Skip to main content

RIP Onload: Finding a Better Measurement Yardstick

Buddy Brewer (SOASTA), Philip Tellis (SOASTA)
Location: 212 Level: Intermediate
Average rating: ****.
(4.14, 7 ratings)
Slides:   external link

One of the enduring holy grails of web performance is finding a better speed metric than time to onload. Often however this is not the best marker of when a user perceives that a page has loaded. In this talk we’ll show you why onload often isn’t the best metric. Then we’ll analyze the available alternatives, dig into their strengths and weaknesses, and try to pick a winner. And what if their isn’t a clear winner, or if the winner depends on a set of variable conditions? We’ll share our tips for figuring out which approach is best for your site by measuring the impact of each potential metric on your users.

Photo of Buddy Brewer

Buddy Brewer


Buddy Brewer is VP Engineering responsible for SOASTA mPulse. Prior to SOASTA, he co-founded Real User Measurement company, LogNormal which was acquired by SOASTA in 2012. Prior to LogNormal, Buddy spent ten years in the web performance industry building synthetic monitoring tools helping the world’s largest web sites solve their performance challenges.

Photo of Philip Tellis

Philip Tellis


Philip Tellis is a geek who likes to make the computer do his work for him. As Chief Architect at SOASTA, he analyses the impact of various design decisions on web application performance, scalability and security. He is also the lead developer of “boomerang” — a JavaScript based web performance testing tool.

In his spare time, Philip enjoys cycling, reading, cooking and learning spoken languages.

Philip has spoken at several conferences in the past, including FOSS.IN, FREED.IN, Ubuntulive, Linux Symposium, PHP Quebec, ConFoo, FOSDEM, IPC, WebDU, Velocity and JSConf.