Practical guide to controlled experiments on the web: listen to your customers not to the hippo

400 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Collaborative Filtering Recommender Systems (2011, background): “Field trials, frequently in the form of A/B tests, are commonly used in commercial recommender deployments where sites have already established user bases and can test a new algorithm or other site change on a small subset of their users [76, 77].”
The cost of a cloud: research problems in data center networks (2008, background): “There is substantial empirical evidence suggesting that performance directly impacts revenue [22].”
Data center TCP (DCTCP) (2010, background): “For interactive, soft-real-time applications like these, latency is the key metric, with total permissible latency being determined by factors including customer impact studies [21].”
Automatically generating personalized user interfaces with Supple (2010, neutral): “[43] offer some helpful initial insights.”
WTF: the who to follow service at Twitter (2013, neutral): “, [14] for more discussion.”
Crowdsourcing graphical perception: using mechanical turk to assess visualization design (2010, background): “[15] provide a brief survey of experiments and recommendations for web experiment design.”