same kind of mistakes.
For example, there are hundreds of Linux distros out there. Of those, no more than a dozen are commercially viable. I mean, if you want to build your own Linux, go for it. It’s fun. (Really, I’ve done it myself.) For some of us, it is fun. But there’s a big gap between building software for fun and building a viable business.
Simply inventing a better mousetrap is great, but if you can’t market, ship, support, and sell that mousetrap, you don’t have a business. You don’t even have a business plan. You need a great idea, execution, and all those other things I’ve mentioned.
So, again, and with feeling, stick with your own expertise, and let the real software pros help you with your company. They’ll be happier, you’ll be happier, and—oh yes—you’ll have a greater chance of being profitable as well.
