I have been noticing a pattern among many of my large clients recently. They’ve customized themselves out of business. They’ve spent so much time, money, and energy customizing their marketing technology that they have painted themselves into a corner, with no easy way out. I’m not saying you should never customize technology, but you should be thinking very clearly about the kind of customization you are implementing, as well as what its true cost is.
If you work for a small company, I’ll wait here until you stop laughing. I understand that except for configuring settings, customization is beyond the budget of any small business. But medium and larger businesses customize technology all the time–often in ways that the vendors of the technology could never have dreamed of. And, in my experience, they often do so without understanding the true costs of their changes.
So, first, I am not talking about configurations. I am not talking about tweaks to the software that the vendor expects you to make. I am talking about actually reprogramming the software. I’ve seen so many of my clients find some objectionable piece of their analytics system, or their marketing automation system, or their content management system–and they immediately “fix” it.
When they decide to fix it, they ask their IT team how much it will cost to develop the workaround that lets them do things the way they want, instead of the way the software wants. There are a number of problems here:
- You might be making things worse. I know that you think your existing process is essential to proper functioning of your business, but that is honestly a rare case. What I see far too often is that you are changing a perfectly good, well thought-out process that follows industry best practices into your own custom process that is none of those things. Your process hasn’t been thought through–it has just evolved. It’s just the way you do things. You are usually better off figuring out how to do it the standard way that the software works, rather than twisting the software into a pretzel to fit your preconceived notions. Think about the cost of the system not working as well as it can.
- Your change probably leads to many more changes. Often, I am told that this change is critical to work with some other existing system. But performing an unnatural act with the vendor software to keep you from having to modify your legacy software is usually a fool’s bargain. You’ll likely be changing more areas that you first believe to really make everything work. In the long run, changing your custom software is usually a lot cheaper. Think about the cost of the ripple effect of changes stemming from this first one.
- The cost to implement is the tip of the iceberg. We focus on what it costs to implement the workaround, but that is often the least of your problems. As the IT budget gets cut each year, it’s harder and harder to justify the constant maintenance of the customized code–this is often referred to as technical debt that you must pay interest on every day. Worse, I’ve seen clients that can’t even afford to move to the latest release of the vendors software, because it is so excruciating, slow, and expensive to carry the customizations into the new version. Think about the costs of maintenance and of being on old releases.
There are situations where you seemingly have no choice. You must customize. In those situations, I would recommend that such customization is a last resort and should have an expiration date–you need to have a plan for migrating back to the standard mode of operation. That plan must be executed in months, not years, so that you pay off the technical debt and avoid the ripple effect of more changes.
