A huge number of words, nerves, time, emotions in the work of the site developer goes to fight IE6. On Habré, one or other suggestions appear periodically, what to do about it, I dare to suggest my own version.

In our work, we somehow began to consider the statistics of labor costs in the layout. In particular, in the variants with and without support for IE6, labor costs differed on average by 2.5 times. Significantly, I agree. But sometimes it was more, 2.5 - the average. A little thought, I came to the conclusion, why should I determine my own policy on IE6? After all, it is possible to offer the client both cheaper and more pleasant (more buns, cleaner code), but without IE6 support, or we will ensure that everything indicated in the TOR works in IE6 as well as in more recent browsers, but it will cost expensive.
On the one hand, this is not a complete refusal of support, on the other, encouraging the client to abandon it. Again, this can be served under a different sauce - as a rise in the cost of services when using outdated technology or as facilitating the price while following the progress.
')
In our case, to support IE6, we use a factor of 2.5. It works - not for everyone, but over time, the percentage of people willing to pay for “nine-year-old milk” two and a half of the price gradually decreases.
PS: In the winter, I wrote a post on this topic in the sandbox, but then for some reason they did not appreciate it, now as a habrovchan I would venture to repeat myself.PPS: The idea is far from new, and for some readers it will be an articulation of evidence, actually written for the rest of the audience.