How to define website capacity and bottlenecks to increase performance

We’re excited to have Niels Henrik Sodemann, CEO and Co-Founder of our partner Queue-it on the blog today discussing website performance trends and the importance of addressing your sites bottlenecks.

Are you defining capacity correctly?

High-interest online retail campaigns or ticketing onsales, for example, can result in website outages that are highly costly in terms of both revenue and reputation.  Even though large investments are made in infrastructure, and the nature of end-user peaks are sporadic, outages continue to plague webshops.  This can be resolved by understanding capacity in a flow-based approach, which is critical when you plan for a timed release on a webshop. 

The traditional way of defining the capacity of a web-based system is by “number of concurrent users”.  This way of defining capacity is based on the assumption that end-users are spread throughout the entire user-journey.  However, it does not take users entering and exiting the user-journey into account.  Therefore, Queue-it uses a flow-based approach to capacity, derived from Little’s law.  A flow-based approach to understanding capacity centers on the idea that, at optimum over a longer period, the throughput per time unit in a given transactional system must have average inflow equal average outflow.  If average inflow is below average outflow, the capacity is not used. If inflow exceeds outflow, the system will congest.

Queue-it has now published our 2014 Performance Trend Report, which analyzes performance trends over the past year and also provides more insight into the flow-based approach to capacity and website bottlenecks.  Queue-it specializes in helping website owners manage end-user peak situations with an online queue system, designed to provide fairness and transparency to end-users while maintaining website performance and optimizing revenue.  You can see some highlights from the report here:



Do you know your main website bottlenecks?

These 2014 trends were noted by Queue-it experts to be due to the increasing size of online campaign days, and the growth of a multi-device preference by end-users in all industries. While major bottlenecks include web services and gateways, and databases, other bottlenecks to watch out for include web servers, internet connections, and your network infrastructure. In particular, payment systems seemed to cause some main issues to website performance, which is why webshops should also analyze their third party content and subsequent fourth party calls.

Third party content added on to your website can significantly affect website performance by either slowing down load time or breaking down completely during user influxes.  Fourth party calls are essentially calls from an authorized third party tag to a myriad of unauthorized websites, each of which can have redirects, are done over SSL, and add approximately 1-2 seconds of page load time. Therefore, website performance is greatly affected by a loss of webshop control with both website load time and overall security. Webshops should analyze their third party content and fourth party calls, and assess the website performance impact.

In preparation for your next big webshop campaign, remember to check your website performance against Queue-it’s recommendations to ensure that peak situations don’t cause any unwanted issues. To read more insights, click here to download the full report, or contact Queue-it for more information.


Aisling McCaffrey

Demand Marketing Specialist at Thinkwrap
Aisling is our Demand Marketing Specialist at Thinkwrap, and loves working with both technology and humans. She studied International Business (concentrating in Marketing) and has spent several years living and working in China, mostly in Shanghai, where she became passionate about global innovation and how the use of social media changes in different cultures. Aisling likes to keep up on internet trends - from business to memes - and is always looking for new ways to learn or entertain herself.