Legacy, technology, and band-aids

Legacy in the context of technology most often has negative connotations. A “legacy system” usually means old, deprecated¹, outdated, etc. “Legacy code” is similar: it’s old. It might work, but it isn’t in line with best practices anymore. But what if we consider legacy in the traditional sense, like we’d do for estate planning, trusts, inheritance, and the like? How can we consider the websites and technology we use in a way that promotes legacy—that considers the future and not just the now? And how do I maintain a powerful vision of the future while not neglecting the day-to-day? A lot of that comes down to a mindset: choosing platforms and partners with the long game in mind, doing it right the first time when you can, and knowing when a shortcut is deliberate versus a band-aid that will haunt you. Tech changes. We can’t set and forget. But we can build and decide in ways that leave us in a position to change when we need to, instead of being stuck. One of the clearest places that mindset shows up is how we respond when something goes wrong. Let’s look at some practical ideas with examples to inform our decision making as it relates to technology.

The goal

First and foremost, what is the goal…the end state? Is my website just a temporary landing page for a political candidate during election season? Am I starting a new business? Am I in charge of operating an 8-figure eCommerce² store? In other words, how should my website be defined as a success? By what metrics? Whose standards? Is the main goal just the bottom line? From an investor or shareholder perspective, perhaps. But what about the less-obvious ones, like these:

• The website is easy to maintain.
• The website is not brittle, i.e., not a house of cards where even one structural change could spell doom.
• The website is clearly set up so that a change of hands is effortless and intuitive.
• The website is approachable to non-techie members of the team. They can adjust content without fear of breaking the site or ruining things.
• The website amplifies the business’s message instead of being a bottleneck.

My point here is that it’s not enough to judge a site by how it’s performing today. How valuable will it be to the business tomorrow? Next week? Next year? Simply put, how future-proof is your website? That’s why it matters how we respond when something goes wrong. Patch the symptom or fix the cause? It’s one of the main ways we either build toward that future or undermine it.

When a band-aid fix matters

In most cases, a short-term fix is the wrong choice. It yields immediate but temporary rewards, and often at the expense of long term stability, productivity, and value. But sometimes, a band-aid fix is truly called for. When a child skins a knee, we clean the wound and cover it with a band-aid to prevent it from being infected and irritated. Root cause analysis of the injury doesn’t help protect the open wound right now. In the same way, certain situations call for band-aid fixes. Here are some examples:

• Your eCommerce website checkout flow is broken. This is bad..really bad. You’ve determined the source is a coupon plugin. The band-aid fix is to immediately turn off this coupon plugin. This doesn’t fix the real issue (the coupon plugin is interacting unfavorably with another part of your eCommerce website), but it immediately allows your customers to transact with you.
• A new website feature is creating a potential security vulnerability, as you are noticing an increase in fraudulent access attempts. The band-aid fix is to immediately shut down that feature to keep the bad actors out. It doesn’t solve the “why” but does protect you until you can figure out the real issue.
• You notice that the pricing for a product or service on your site shows up in two separate places as two different prices! The band-aid fix is to manually fix the discrepancy. Now, this violates the principle of “Don’t repeat yourself,” also known as “DRY.” But in this moment, it is more important to violate that principle so your site isn’t at odds with itself. You can determine the cause of the price deviation later and set things up so you have a single source of truth³ for your pricing.

When a band-aid fix is a thorn in your side

In general, band-aid fixes are a measure of last resort. If you can find the root of the issue and correct it there, you have set yourself up long-term. This is future-proof troubleshooting. You’re getting past the symptom and looking for the cause. But for many, it has become habit to default to short-term fixes. This is preposterous. It is akin to running a marathon, realizing your shoes are untied, but then stating you don’t have time to stop running to tie your shoes. Or there’s the axe analogy: “I don’t have time to stop chopping the tree to sharpen the axe!” Here are reasons to avoid band-aid fixes:

• Band-aid fixes can cause regression errors. This is a fancy way of saying that quick, surface-level fixes are akin to whackamole. Sure…that one issue is gone, but we just broke the website in three other places.
• Band-aid fixes mask the real issue. If we remove the dashboard lightbulb because we’re tired of the check engine light being on…what good can come of that?! Ignorance is not bliss.
• Band-aid fixes are not future-proof; they’re not sustainable. Do we value long-term success in our organization? Do we want our business to succeed? Do we care about our coworkers or the folks who will have to deal with this same issue weeks or months later because we merely patched what should’ve been replaced or overhauled?

Technology won’t stay current forever. What we can do is build and fix in ways that keep it maintainable, so when the next change comes (a new platform, a big update), we’re not buried in band-aids and shortcuts. Band-aids have their place, but as a deliberate exception, not the default. That’s how you stay in a position to change when you need to.

Footnotes

1. Deprecated means no longer recommended or supported; it’s being phased out in favor of something else. ↩

2. eCommerce means selling (or buying) online; an eCommerce store is your online shop. ↩

3. A single source of truth means one place that defines the data (e.g., a product database or settings page); everything else uses that so you don’t have conflicting versions in different spots. ↩