Enough with the excuses. Here's how we're going to stop these outages.
News · 1 July 2018Nik Storonsky
About 10 months ago, I had to write a similar letter apologising for card payments being declined for a substantial amount of time, and yet here we are again.
I'm not coming to you with excuses or meaningless apologies. I'm here to explain what happened and what we are doing behind the scenes to stop this from happening again.
Earlier this morning, we could see that a large number of payments were either timing out or completely failing. I got straight onto our engineers and we soon discovered that the issue was with a third-party payment processor.
As some of you will know, early stage fintech companies often rely on third-party companies to provide them with certain services, whether it be payment processing or international money transfers, for example.
And in our case, we work with a third-party company to process our payments. More often than not these partnerships work brilliantly without disruption, but there is always that 1% chance that they don't.
From what I understand right now, our payment processor experienced severe latency when processing data from merchants, which caused card payments to time out and then ultimately fail. This happens when data is not pulled fast enough from a merchant.
The good news is that our partner has since fixed the issue and all card payments should be processing again. If you made a payment before and it failed, this will show up as a pending transaction which will be automatically reverted back to your account. This should only take about an hour or so, but it can take as long as 72 hours in some instances.
There is a glimmer of hope, though. Over the last nine months or so, we have been building our own in-house payment processor to help reduce any downtime and allow us to have full control over any issues like this in the future. We're on track to deliver this project by the end of the year and proceed to roll it out across Europe.
I know that one of the main frustrations right now is the length of time it takes to speak with customer support. As I'm sure you can imagine, with anywhere between 300,000 and 400,000 daily active users, our support team simply become overwhelmed and therefore it takes a little time for us to work through the queue.
The solution here is not hiring hundreds of more support agents, but instead building as much of our tech in-house as possible and improving our communications on these issues.
Our in-house communications team do an amazing job, but they are incredibly limited right now due to an old-age engagement system. Right now the process is to update all social media channels as well as place a banner on the main screen in the app.
However, we are working on a new engagement system that will allow us to instantly send out push notifications and emails to all customers in minutes, not days. Unfortunately, we neglected this system in the past and that was a mistake.
These are teething problems. Please bear with us.
While we would love to be up and running 100% of the time, I plead for a little understanding. Revolut is doing amazing things for banking and for consumers, but we're not immune to issues along the way.
I hope the explanation above gives you a little context as to what went wrong and demonstrates that we are actively working on numerous projects to reduce these kind of problems in the future.
We're not going to point fingers or deflect the blame. The responsibility here lies with Revolut and these kind of issues will only force us to do better.
Founder & CEO