Achieving a 20% decrease in Load Time
Online Employment Services
Our client is a diverse group of companies in the Online Employment Services sector that have a unified purpose to help people live more fulfilling and productive working lives, and help organisations succeed on a global basis. They create world-class product technology solutions to address the needs and facilitate the matching between job seekers and hirers with exposure to 4.1 billion people globally and greater than 30% of Global GDP.
Across their employment marketplaces they receive over 375 million visits to their sites every month and have over 3 million job opportunities available at any given time with over 100 million jobseeker profiles.
Jobs listed on our client’s website contain links to the company profiles. Recently, 20% of the traffic was routed to a brand-new solution for the homepage, search results and job pages that had more prominent showing of the company reviews (1-5 stars right next to the company name). The volume of traffic to company profiles increased by 40% within a week causing performance issues due to the Content Delivery Network (global cache) failing to remember what it had previously served.
A replica of the production Content Delivery Network settings was built in a development account and investigated to steadily narrow down the root cause. Once identified, a fix was trialled and tested, with the positive response allowing a deployment to production.
The improved responses from the Content Delivery Network reduced the average page load time of job postings by 20% within hours, giving our client’s customers a better front end User Experience. The findings were also documented and shared with the other delivery teams in the organisation to ensure an improvement in future deployments.