The reality is that you may well have to deal with the issues mentioned in the previous entry, and more, when trying to make your site work in IE. Many businesses have standardized to “Microsoft only” and use Internet Explorer as their default browser, despite well-known security risks associated with doing so. Even worse, many institutions have built their websites solely to the features of IE, to such a degree that upgrading to a browser that attempts to follow standards – even later versions of IE – breaks the website.

In most cases you can at least attempt to convince the client to upgrade their browser, if they choose to stick with IE: any later version is better than IE6. You may also be able to build an argument for dropping support for certain browsers based on market share statistics.

Please keep in mind that web browser market share numbers are notoriously imprecise: different sites will draw different types of audiences, and thus a different array of browsers. Ultimately, the only way to have any degree of certainty is to look at the server logs for your own website, or to use a tool like Google Analytics. Of course, those techniques only provide information once your website is up and running.

Personally, I refuse to develop for IE7. The browser is now more than a decade old, represents less than 0.3% of visitors to this site, and is not worth the time and associated stress to develop for, at least to me.

However, you may not have that choice… and your client will always expect the site to appear perfectly in whatever browser they use, including IE 7 or 8.

Even if you do convince the client to drop support for earlier versions of IE, you should ideally continue to provide a path for users of the browser to view and use the site. Blocking your website to particular browsers is not the way forward, as a general rule. My personal recommendation is to use ChromeFrame, equivalent to giving IE a brain transplant). If the client is absolutely adamant in her support of IE, or if the audience numbers demand supporting the browser, you should add the extra demanded time to your contracted hours in developing the website. As a very rough heuristic:

  • Supporting IE7 & 8 and ignoring IE6 adds 20 ~ 25% to estimated development time.
  • Supporting IE9 and 10 only and disregarding earlier versions adds only 10% to development schedules, in most cases.