RUM only beginning its APM relationship

Application performance management (APM) has a lot to learn from real-user monitoring (RUM). Joshua Bixby argues the point at length in a recent posting, and he’s right: “RUM is really, really cool.” At the same time, though, “… only about 20% of major ecommerce sites have adopted RUM …”

There’s a big mistake here, and it’s not Bixby’s.

RUM fills big APM holes

Testing is crucial; RUM doesn’t change that. There’s a lot that the testing traditional in APM simply misses, though, and that only turns up in RUM or something very much like RUM. The foundation of RUM is to test actual user experience, so all the variations in browser, geography, latency, firewalling, packet loss, user data, traffic proxying, navigational trajectory, and so on that are hard to model in-house come through automatically.

RUM’s focus on Web experience also gives it a kind of universality that the first generation of APM lacks. APM originally was restricted to instrumentation of Java and .Net sources. While RUM’s reliance on JavaScript entails its own difficulties, it’s at least fair to say that nearly all e-commerce consumers enable JavaScript.

With all this going for it, why haven’t more sites adopted RUM? Why, in particular, is the RUM screen in Google Analytics so infrequently used by customers who already depend on Google Analytics?

Bixby’s answer isn’t particularly satisfying. He concludes RUM simply “… hasn’t trickled down …” yet from “the mega-players like Google, Amazon, and Etsy …” True, but what’s the reason? As Bixby himself underlines, “What does this say about the potential long-term success of RUM?

Not much, probably. RUM is still young. While it will take several follow-up postings to explore all the dynamics of RUM adoption, most important is that RUM remains so young. Even though it looks like a natural candidate for attention in a world focused on such hot topics as Big Data, end-user conversion, and information technology (IT) strategy, my best guess is that too few decision-makers yet understand the benefits RUM brings.

The good news is that the market leaders in RUM are largely intelligent and realistic: they recognize that RUM complements more than replaces traditional APM, they know there’s a lot of education left to do, they’re all aware that “education” is a priority, and they generally agree that their “out-of-the-box” experience still has room for improvement. RUM truly has a lot to give. I applaud Bixby for his estimations of how distant the market is from saturation. The “Application Monitor” will return several times over the next six months to look at adoptions by specific organizations and the results of those adoptions.