Average Response Time Calculation: How to Test Response Time

How is Average Response Time Calculated?

Average Response Time is calculated as an average of durations of the web transactions simulated on the target website at a given time interval:

Average Response Time = ∑ time of transaction duration / the number of transactions started

What is a Transaction?

A transaction is defined as a sequence of completed operations, performed on a web resource by a visitor, or a sequence of HTTP/S requests and replies. The time of transaction duration is the time elapsed, from the moment the transaction is started, to the moment the transaction is completed. For example, a transaction can be defined as sequence of operations, such as loading a web page, logging in to the website, navigating to another webpage, and finally, submitting a web form.

User Behavior Profiles and Delays

Configuring user behavior profiles provides you with the ability to simulate how typical users interact with your websites or web applications. For more information about user behavior delays, see our User Behavior Profile Knowledge Base article.

Web Page Tasks

When creating web page tasks, the LoadView platform provides Normal and Custom user behavior profiles in the Test Scenario page. Choosing the Normal option will slow page interaction and add random delays (from 3 to 6 seconds) between actions to simulate how real users navigate through your website. Selecting the Custom option allows you to set minimum and maximum delays, from 0 to 30 seconds. Setting the delay with a minimum and maximum of 0 seconds will execute the test scripts as quickly as possible. This option is designed for stress testing to see how your system responds.

Web Application Tasks

For web application tasks, user behavior delays will be included in the transaction duration. Once your device has been created, you can customize the profile according to the needs of your specific device. Just like the user behavior profiles for web pages, he same user behavior profile options, Normal and Custom, are provided, but include further configuration settings to simulate specific user actions, such as mouse movement speed, mouse click speed, and typing speed, based on the requirements for your specific web application task. For more information about configuring web application tests, see our Web Application Load Test Knowledge Base article.

Why is Average Response Time Important?

Your users expect your websites and applications to always to be available and perform without experiencing any setbacks, regardless of the time or day. Applications or sites that take too long to load or respond slowly can cause your users to quickly become frustrated and abandon the tasks and actions they were intending to carry out, which could potentially result in lost sales. Even just a second or more of a delay can make the difference between keeping a user from bouncing off your site or application.

Factors that Impact Response Times

While executing performance tests can help pinpoint where issues and bottlenecks are occurring, remedying slow response times can be difficult to carry out. Slow response times can be an indicator of more complex issues, and could be due to an overloaded server, hosting provider issues, or even client-side issues. While there are numerous factors that can contribute to slow response times, the list below comprises of some of the more common causes.

Complex Environments

Complexity is one of the key factors that lead to slow response times.  Many of today’s websites and applications rely on a variety of third-party services, networks, technologies, platforms, etc., making it difficult to precisely determine what specific component or elements could be the cause.

Heavy Web Pages

Furthermore, today’s website and application frameworks can lead to “bloated” web pages that are too big in page size, contain too much JavaScript, or simply are not optimized appropriately, leading to sluggish page performance. While it is important to build a website that is eye catching to users, web developers must carefully balance website and application content with user experience, and how each of them impacts the overall response times. It can be easy to get carried away with building a content-heavy site, but if you find users are bouncing early and often, it is time to consider pulling back on the amount and type of content and optimizing the pages for a better user experience.

Scalability

Scalability is another key factor that can contribute to slow response times, especially during peak traffic times and busy online shopping periods, like Black Friday/Cyber Monday. When traffic increases suddenly, that can result in the server receiving more requests than it can handle, leading to a performance bottleneck as resources become uses. Performance testing can help identify gaps in infrastructure to ensure you sites and applications can scale with the demands of users. When demand is high, your server should be able allocate, or ramp up, the necessary resources appropriately to handle the demand, as well as ramp down when demand decreases.

Continuous Monitoring with Dotcom-Monitor

Once your website or application is ready and has been pushed to production, it is important that you continually monitor load times and response times to ensure users are not receiving a poor user experience. Setting up monitors will give you the insights and data you need to ensure your websites, web applications, and any third-party services and APIs are continually up and running. And if they are not, you and teams get alerted immediately, so you can remedy the issues before a larger percentage of your users are affected.

The Dotcom-Monitor platform allows you to monitor from 30 locations around the world and provides a variety of solutions and features, such as alerting options, schedules, filters, integrations, and more, for complete end-to-end monitoring for all your needs. Learn more about the Dotcom-Monitor solutions.