Every now and then, a software update goes wrong, as we saw with the Crowstrike update that took down a good portion of global IT infrastructure. Your own software issue could be something small, like a feature moving to a new spot and confusing users. Or it might be bigger, like login failures, broken forms, or crashes that spread across your customer base before you even realise what’s happened, but trust us, your consumer base will let you know.
It’s important to note that no dev team on Earth is going to want this outcome, but it still happens. So the question becomes less about how to avoid every issue and more about how to respond when something difficult occurs.
Let’s explore that below:
Slow Down Before You Rush To Fix Everything
The first instinct is usually to get in and patch the problem as fast as possible. That’s understandable, of course, and yes, some tempo is needed here. Yet rushing often makes things worse, especially if you’re not fully sure where the fault sits. Like in real life, a problem that may require urgency but not all-out panic can still allow you to take a pause, a breath, and figure out what exactly changed.
Is it a front-end display issue only working on some devices? Did your API routes shift without warning? Did you avoid the advice to go with Laravel cloud hosting and now your PHP environment isn’t working as well as it could? If you can, be sure to open pathways for your users to report specific bugs and device combinations to give you a better idea.
Keep Talking, Even If There’s Nothing New To Say
Customers get more frustrated when they feel ignored than when something actually breaks, because while the latter is forgivable, the former isn’t. You don’t need to have all the answers right away, but people do want to know that you’re working on it with updates.
A quick message that says you’re aware of the issue, you’re looking into it, and you’ll post again in an hour is sometimes all that’s needed. You can also update when you’ve isolated the issues and are deploying fixes. If you handle this well, some may trust you more as opposed to less.
Have A Rollback Plan That Provides Flexibility
If you don’t already have a way to roll back updates safely, it’s worth making that a priority. Assuming you’re not reading this post trying to figure what to do in a current emergency, this is a good thing to put in place ASAP.
For instance, that may require hosting on a platform that lets you stage changes without pushing live immediately. If you’re using a framework as discussed above, for example, pairing it with cloud hosting can make it easier to preview updates in a safe environment and avoid downtime. It also gives you better options for version control and isolated testing if you need to figure out what went wrong.
With this advice, you’re sure to respond with confidence even after a tough software update catches you unaware.
Discover more from The Cameron Journal
Subscribe to get the latest posts sent to your email.
