Blog Post

Retail Resilience: Lessons Learned from Cyber Week 2023

Sales were up compared to last year, and while there were no major outages, availability and performance issues were rampant and disrupting the customer experience. Let's explore some examples.

Black Friday and Cyber Monday this year marked a strong recovery for the retail and e-commerce sectors. Consumers were more eager to spend compared to 2022.  

Adobe Analytics highlights a significant jump in online sales, reaching $9.8 billion on Black Friday, up 7.5% from last year. Cyber Monday also saw an impressive rise, with sales hitting $12.4 billion, a 9.6% increase from 2022. These trends were reflected in our Real User Monitoring (RUM) data, especially across various luxury brands, where we noticed a surge in traffic and conversions.  

The graph below shows a 25x increase in products sold during Cyber Week for one of our customers.  

Cyber Week challenges

Over the years, we’ve noticed that although retailers put extra effort into monitoring their applications, there is always a gap in the monitoring strategy during peak periods. The monitoring strategy is often focused on handling complete outages or site reachability, leaving retailers and service providers vulnerable to other performance issues.  

It’s tempting to assume that a Cyber Week without any significant outages like the one we’ve just had means everything went smoothly. Major issues with online retailers were rare, and social media was relatively calm. That said, our Internet Performance Monitoring (IPM) platform detected various availability and performance issues across Cyber Week that hindered users from effectively navigating sites and completing purchases, impacting user experience and revenue. That’s why it’s crucial to ensure the resilience of your Internet Stack during peak periods, which is the main objective of our Internet Resilience Program (IRP).  

Internet Resilience Program: Delivering performance when it counts  

Customers who joined the IRP had access to a team of expert engineers to help ensure the performance and resilience of their websites and applications during Cyber Week.

Here’s an overview of what the IRP offered:  

Catchpoint’s IPM platform and dedicated performance team:

  • Access to Catchpoint’s industry-leading IPM platform.
  • Support from a dedicated performance team to develop tailored testing strategies.  

Real-Time Issue Detection and Resolution:

  • Continuous 24/7 monitoring of applications and websites.
  • Prompt identification, reporting, troubleshooting, and resolution of potential issues.

Comprehensive Post-Event Analysis:

  • Detailed analysis of monitoring data post-Cyber Week.
  • Benchmarking against key competitors and recommendations for performance optimization.  

Catchpoint detected numerous application and delivery issues during Black Friday through Cyber Monday, including:  

  • Cart and checkout issues
  • Latency due to third-party tags
  • Connection failures
  • Search result page not working
  • Sign-in/login page performance drop
  • Broken content
  • Server errors  

500 – Internal Server error at checkout page

Error on Cart page after adding product to the cart

Unreturned search result

Below is a graph illustrating a spike in Largest Contentful Paint (LCP) for a Homepage we monitored during the sales event. LCP is a key metric for measuring user experience, as it indicates the time taken for the largest content element to become visible on the screen. A higher LCP can lead to slower page load times, negatively impacting user experience by making users wait longer to interact with the site's content. In this instance, the LCP increased from 1.5 seconds to over 5 seconds.  

 

Before Spike

After Spike

Before the spike, users experienced a page rendering time of under 2 seconds based on the filmstrip. However, following the spike, the page's content took more than six seconds to appear for users. This delay was mainly attributed to the inclusion of a sizable video file, exceeding 2MB, causing a significant slowdown in the page's loading speed.  

Upon detecting this LCP spike, we immediately took action and alerted our customer. By promptly detecting and notifying this incident, we ensured a smoother experience for users and maintained the site’s performance during this critical sales event.  

Why monitor the Internet Stack?

This year, the Internet Resilience Program noted a 14% increase in Internet Stack incidents compared to 2022, emphasizing the importance of IPM to gain deep visibility of the entire Internet Stack during times of high traffic.  

 

Given the complexity of the Internet Stack, when your business is disrupted, how do you figure out where the disruption is? The only way is by having visibility into all the layers of the Internet Stack. Considering the financial impact of Internet disruptions, a robust IPM strategy to monitor your entire service delivery chain during the busiest time of the year becomes even more critical.  

Beyond Cyber Week  

The Internet Resilience Program goes beyond managing the Cyber Week rush. It’s essential for any high-traffic period, ensuring the resilience of your Internet Stack when you need it the most. Whether it’s a major sporting event like the Super Bowl, an iconic entertainment occasion like the Oscars, or a surge in traffic due to a celebrity endorsement – think Taylor Swift sporting your brand – Catchpoint will be there to support you.  

Discover more about the IRP and how it enables the resilience of your Internet Stack during peak periods. Visit our IRP page now.

For tips on how to prepare your website for times of high traffic, download our ebook.

Internet Performance Monitoring
Internet Synthetic Monitoring
Real User Monitoring
Customer Experience
News & Trends
eCommerce
This is some text inside of a div block.

You might also like

Blog post

Bridging the IT-business comms gap comes down to this one word: Ask

Blog post

A Gulf Tale: Navigating the Potholes of Customer Experience in the Digital Era

Blog post

The Power of Synthetic Data to Drive Accurate AI and Data Models