If you’re just getting started as a web designer (freelance or an agency) this one’s for you.

Here are three things I’ve learned over the years while running a web design agency. Some were harder lessons than others. And all of them, in hindsight, are so obvious. Not so much in the moment though.

Never deploy on Fridays or the end of the day

Remember Murphy’s Law. ”Anything that can go wrong will go wrong.” I get it. You have deadlines. You made promises. It doesn’t matter – never deploy on a Friday or at the end of the work day. It doesn’t matter how good you are or how perfect your code base is. Sometimes problems arise and it’s totally unrelated to you (and sometimes it is).

If it’s 3:00 just wait. Take the heat on the delay if you have to and explain why everyone will be better off. Then do a better job at project management next time.

• Problems may go unnoticed until the next day
• If a problem does get noticed – you’ll be working late
• Oh, and your employees will have to stay late too…good luck with that
• Nobody wins

Do releases in small increments

When web designers are working on a large project there’s a tendency to do one huge launch. I get it. While it may seem more efficient from a project management view, it actually becomes counter productive. Here’s why:

1. Bugs. We hate bugs. You change one thing here, and it messes with something way over there. Releases in small increments lets you identify issues before they’re out of control. It can save you some serious time in QA and in development.
2. Feedback and Communication. Sure, everyone agreed on mockups and the UI/UX design. And yep, people sometimes change their minds. Small releases make those changes much more manageable on every front and keeps the project moving forward. 
3. Wildcards. Servers, databases, compliance, and the one off’s that you never saw coming. You have to plan for wildcards. Releasing in small increments makes managing wildcards much easier because the project has still progressed – which makes everyone happy.

Get feedback from the wild

Feedback can be tough. But brainstorming and QA testing only goes so far (before it gets really, really expensive). Do your best and then get feedback from live users in the wild. And if you can do that in a controlled live environment before a global launch even better. Seriously – some people still use Explorer. On Windows XP. The world is a wild place and people do (and think) things you’d never expect. So get feedback from them. You’ll catch problems that your QA team missed fast and you might get some brutal (but good) feedback on your design and UI/UX.