Q: How is your IT department like the D.C. Metro?
A: Both of them have huge incentives to put off essential upgrades and improvements until you have a genuine emergency on your hands.
For years I’ve advised publishers to stick to their knitting and outsource as much technology as they can. The main reason for this is that technology requires regular maintenance and upgrades, and (1) that’s not your business, and (2) you probably don’t have the temperament for it.
Running a technology company requires you to allocate about 20 percent of your precious resources to keeping things current. Patches. Upgrades. Integrations with new systems.
Publishing companies simply don’t think that way.
I’ve been in the meetings where this conflict comes out. Marketing wants the new shiny object. Editorial wants a new subscription feature on the website. Sales wants to integrate the lead pipeline with the fancy new tool. Management wants a new product. All these projects are tied to revenue, but the IT director wants to upgrade infrastructure, which is seen as nothing but an expense.
The conclusion in those meetings is often that if you can make do with what you have, do that and focus your development resources on the new stuff. After all, good enough is good enough.
It’s like the DC Metro board. Short-term thinking led to a “run it until it breaks” mentality. They deferred maintenance, which reached a breaking point in 2015. They faced track fires, equipment failures, and, tragically, loss of life. Emergency repairs created a year of chaos in the system.
The Metro didn’t fail because of one bad decision, but because maintenance was always postponed in favor of short-term needs.
Publishing organizations can do the same thing with their own technology.
Technology debt is what accumulates when an organization consistently chooses new projects over maintaining and upgrading the systems it already depends on. Nothing is “wrong” at first. Everything works — until it doesn’t. And when it fails, it fails spectacularly.
What’s a publisher to do?
Here’s some advice to steer you away from this kind of failure.
1. Stay out of the technology business as much as you can.
I don’t mean “don’t use technology.” That would be self-destructive.
I mean use technology that other people have to manage. Let them worry about keeping things up to date.
2. Where you must develop your own technology, put a hard-headed manager in charge.
I remember a meeting where an operations guy had a great idea for reducing friction in one aspect of a publisher’s business. My lead programmer said no way — it would create huge security risks. The operations guy outranked him and tried to force him to do it anyway. My programmer said, “It is my professional judgment that this is reckless and irresponsible.”
You need that guy — not just for security issues, but to keep the wheels on the bus.
3. Budget for maintenance.
Assume that 20 percent of your IT labor will be spent on completely unsexy stuff that doesn’t make you any money.
4. Audit your system for fragility.
I used to tell my kids, “think of what might go wrong and take steps to make sure it doesn’t.”
Ask your head of IT questions like this.
- Which part of our system worries you the most?
- If you left tomorrow, what would no one else understand?
- Where are we relying on workarounds?
- What would you simplify if you had the chance?
5. Recognize the secret enemy
There are lots of missteps you can make with technology, but this one is often overlooked. Short-term thinking.
This is a huge challenge when “long term” means the next quarterly meeting with the board, and you have to show a profit.
That’s when you have to be a leader and make the hard case for stability and sustainability.
Conclusion
Avoiding technology debt is about as exciting as flossing and brushing your teeth. And just as necessary.
Outsource as much as you can, but you’ll inevitably have some internal tech. Make sure you invest in keeping it up to date.
