Running your Best Holiday Season with AppDynamics

By | | 6 min read


Online shopping is one of the biggest drivers of revenue and customer loyalty for the retail industry, especially during holiday season. However, preparing for the holidays comes with its own requirements for stakeholders as well. Any potential downtime or system crashes can directly impact a customer’s experience, and therefore the bottom line of a business. So how should these e-commerce companies protect the potential for increased income and a growing customer base by preparing their site and applications properly to handle the drastic increase in load? Read on to see the different methods that AppDynamics provides to protect and enhance your omnichannel retail experience throughout holiday season.


When more users start arriving on your internet application over the holiday rush, providing a consistently high performing experience is critical. During critical times, it is important to also measure a unique user’s experience and capture the resources that may be impacting their overall performance. Understanding a user’s time to onLoad will give you a good perspective on how quickly they are viewing content and interacting with each page/action of your business transaction. Using the tools available in AppDynamics allow you to inspect snapshots of poor performance and determine if they were caused by your application’s latency or by that of a third party. During high traffic times like Black Friday or Cyber Monday, third parties are delivering a large amount of content to many sites and may get overwhelmed, slowing down the delivery of content to your site.

The AppDynamics APM Dashboard (Figure 1) graphically shows performance across the all tiers, nodes and backends for all business transactions in the application. It displays the combined health for all business transactions within the application. The health of each business transaction is based on any health rule violations triggered during the selected time range. Each business transaction is scored and a transaction scorecard shows the number and percentage of transactions that are normal, slow, very slow, stalled or errors.

Screen Shot 2015-10-25 at 11.22.52 PM.png

Figure 1

AppDynamics dynamically baselines normal performance of any business transaction based on historical performances and then scores business transactions as slow, very slow or stalled depending on the deviation from the normal performance. Monitoring the application performance in real time, gives you the leverage of catching and addressing the early warnings by passing it over to a subject matter expert or automating the resolution before these warnings significantly impact end user experience.

Customers are increasingly going mobile both with your native mobile and web applications and which present their own set of challenges. In order to deliver a five star mobile experience to your customers, you need to ensure that your mobile apps and websites are performing flawlessly, and that you can react quickly should any issue arise regardless of whether the problem is on the front-end or the back-end service.

Screen Shot 2015-10-29 at 4.39.36 PM.png

Figure 2

By using AppDynamics Mobile and Browser Real-User Monitoring (Figure 2), you can get visibility into the actual experience as perceived by the customer. So make sure you are delighting your customers throughout their journey with your products and services across channels and devices with an excellent end-user experience.


During the holiday season, application performance issues can make headlines via social media almost instantly. Retailers must address emerging application performance issues before they impact the customer. The challenge is that application issues can arise from a variety of locations–it could be the database being flooded with multiple requests due to a popular product, or capacity issues on the app servers, or even problems on a third party system linked to a website or mobile app.

Therefore, ensuring that all teams (internal and external) are looking at the same data is critical to resolving emerging problems quickly. This is where our virtual war room (Figure 3) becomes useful:  

Figure 3

There are going to be a number of days in the fiscal year when your websites and mobile apps will be experiencing peak load. Being able to compare specific time periods easily is key to being able to plan capacity requirements better for upcoming peak days during this season. The AppDynamics console includes a ‘Compare Releases’ feature that allows two time periods to be easily compared. Comparisons can be made for a number of performance metrics and can be made on Business Transactions which means you can ensure that key transactions central to buyer experience are performing as expected (Figure 4).


Figure 4

It’s likely that the main cause of performance or availability issues during this period will be capacity issues caused by unexpected buyer demand. There are two choices that retailers will have during this period to deal with this. First, you could base application workload sizing on maximum peak expected. The problem with this is that you would be paying for capacity that may not be needed and if your maximum peak estimates are incorrect, then you will likely be hit with performance issues. Alternatively, AppDynamics with our Cloud Auto-Scaling feature gives you the ability to automatically scale based on capacity and demand. So on emerging capacity issues, compute can be scaled up appropriately into your chosen cloud environment (Figure 5).


Figure 5


Holidays present a unique set of challenges and opportunities for retailers. The consequences of poor application performance are dire and can spread like wildfire. On the flip side, opportunities are ripe as retailers make 20%-40% of their annual revenue during this period. Delivering an optimal customer experience of applications during peak times means more than monitoring the performance of key business transactions; it means gaining the ability to automatically collect and correlate operational analytics to provide actionable insights to DevOps and IT operations in real-time.

AppDynamics Application Analytics is a real-time analytics solution that automatically collects and correlates application performance and business data to provide actionable insights to DevOps and IT teams, influencing business outcomes and provide optimal customer experience.

Screen Shot 2015-11-05 at 11.39.13 AM.png

Figure 6

Retail enterprises that proactively reach out to customers during performance issues are 75% more likely to keep the customers, maintain their loyalty and protect their brand. Knowing which customers had performance issues, slicing that data to know which ones had items in shopping carts, and then diving deeper to understand the population by customer type and those with greater than minimum threshold of monetary value, all in real-time can be a powerful way to minimize revenue impact and provide exceptional customer service.

DevOps and IT operations teams can quickly provide lists like these to marketing teams who can run win-back campaigns in the post-holiday season. Marketers and business owners who need visibility into unique customer data, like insight into consumer shopping behaviors, their ideal shopping preference, the device they visited from, or top features or application version(s) can rely on usage analytics to segment the customer population in real-time to build a custom dashboard.

As more customers flock to making online purchases during holiday season, retailers need a single solution to collectively analyze application performance, user experience and interaction, and business metrics, to identify and optimize their intricate interdependencies. AppDynamics is the answer to see, act, and know the impact your performance makes on your end users, all to drive your best holiday season.

Saba Anees

Saba Anees

Saba is the Content Marketing Specialist at AppDynamics. She grew up in the SF Bay Area, graduated from UC Berkeley, and joined AppDynamics' Marketing team in 2015. You can follow her on Twitter at @sabaanees.