The U.S. Bump – A Quick Look at Web Traffic Patterns

After launching Real User Monitoring earlier this year, developers around the world were able to see performance data from all of their customers in one place. They started to see end-user performance right next to application performance, and embarked on a whole new tier of app performance improvements as a result.

By providing Real User Monitoring, we see requests from our customers’ customers. Every user of our customers’ web apps executes a tiny async javascript snippet. This snippet sends the browser page performance back to us, so we can display performance metrics to the folks who made that web application. (Here’s a more detailed post on How RUM Works if you’re interested).

This means that our RUM service gets hit once for every single page view of every RUM-enabled customer site. We’re peaking around 300k requests per minute right now on our RUM service and growing every day.

US Consumers Dominate RUM Customer’s Traffic

While we have many international RUM customers, often those sites are also visited by US customers. Let’s take a look at some traffic patterns we see on behalf of our customers:

We see a significant increase in traffic around 9AM Pacific Time every day.

Here’s another view comparing a particular day, the previous day and a week earlier. 9AM Pacific… proof us Californian’s get to work a bit later in the day, right? This boost must be the California morning rush, right?

Not so fast! With our Real User Monitoring we can see app performance by browser and geography! Let’s take a look at where this bump is coming from.

California does show some increased throughput during our lazy 9AM mornings, but more importantly we see an even more significant increase in traffic from New York. New Yorkers lunching are the primary drivers of our bump!

Another thing we notice is that weekends tend to have less traffic than week days. These two smaller humps are from a recent Saturday and Sunday. In aggregate, traffic tends to be highest on Mondays and slowly tapers off during the week.

Looking at these weekend graphs has me wondering if the percentage of mobile browsers changes significantly during the weekends or not. Are more of us using mobile browsers to sneak in some web time on the weekends, as compared to our laptop browsing mid-week? I’ll take a look next week and let you know. Until then, happy browsing!'

Marketing at Github View posts by .

Interested in writing for New Relic Blog? Send us a pitch!