Imagine that it’s “show time” for your company’s annual peak period of e-commerce traffic. If you’ve ever been an e-commerce manager for a toy company on Black Friday, a floral company the day before Mother’s Day, or a sporting events ticketing company a month before the Super Bowl, then you can surely relate.
Your online customer service representatives are trained and primed to help your customers. Your warehouse shelves are stocked, and your logistics providers are all lined up at the door. Your marketing promotions and campaigns are in full throttle, and your feature-rich Web applications including search, online catalogs, shopping carts, order status information, ratings and reviews, streaming video and more are all ready to roll.
Then — boom! — something in your most critical Web application goes awry the very day or days that exceptional performance is needed most, bringing your e-commerce operation to a screeching halt. Thousands of shopping carts and product searches go abandoned, and you have more disgruntled customers than you can imagine.
“Impossible,” you might say, “we’ve conducted internal testing of our Web applications inside and out! All our tests have passed, and we are confident our internal infrastructure can handle our best-case traffic and then some.” I’m here to tell you — testing internal components is not enough to “get it right.” With potentially huge revenue hits and brand image on the line, is anything less than exceptional performance really something you can live with?
You’re Only as Strong as Your Web App’s Weakest Link
Let’s start with a look at today’s Web applications, which have evolved from single-function tools to extended, interdependent, multi-tier delivery chains comprising numerous third-party applications and services. The performance of your Web application in its entirety hinges on the performance of each and every third-party application or service comprising it. Consider an online sales application which includes search, shopping cart and check-out functionalities. Together, these functionalities comprise a highly interdependent Web application delivery chain, and poor performance at any step can bring down performance of the entire application.
Today’s modern Web sites incorporate an average of six third-party applications and services delivering content and functionalities from beyond the firewall, all converging and assembling in your customers’ browsers. Third-party applications and services are more prevalent than one might think and include such commonly used features as CDNs, Omniture and Google Analytics. While these third-party applications and services are designed to enable a richer online experience, they also present a liability since it’s estimated they comprise 50 percent or more of the time a user spends waiting for a Web site or application to load.
In a recent study, Aberdeen Group found that a 1-second increase in response time can reduce online sales conversions by 7 percent. In the event of poor application performance, customers simply don’t care which of your third-party application providers is to blame. Instead, they will hold you responsible, and failure to guarantee performance anywhere in your Web application delivery chain can result in significant damage to your brand and revenue.
For these reasons, online businesses cannot shrug off responsibility for the external components that come between them and their customers. These components include not just third-party services, but an extremely varied set of individual user circumstances (including customers’ geographies, Internet service providers, connections speeds, browsers and desktops) known as the “last mile.”
In this context, Web applications require a new approach to load testing — known as “load testing 2.0” — to manage the total user experience across the entire Web application delivery chain and from a last mile customer perspective, during periods of both normal and peak traffic. Load testing 2.0 is the only way true way to see through your customer’s eyes and expose the actual performance of your applications under various load sizes, to find and fix problems before launch and ensure consistently strong Web application performance.
Getting Real With Load Testing 2.0
Traditional, older-style load testing tools are usually found deployed in massive (and often expensive) test labs and are usually the realm of application developers and quality assurance teams. Load Testing 2.0 differs from the traditional labs-based approach in three primary ways:
- Load testing 2.0 measures how the extremely broad range of variables in a complex Web application delivery chain perform for your customers under peak loads, helping to identify, isolate and fix performance problems anywhere in the chain, and before customers are impacted. This is in contrast to a purely lab-based approach, which is walled off from external variables and therefore incapable of rendering a complete view of the customer experience. Remember, even if your internal components passed performance tests with flying colors, your application performance can still be failing.
- Load testing 2.0 delivers the realistic, last-mile perspective of customers. Chances are, your customers don’t live in your data center or anyone else’s. They’re not running on a server-grade machine and aren’t hooked up to a data center-grade Internet connection. They’ll be in homes and businesses around the world, using a wide variety of ISPs and consumer grade connections on a variety of desktop machines and browsers.
