Thank God it's Black Friday, countdown to Black Friday, shop till you drop, I like my Fridays Black, … Mediamarkt, Fnac, Zalando, bol.com, they all have their Black Fridays, but can their website handle all the crowds?
Back in the days
To know the origin of the term, we have to go to the United States where Black Friday is the day after Thanksgiving. Also known as the busiest shopping day of the year. Stores open very early that day, doing all kind of promotions, early bird specials, … It's considered to be the start of the holiday shopping season.
In the early 50's the term Black Friday was first used by the police to describe the traffic chaos in Philadelphia on that day. Later, since the 80's, it is used more often to describe the profits the merchants made. The losses were written down in red, however on that day, the red figures turned black, hence the term.
In Europe however, not a lot of people are celebrating Thanksgiving. Nevertheless, shops started to adopt the term to describe the busy shopping period with promotions and discounts like for example during Christmas.
Is your website or application up for the job?
Now in a global market with growing e-commerce importance, these Black Fridays are a challenge. More shoppers are leaving their cars in the garage and buy online. So the question for the companies is whether their website or application is up for the job? How to prevent it from slowing down or even crashing?
In 2018, Walmart's website had difficulties because the site couldn't handle the increased load on that day. It affected around 3.6 million shoppers, costing Walmart 9 million dollars. A similar story at J. Crem, a clothing retailer. Due to heavy traffic, their website crashed in Black Friday. The outage lasted for 5 hours impacting 320 000 customers and costing them 775 000 dollars. Some painful examples that show how important it is for your website to cope with this pressure.
This is where performance testing comes in. How can we get the most out of our website? How do we make sure that our application does what it needs to do, even during peak times?
Plan of action
The first thing we need to do is to make an estimate of the situation. How many users will visit our site? Will there be peaks and around what time? What will users do on our site? Browse, compare goods, put something in their shopping cart, pay, … Make sure that you have a good understanding of the situation?
Next step is to sit together with the technical team and ask yourselves following questions:
- What are the weaknesses in your system?
- What are the tests that we need to do to make sure that our website or application can handle the (continuous) load and that it can handle the spikes?
These tests are simulated with a load testing tool. There are different tools on the market, all with their pros and cons. So, choose the tool that best fits your needs.
And the final step is, of course, is running and analyzing your tests. Identify the bottlenecks and weaknesses in your system. Determine which solutions need to be implemented and rerun your tests until you have a stable application/website. One that can handle the expected load.
What happens if there are two or three times more users than normal?
To conclude, e-commerce keeps on growing in importance. Most companies are aware of that and already have a stable website or application that can fulfill the needs of their customers.
However, what happens if all of a sudden there are two or three times more users than normal? What if there are peaks because of special actions like Black Friday? Can our infrastructure handle all this? To make sure of that, more and more companies are starting to use performance testing so that their customers won't have issues during those busy shopping periods. We can conclude that e-commerce and performance testing are a match made in heaven.