How Bambora used to Drive Business Value Across Multiple Teams

How Bambora used to Drive Business Value Across Multiple TeamsHow Bambora used to Drive Business Value Across Multiple Teams How Bambora used to Drive Business Value Across Multiple TeamsHow Bambora used to Drive Business Value Across Multiple Teams
"We are able to count on to ensure our performance remains reliable as we scale."
Patrik Nilsson
Stockholm, SE

About Bambora

Bambora is an international payments company based in Stockholm, Sweden that processes payments in 65 markets around the world. The company began as Beanstream in 2000 and evolved into Bambora in 2015, joining 12 other companies around the world. In July 2017, Bambora joined the Ingenico Group, one of the world’s largest payments companies, to further streamline the payment experience. Together, Bambora and Ingenico offer payment solutions that scale as companies grow their businesses globally.

In Online Payments, Every Transaction and Second Counts

Bambora is a payment gateway that is designed to help small and medium-sized businesses accept and receive payments online. The company’s platform engineering team is tasked with ensuring Bambora is operational and that transactions are processed quickly and accurately. 

With money involved and millions of transactions taking place each day, Bambora is entrusted to have a highly reliable service built on a performant production environment that delivers for their customers.

To help monitor the performance health of their cloud infrastructure and overall production environment, the platform engineering team actively manages and analyzes their logs to help with troubleshooting and debugging.

For a number of years, the team used Amazon’s Elasticsearch service, but while they were comfortable using ELK, they felt that their engineers were spending too much of their time maintaining, scaling, and backing up Kibana. In addition, the solution’s high storage costs and lack of PCI certification led Bambora to search for another solution. 

“The original solution we used was AWS Elasticsearch Service. But, we had two main reasons for finding another solution. One was the amount of maintenance for ELK that was required, as well as the associated backup and storage  costs. Another main issue was that we were moving towards PCI certification and needed a solution that could comply with PCI standards” – Patrik Nilsson, DevOps, Bambora 

Transitioning to to Centralize Log Management 

Because it’s a managed service, enabled Bambora to offload Kibana maintenance. As a result, they no longer needed to worry about updating Kibana to the newest version or scaling it manually as they grew. 

“First and foremost, having a fully updated Kibana version always running, and with all of those added insights is fantastic. Upgrading and scaling  was a significant issue for us as it put a lot of work and demands on our team. “ Patrik Nilsson

Since Bambora already used and were familiar with ELK, the move to was simple. They were able to ship a variety of log data coming from hardware encryption modules, a Cassandra database, and other external systems.

Furthermore, because already adhered to PCI compliance standards, Bambora’s engineers were free to focus on improving their product and driving their business forward rather than handling the complications of compliance on their own. Their team only needed to prevent customer credit card information from being sent through the system, while handled the rest of the PCI requirements. 

Most of all, Bambora found a partner in’s support team. 

“Our relationship with support is extremely transparent. Tickets are always being shared and there’s always a follow up from the account manager to make sure that our issues have been taken care of. This provides a good sense of ease when trying to reach out to support for any queries we may have.”  Manish Sejpal

Improving Troubleshooting with Insights 

Having a scalable, reliable, and PCI compliant ELK-based solution is what drove the Bambora team to switch to initially. But they soon found that’s advanced features such as machine learning Insights, were extremely useful for identifying and resolving performance issues. By using Insights, Bambora engineers were able to easily pinpoint the root cause of various issues, and resolve them using data collected from crowdsourced feeds.

 “Insights has provided recommendations that enabled our developers to speed up troubleshooting and resolution time for our services and the business overall. It has provided avenues where developers can quickly examine issues and then, if not pinpoint the answer, at least be pointed in the right direction much faster than they could on their own.”

Driving Value Across the Business through Log Visualization

Bambora’s support team found value in as well. In the payment industry, time is of the essence and if a transaction fails to go through or simply takes too long to process, a customer may choose to take their money elsewhere or at the very least, contact support to ensure their payments are processed. helped Bambora’s support team handle these issues through a specialized customer support dashboard which enabled them to easily track transaction flows. As a result, they were able to debug issues faster and increase customer satisfaction. This use case helps generate strong ROI for Bambora and their engineering organization.

“Using we were able to set up a dashboard where the Bambora support team could follow all operations within a transaction flow in order to answer questions like ‘why did X fail to capture transaction Y’  The dashboard shows all the info needed to quickly get an overview of the transaction insights, including, for example, truncated PAN, transaction/operation type and timing information. With this dashboard I do not need any ‘super search features’ to gather/filter data. All that is already done by the dashboard.” Magnus Beronius.

Ensuring High Availability of Logs to Deliver Observability

All in all, helps Bambora maximize product performance, resolve support issues fast, and effectively scale production. Once logs are shipped, they are automatically processed so the team can easily troubleshoot any issues that may arise.  

When we run performance tests and send 10 times, or even 50 times the amount of traffic we usually have in production, we can see that logs are piling up, but we always know that when the time comes, the needed resources will be in place to have the logs available when we need them. So in the future, when we have more and more traffic, we will be able to count on to ensure our performance remains reliable as we scale.”

Get started for free

Completely free for 14 days, no strings attached.