The Need for Speed: How a Consumer Electronics Brand Sprinted to a 6x Faster Adobe Commerce

From time to time, I’ll be using this newsletter to share success stories of our ecommerce clients at Elogic Commerce. Not because I want to show off the work of my team, no (although I’ll be honest, I do take pride in it!). Rather because these stories of industry disruptors deserve to be told, shared, and adored.

This one goes to a Norwegian consumer electronics brand, Benum, who has been an invaluable client of Elogic for years. We’ve worked tirelessly to overcome business challenges together, take risks, and ultimately achieve their goals. 

Check out how the brand speeded up their Magento (Adobe Commerce) store x6!

Powering through Magento 2 performance issues — succeeding with Elogic

Based in Oslo, Norway, Benum is a leader in the distribution of professional audio equipment in the Nordics region. Benum began their online journey long ago with a small business solution (see the beginning of their success story in a related case study). But it wasn’t until 2020 that the brand started scaling and migrated to Adobe Commerce under the expert guidance of the Elogic team.

Years passed by, and Benum remained committed to improving customer experience (CX) for its B2B buyers. Together, we have revamped the brand’s UI/UX following the buyer’s journey, integrated Visma ERP, and built some disruptive custom modules. But as new technologies layered up, website performance started to slow down. And that’s when the Elogic team set out to work with vim and vigor.

Magento loading speed could reach up to 25 sec leading up to lost prospects and low sales

Little known things affecting Adobe Commerce performance

Poorly optimized database queries

Platform upgrades may cause some modules to behave incorrectly, which calls for additional optimization of database queries. In the case of Benum, these were the Mageplaza BetterSorting and Magento MSI modules which slowed down the website response. Rewriting database queries for the new platform version should do the magic. 

Insufficient resources to run the site

Benum is hosted on AWS servers, which already guarantees 99.9% uptime and easy management. This is true for low season sales, so the website should always undergo load and stress testing. In fact, the production server had only 5% of the free space at peak times, which resulted in performance issues.

Non-configured caching 

They say caching is a great way to optimize webpage speed. But a few know that it works only for retailers with a small number of pages and user groups. Benum is a large B2B store with 1.6 million pages that need to be cached and warmed up for over 120 user groups and shared catalogs. Caching for such big websites is a performance bottleneck rather than a cure.

From MySQL to MariaDM: How a simple database migration reduced page loading speed x6

Moving the database is only the first step to improve Magento site performance. Initially, Benum used an MySQL database hosted on the same production server as the website itself. It was taking about 5 seconds on average for a database to process the request (which would grow as more customers would place their orders).

To tackle the issue, the Elogic experts suggested hosting the database on a separate server so that it does not depend on the platform. Moving the MySQL database to MariaDB and hosting it on Amazon RDS (Amazon Relational Database Service) brought immediate results:

→ Amazon RDS comes with the Amazon Aurora core, which provides default security and continuous backup;

→ development team has become much more efficient thanks to serverless computing, allowing for automated replication across multiple regions and integration with other AWS services;

→ parallel queries and their caching in Redis on a new Maria DM database now load process the query in 0.3 seconds.

What’s next for Benum?

Adobe Commerce performance optimization is an ongoing process that requires constant monitoring of the system. Despite the outstanding results achieved so far, Magento site performance can still be optimized by moving Elasticsearch to a separate instance. 

Find the full case study following the link: 

https://elogic.co/projects/performance-optimization-case/?utm_source=LI&utm_medium=Newsletter 

Would you like to see more success stories like this? Would love to get your feedback on newsletters like this in comments below.