Imagine yourself at a concert by your favorite artist. You even sprang for VIP seats because the venue is the hometown stop on their latest world tour. The hefty premium you paid bought you tickets with a direct sightline to the stage—and access to a mobile app where you can order food and beverages delivered right to your seat. You don’t have to miss a minute of the show. “This is incredible,” you think, “Technology is magic!”
Alas, there are a few hiccups.
After scrolling through the options on the app, you order a $57 bottle of wine. That’s more than you would normally spend, but the mark-up seems worth it for the convenience. As you step through the app flow—selecting the wine, entering your seat number, entering your credit card info—everything is great … right up until the moment the app crashes.
Undaunted, you start up the process again … and the app crashes again. And again. After three crashes in a row, you give up and walk back to the concession stand where you spend $25 less for the exact same bottle of wine. But you miss your favorite song while waiting in line—and you end up annoyed at the artist, the venue, and especially the app maker. You take to social media to make sure the whole world knows you’re never going to use any app or service from them again.
New Relic could have helped!
Something like this actually happened to me recently, and it drove me crazy. As a New Relic account executive, I work with my clients to prevent these kinds of problems.
Now, it’s not always clear in these cases where the problems occurred. Maybe the venue’s network was overloaded by tens of thousands of thirsty music fans. Maybe it was something else.
But the cause really doesn’t matter, at least not to the frustrated concertgoers. Somewhere along the way, in the application, in the browser, in the technology stack, in the network, or in the complex interactions among all these things, the desire to drink a little wine resulted in errors that ruined the customer experience. The vendor lost a sale and created an unhappy customer. Worse, the team behind the app may not have had any idea that things were going wrong.
5 key questions
Given the situation, I see five key questions that the engineers and business executives behind this application should be asking themselves:
- Did we detect this problem in first place?
- If we did detect it, do we know what combination of applications, containers, databases, infrastructure, or third-party services caused the problem?
- Could the issues have been quickly resolved in real time?
- Could the problems have been prevented before they even happened?
- Do we know how widespread the problems were, and how much revenue we lost because of them?
To address the first two questions, the vendor needs to have a set of monitoring tools in place that tie into an alerting platform. One reason why customers love using the New Relic platform is because it monitors their entire technology stack from a single place. It includes a robust alerting system that could help answer the first two questions. And because New Relic can house all of your application and user experience data in one place, it can help you answer questions three and four.
Ultimately, though, question five is the key. At the end of the day, if you cannot quantify and understand how much money you’re losing from crashes like these, you can’t know how serious the problem is and how much resources to devote to fixing it.
To help our customers measure the money metric, New Relic provides an API that you can use to pull in financial data from POS systems, ERP tools, and third-party applications. You can correlate this data with the application performance and infrastructure data that New Relic helps you collect to understand “What is the sum of failed transactions that occurred at the concert due to errors and/or slow loading?”
You never know … maybe if the dev and ops teams behind the venue app had been using New Relic, they would have known things were going wrong, and been able to fix them in real time. Maybe then I’d have been able to stay in my seat and still get my bottle of wine—without missing my favorite song. And maybe, just maybe, the vendor would not have lost potential sales and turned off long-term potential customers.
I would raise a glass to that!