For one of our larger Rails clients, the answer’s a definite yes. This site is a shopping cart application where user experience, and retention converts to clicks which convert to dollars.
Shoppers on this site now feel the benefit of a 25% improvement in average response time (sampled across 36hrs) since the upgrade to 1.9.3 was deployed.
Another way of cashing in on an improvement in your sites capacity to deliver content, is of course to shed hardware and reduce costs. Theoretically, on this cluster of 6 application servers, we could drop 1 of these servers, and we’d still be ahead.
An unexpected benefit of upgrading to 1.9.3, is that the carrying capacity of the site perhaps doubled. By this I mean, we could experience a doubling in traffic before we have any adverse effects.
Co-incidentally perhaps, the business gets the benefit of a 25% increase in throughput, suggesting that users are doing more while visiting the site.
