I am frequently asked to provide some financial justification around doing certain things to improve response time of web applications, especially in the e-commerce space.

I usually end up building a model based on response time vs conversion rate, showing how the reduction (or increase) of the first is causing a change to the latter.

I base the estimates on the three different models:

a) Amazon, where a 100ms increase in latency has a 1% hit on revenues [0] b) Google, where a 0.5 seconds increase in latency decreases traffic by 20% [1] c) Walmart, where a 1 second reduction in the response time causes a 2% increase in conversion rates and for every 100ms they grow incremental revenues by 1% [2]

All those numbers have been circulating from multiple sources and shown in a number of presentations, but they're all fairly dated (some of them are from 2006!).

If you do, how do you handle this task? Do you have more recent statistics to base models on?

Thank you!

[0] http://blog.gigaspaces.com/amazon-found-every-100ms-of-latency-cost-them-1-in-sales/

[1] http://glinden.blogspot.com/2006/11/marissa-mayer-at-web-20.html

[2] http://www.webperformancetoday.com/2012/02/28/4-awesome-slides-showing-how-page-speed-correlates-to-business-metrics-at-walmart-com/