The key to gauging how your applications perform for different customers is to therefore use a testing network that measures performance in a wide range of geographies across an extremely broad range of last-mile scenarios. Since building a large, on-demand network of desktop browsers is likely beyond the capabilities of most load testing labs, load testing 2.0 taps established global end-user owned desktop monitoring networks to provide exact measurements on how different customers in different geographies experience applications under variousload sizes. Armed with this information, you can prioritize and fix potential performance problems for key geographies and user segments.
- Load testing 2.0 solutions are available on-demand as self-service, user-friendly tools, making them much more accessible to more individuals within your organization. Any party with a stake in your Web operations performance — including e-commerce managers, IT staffs and application development teams — can take advantage of these tools whenever they want and as often as they want, and test more last-mile scenarios.
Also, by using a worldwide network of “cloud computers” for testing, load testing 2.0 solutions help eliminate expensive investments in test labs or consulting arrangements. Worth noting: The on-demand nature of load testing 2.0 means you don’t have to schedule tests or wait to conduct them, which is often the case with more traditional labs-based approaches and can lead to application deployment delays. However, this doesn’t render existing investments in lab-based approaches obsolete; rather, you can reserve load testing 2.0 for your most time-sensitive, business-critical and/or complex applications, thereby supplementing lab-based efforts.
The bottom line is this: For businesses operating at “Internet speed,” greater load testing flexibility can help ease and speed application development and deployment, which helps ensure optimal performance for time-sensitive, business-critical applications.
When Do You Need Load Testing 2.0?
Thanks to their ease of use and on-demand nature, Load Testing 2.0 solutions are appropriate and valuable for a number of marketing or IT initiatives, including when:
- you are launching new marketing campaigns, offering major promotions and/or hardening for peak sales periods. In the current economic environment, online retailers and businesses need to protect every dollar of revenue and can’t afford to lose sales to poor Web application performance. Applications simply must perform well, especially during high-traffic, high-visibility periods like marketing promotions and holidays.
- you are releasing new features and/or incorporating a new third-party application or service into a Web application. The key is to identify and fix potentially costly and critical problems in your Web application delivery chain before launch, not after.
- you are upgrading or virtualizing your IT infrastructure. Here, load testing 2.0 can deliver valuable peace of mind that your most important Web applications can scale and will continue to perform properly during or after any major infrastructure change.
- you are adding cloud-based services. Load Testing 2.0 can give valuable assurance of the ability of both internal and external cloud-based applications and services to scale to load. Load testing 2.0 also offers validation of service level agreements (SLAs) and a knowledge that you’re getting what they’re paying for.
- you are entering a new market or a new geography and view Web application performance as a key to winning customers and improving market penetration.
In a down economy,competition is fierce. Your customers’ Web experiences are paramount to winning and maximizing fleeting revenue opportunities. Traditional load testing methodologies can show the load “breaking point” for your internal infrastructure, but only load testing 2.0 can show the load “breaking point” for your customers, no matter where it may be in the Web application delivery chain.
Businesses that fail to conduct rigorous, comprehensive, end-to-end testing to ensure their most critical applications can scale under load risk turning their customers over to their competition — and rest assured, your competition is ready to pounce by ensuring nimble, high-performance Web sites of their own.
The key to guaranteeing application performance under any size or type of load lies in quickly and easily gathering performance data across the full breadth of your Web application delivery chains, from a realistic, last-mile customer perspective. Fortunately, today’s tools deliver these capabilities to enable rapid identification and resolution of problems, while being cost-effective, available on demand, flexible and easy to use — in sum, expressly tailored to the needs of competitive, agile businesses that understand the critical “make or break” nature of online customer experiences.
Imad Mouline is CTO of Gomez..