Soon it will be Black Friday again, directly followed by Cyber Monday. We’ve all seen the videos of people racing to get the best articles with the highest discounts from a psychical store, but the actual surge in shoppers is in the online space.
The term Black Friday came into being in the 1950s, when factory managers in America saw an increase in employees that called in (presumably sick), the Friday after Thanksgiving. For most, it was time to recover and return the gifts they received. In the 1980s it came to stand for the phenomenon as we know it today, a seriously busy shopping day. It marks the start of the holiday season, which for many of us equals a period in which we buy and gift things to friends and loved ones. Cyber Monday is a more recent development, for stores to add an additional moment to the sales frenzy.
In recent years for most eCommerce stores, Black Friday has become an important marketing event in the year, a chance to boost their annual revenue. But for online commerce it’s not only about the best discount, delivering best experience is equally important to make shoppers buy from you. When it’s slow, the link is not working or your site is unavailable, discounts don’t matter, shoppers will click away and likely buy elsewhere.
So, each year the question you should ask yourself, is my website ready for an influx of visitors during the Black Friday? In this blog we will provide 3 questions to ask.
1. Prepare for a spike in traffic, get the numbers straight
Since an ever-growing part of our (gift)shopping is done online, websites that offer products need to be ready for a sudden increase of visitors. You do not want a virtual version of people queuing in front of your store and rushing in to plunder your goods, as we’ve seen happening in real life.
The best way to prepare yourself is to look at your numbers and build a forecast model:
- Check your analytics data to find out the busiest periods per day and per month.
- Use this data and match it with the site speed metrics at the same moment of time.
- As a final step, talk to your DevOps teams and correlate the data with the capacity of the infrastructure. Together this will create a capacity model as shown in the graph model.
Now you can update this model each week/month and add forecasted traffic, consisting of organic growth and marketing events. Over time you will start to see trends and validate if an increase in visitors has impact on performance.
2. Speed of interaction matters, stay ahead of the competition
Humans tend to respond better to data if this is presented fast and reliable. The faster you can deliver a response on a mouse click, the more engaged a visitor will be.
The most asked question is, how fast do I need to be? This is not a straightforward answer, as it depends on your product, competition, and customer profile. But there is a simple way to start, just look at your competition. Visitors usually compare new experiences with experiences from the past. When your site performs better than the last ones they visited, this will give you an edge as visitors will notice the difference and become more engaged.
For this reason, we at MeasureWorks have created our own benchmark tool. This allows us to always check how your site is performing against direct competition, and thus create initial insights if your website deviates enough to bother visitors.
In our blog from last year, you can also see which online company had the best performance, and where this performance stood out from the crowd
3. Test, test, test!
It is a great start to build a capacity model and compare yourself to direct competition. But with a lot more visitors the slightest bug can already result in downtime or slowdowns, or worse; customer complaints. And if you’re down, it will have a big impact on sales and brand loyalty. The only way to make sure you’re actually ready for peak volume is to test for the maximum capacity.
But where do you start and what things are essential for you to consider during the preparations for Black Friday?
- Look (again) at your analytics data to discover the top 4-6 customer journeys
- Combine this with your marketing plans for Black Friday, including landing pages and banners
- Simulate these customer journeys to create a mix for testing similar to actual behavior on your website
- Execute a first test at normal levels (eg. Busiest day from last month) to set a baseline
- If successful, increase the volume with 20-30-50% until you exceed the acceptable performance thresholds (see step 2)
- If the application breaks at any time, fix it and repeat the test again.
By putting your website under pressure you will discover in the early stages if your website cracks or discover if you can deliver a consistent user experience at higher volumes.