Customer Sentiment: The Missing Metric in Monitoring

Over in the utility industry, many power companies have a foolproof method for detecting an outage: wait for angry customers to call. Granted, your service is already disrupted at that point. Your customers are ticked off. Your company’s reputation is getting dragged through the mud. But in terms of measuring customer experience, you won’t get more accurate.

Of course, for companies providing digital products and services, this doesn’t sound like much of a tradeoff. You’d much rather detect problems before they affect your customers and your bottom line. This is, after all, why monitoring is a multi-billion-dollar global industry. But there’s a nugget of something useful in the power company model: hearing firsthand what customers are saying. As powerful as modern monitoring tools are, they just don’t capture that kind of insight. And the picture you maintain of real-world user experience is poorer as a result.

Building Modern Monitoring

Today, enterprises have a few different methods they use to detect when customers are having issues accessing their digital products and services. (That is, when an online app won’t load, an e-commerce transaction gets hung up before completing, users see strange error messages instead of content, and so on.) These include:

• Internal system and service monitoring: Companies employ a fairly complex architecture to track all sorts of things (servers, infrastructure, application resources) that may or may not equate to a problem affecting end users. But these metrics provide an important piece of the puzzle in staying ahead of issues.
• Analytics: Analytics can come from tools like Google Analytics, StatCounter, a hundred others. These tools are geared more toward understanding customer behavior than detecting outages, but they can provide useful warning signs if, for example, you see a sudden dip in visitors.
• Specialized monitoring tools: The third method and most effective method is to employ tools like synthetic monitoring, endpoint monitoring, real user monitoring, and others that explicitly measure user experience. These tools use bots deployed in locations all over the world to simulate user behavior and measure the experience from their point of view.

By combining these approaches, you can get a pretty good gauge of the experience users are having when accessing your digital products and services. Often, you can detect problems when they’re still small and localized, before they blow up. But they can’t monitor everything, everywhere.

Watch Out for Blind Spots

Even the best tools inevitably have blind spots. And even if you detect that some users, in some location, are having issues, it can still be hard to validate what’s happening.

Seemingly straightforward digital content can encompass a complex web of third-party services and interdependencies. If users in Ireland can’t access your site, the problem might be with your infrastructure serving that region. It might be an issue with a regional ISP. It might even be that services you rely on from a third-party provider, like AWS or Akamai, are down. 

There’s another, potentially bigger problem though: damage to your brand. When users can’t access your digital products and services, they’re unlikely to be quiet about it. Instead, they’ll to take to social media (often Twitter). That might start with questions: “Is this site down? Is anybody else having problems?” But it can quickly turn into complaints. And the last thing you want to be is a trending topic on Twitter for the wrong reasons.

Tapping Social Sentiment

Social media sentiment can cause real problems, but it can also represent a valuable tool to spot problems. After all, no matter where your users are, if they have internet access, they can get on Twitter. It’s like a grassroots monitoring architecture that gives you real-time information about real-world users pretty much everywhere—beyond just the places you’re actively monitoring.

By keeping tabs on what users are saying about their experience with your digital products and services on social media, you can:

• Validate issues and outages earlier
• More effectively localize problems
• Validate what your monitoring tools are telling you
• Keep your finger on the pulse of user sentiment around your brand

In theory, you could take user sentiment monitoring even further. Why just track what they’re saying about your own company? You could also capture sentiment around partners and competitors. You could even keep tabs on user experience with cloud services (like AWS or Microsoft Azure) that your content relies on and get ahead of complaints when you see those services having issues.

Listen to Your Users

Modern monitoring tools are incredibly powerful and getting more so all the time. But they’ll never be able to show you everything happening to every user, everywhere. The only ones who can do that are users themselves.

You wouldn’t want to rely only on social media complaints to know when something is wrong with your digital products and services. But these sentiments can be a valuable part of your overall monitoring strategy. If you’re not capturing them, you’re missing a real opportunity to see your business from your customer’s point of view.

Nithyanand Mehta is executive vice president, technical services, and GM at Catchpoint. Mehta leads Catchpoint India’s global technical services teams, which includes professional services, sales engineers, and support. Mehta has 10 years of experience in the monitoring industry and 15-plus years in the internet technology space. Mehta specializes in helping organizations implement digital end user monitoring strategy and enhancing end user experience and is responsible for Catchpoint India’s general management and technical services practices.