Teevio LLC

I had a question posed to me the other day. It happened to be one of my most dreaded, but most inevitable questions as a web developer. The question was, “How does this look in IE?” But to take it a step further (and worse) they followed up by asking, “What about IE 5?”

While everyone is concerned about IE 6 and 7 I haven’t really thought about IE 5 (or anything before 6) that much. I had wrote it off as a dead browser. I quickly came up with a few responses about why IE 5 was obsolete, etc. But, to someone who owns a website and cares very much about all their customers, my ‘excuses’ didn’t cut it.

I decided to come up with a couple of tests that would help a client decide if developing for IE versions less than 6 was necessary. Hopefully, these test will help you in the client world.

The Traffic Test

For websites that are getting redesigned and have some kind of traffic analytics, it is best to see how many people with versions less than IE 6 are actually using your site. This will give you a good feel for whether it is worth the effort. It turned out for the situation I mentioned above that the percentage was about .06%. In terms of traffic this is a negligible amount, you can stand to lose those users. But let’s say that the percentage was 2% of the user base? That’s starting to become a significant portion of your users.

The Demographic Test

Taking into account for the percentages of users, you should also think about the demographics of users who might use the website. I’m not going to specify an age range, but there is point where the computer user starts to become less savvy, and less aware about how to upgrade to newer browser versions.

The Verdict

I’d say if you have a pretty significant user base of IE 5/5.5 users, and your demographics tend to be in the older range, it might be worth developing for IE 5/5.5. Of course, you have to think about time and money constraints. In my case above, there wasn’t much of a user base, but the demographics could dig into the older generation. My solution was to get everything running in IE 6 and 7 and if there was time burn, try and make the lesser versions as usable as possible.