Black Friday to Cyber Monday. That time of the year when we fight the crowds and traffic to squeeze a month’s worth of shopping into just a few fleeting days. Having worked in retail during my college years, I remember the customers lined up outside before 4:00 a.m., chomping at the bit to race into the store and fight (literally) other shoppers for the “scraps” that were left. Ah, the good ol’ days.
2018 Overtakes 2017 with New Records Set
I like to think we’ve come a long way since then and 2018 proved to be a landmark moment. With retailers providing better mobile experiences, decrease of in-store traffic, and consumers opting to buy online and pickup in store, Cyber Monday launched itself into the books by setting the record for becoming the highest U.S. ecommerce sales day in history with $7.9 billion in revenue, according to Adobe Analytics. Adobe Analytics, which takes data from 80 of the top 100 online retailers, reported that there was a 23.6 percent increase in online sales compared to Black Friday 2017, reaching over $6.2 billion.
Mobile is Catching Up
More shoppers are leaving the car in the garage and turning to their mobile devices, specifically their smartphones. Black Friday 2018 was also the first in history to reach over $2 billion in sales from smartphones. Mobile devices alone accounted for over 35 percent of revenue, an increase of over 18 percent from 2017. Online shopping on mobile devices (tablets and smartphones) over the entire five-day period represented over 54% of site visits, with smartphones making up over 46% of that total.
Black Friday Winners
With all that money flying around, there were clearly a lot of winners. eBay, for example, reported record sales on Black Friday and Cyber Monday. Amazon reported it sold more items on Cyber Monday than any other day in its history (include Prime Day 2018), with customers ordering more than 180 million items from Thanksgiving to Cyber Monday.
Black Friday Losers
However, behind every winner, there are unfortunately multiple losers. Lowe’s, J.Crew, Lululemon, Walmart, and Ulta were some of the more notable victims. Due to higher than anticipated levels of traffic, all these companies experienced website disruptions and outages – with devasting effects. According to LovetheSales.com, J.Crew and Walmart experienced the following:
- Crew’s outage lasted 5 hours, impacting over 320,000 customers, and costing J.Crew $775,000 in sales.
- Walmart’s website issues affected an estimated 3.6 million shoppers over 150 minutes, costing them $9 million in lost sales.
A broken website on the biggest shopping day of the year in history is an absolute nightmare for companies. Not only are you losing potentially millions in revenue and customer confidence, customers are forced to find products somewhere else (usually your direct competitors), and your PR team must try to deal with the fallout from customers that take to social media to express their frustration, in real-time, for all the world to see.
It’s clear that the trend of online shopping, particularly from mobile devices is becoming the new normal, so why aren’t companies prepared for these events? It’s a question that obviously gets brought up more often during these shopping days. Hindsight is 20/20, after all.
Black Friday happens at the same time each year, so why aren’t online retailers prepared?
Reasons can be varied, from “we don’t not have the necessary skills/personnel/resources” to “we don’t have budget” or “it’s just not part of our development process.” For smaller companies this can be the case, but for companies as big as Walmart and J.Crew, what’s their excuse? And it’s entirely possible that they did everything right and executed their load/stress tests and were monitoring critical transactions well in advance of the holidays, but something just happened to go wrong.
Whatever the case or reason, companies should continually be performing load/stress tests throughout the year. Even though the time between Black Friday and Cyber Monday are the most popular shopping days of the year, there are other major shopping days practically every month. It’s a perfect time to perform these tests to gauge how well your sites and applications handle a sizable amount of traffic, so when it’s time for the “big dance,” you’re ready to go.
Load Testing – Best Practices and Processes
LoadView, our on-demand, cloud-based load and stress testing platform, takes an outside-in approach to performance testing. It helps assess how a site, web application, or API will respond to various traffic levels, using real browsers, without you having to add any additional hardware or maintain infrastructure. LoadView can simulate thousands of concurrent users during a test, allowing you to create scenarios and distribute load based between multiple geographic locations. The results from LoadView tests can also assist in these critical areas:
- Establishing response time baselines under specific user load numbers
- Identifying performance bottlenecks
- Finding upper limits of your current systems for capacity planning
- Analyzing server performance (CPU, memory, bandwidth, disk I/O) and database response times
Additionally, along with the EveryStep Web Recorder, you can create advanced scripting actions that mimic real user actions with your application, simulating over 40 desktop/mobile browsers and devices. The scripts can be uploaded into the LoadView platform and replayed by a virtually unlimited number of simultaneous users, giving you actual performance from real browsers. The EveryStep Web Recorder is one of the few tools on the market today that allows you to interact with Rich Internet Applications (RIAs), such as AJAX, Flash, HTML5, PHP, Ruby, etc.
The scripts that you create during load testing can also be uploaded into our UserView platform, allowing you to record and playback virtually any set of critical transaction steps for ongoing monitoring. Specific user actions, such as mouse clicks, page navigation, shopping cart, and text inputs can all be monitored through our Dotcom-Monitor platform.
Load testing is also incredibly effective at simulating Black Friday scenarios to root out bad or low capacity web hosting companies. The absolute best WordPress hosting companies should be able to stand up to a thorough load test without crumbling under pressure, and should also have some capacity to spare. However, if you don’t load test your WordPress website, you’ll never know if your host is makes the cut or falls short.
The Case for Load Testing eCommerce Sites
The ability to monitor and test applications by mimicking a user’s path – and being alerted if something goes wrong – gives you and your team time to troubleshoot performance issues before they affect other users, reducing downtime, and ultimately improving the overall user experience. Not only will you protect revenues, but your users and customers be happy. You’ll gain efficiencies, achieve better results, and give your PR team a break from reading negative customer reviews and social media comments.
So, if you don’t want to see your company name in the news listed as a “Black Friday Loser,” set aside some resources to properly performance test (early and often!) your sites and applications. And if you don’t have the personnel, we can provide a team to manage it all for you through our Professional Services offering. You pick and choose what you need done, and we’ll take care of it.
