Salesforce is big, even by tech industry standards. It’s the biggest employer in the San Francisco area, providing a Platform-as-a-Service (PaaS) solution for over 150,000 companies worldwide, including Spotify, AWS, Toyota, and the U.S Bank. It’s not a far fetch to call it one of the biggest CRM providers in the world.
Being that the case, can you imagine a massive Salesforce Outage? Thousands of employees from all over the world turning on their computers and realizing that they can’t log in to their client relationship manager? Well, that’s exactly what happened in 2021.
The whole Salesforce platform went down for over 5 hours. Thankfully, the outage happened in early May, so for most companies, quarters-end was still a few weeks away. Still, 5 hours of downtime is nothing to sneer at.
Much like Facebook in the same year, it seems that one of the causes of the Salesforce outage was related to a DNS issue. Fortunately, since then we haven’t seen any other major scale interruption, but it does open the door to the question: are cloud providers worth the risk?
What is Salesforce Used For?
Salesforce is a company that offers PaaS solutions for several client relationship needs, including Sales Cloud, Service Cloud, Marketing Cloud, and Commerce Cloud and Platform.
While their flagship products can accommodate almost any kind of business, Salesforce also offers a platform where developers can create add-ons that are integrated with the core Salesforce products. These add-ons are hosted on their own infrastructure.
In other words, with a development team behind you, you can tailor your Salesforce experience, developing third-party applications that can later be shared on their main platform.
Keep in mind that Salesforce applications are created using declarative tools, backed by Lightning and Apex, a proprietary Java-like programming language. As for the framework, it uses its own solution called Visualforce.
While Visualforce uses an XML syntax typically used to generate HTTP pages, the fact is that anyone who wants to develop for Salesforce ecosystems has to have the skills to work with Apex and Visualforce. Learning either can be quite a challenge, which is why senior Salesforce developers have an average salary of 130,000$ a year.
Fortunately, Salesforce also has an AppExchange service where users can purchase and install third-party applications including voice solutions, surveys, analytics, meetings, and more. It’s the equivalent of an app store.
Now, as you can see, Salesforce makes for a strong case for cloud solutions. It’s sophisticated, packed with features, and can boost any company’s productivity. However, there’s also another side you need to consider.
How Many Global Outages Have Salesforce Had?
Salesforce has had 3 major outages in the last couple of years. The first one happened in May 2019, the second one in August 2019, and the third one was the aforementioned DNS issue in 2021.
Before we delve any deeper into what happened, it’s important to understand that major outages are rare. Salesforce is a massive network with redundancies, backups, and failsafe which minimize the risk of data loss or long-term suspension of service.
Having said that, Salesforce has been known for sluggish performance, disconnects, and lag, which is an annoyance at best and no different from an actual outage at its worst.
The May 2019 outage affected users who were using the Pardot SaaS marketing automation platform. Due to a bug in a database script, users were able to access all the data from their company, regardless if they should’ve had access or not.
To protect the company’s information, all affected accounts were temporarily suspended by the Salesforce team, which means that a considerable size of their user base was unable to access their systems for over 15 hours.
This happened on a Friday, mind you, so most businesses had to work during the weekend to make up for the lost time.
The August outage, while comparatively smaller, started around 9 AM and lasted for a few hours. The Salesforce team deployed a hotfix and started rebooting their servers throughout the day, solving all connectivity issues by the end of the day.
Unfortunately, there is very little information about what went wrong in this case. That’s understandable in the beginning when the engineers are diagnosing the issue, but the lack of information and lack of follow-through doesn’t build trust with a service provider.
Slack, another Salesforce product, has had many outages in the last few years, more than one would hope. That’s a huge problem considering the popularity of the platform and its widespread adoption across businesses.
Is It Worth the Risk?
A few years ago I was in charge of sunsetting our company’s CRM software after we had hired a service provider. I was about to call it a day on a Friday afternoon when I got a call from the sales manager.
The provider went offline and it was quarter’s end, we had mere hours to close any open procedures and the sales force didn’t have the tools to do so. Fortunately, I was able to spin up the servers and bring the old systems back online for a couple of days. Fast forward a few hours, and by midnight we managed to avert a crisis.
What happened then along with the Salesforce outages show us the other side of the coin: what happens when you lose remote access? It’s not just an issue with the service provider, connections can fail for any number of reasons.
The truth is that when we take our business to the cloud, we are relinquishing control, and we become dependent on our service provider. In exchange we have better control over our expenses, we have access to a plethora of services and we can build custom-tailored solutions even without a software development team.
On the other hand, cloud and platform services are quickly becoming the norm. For many companies, the prospect of outsourcing their technology solutions is a godsend. You don’t have to worry about servers, backups, upgrading hardware, or having a dedicated team of server engineers keeping the show running.
So, what’s the answer? To be perfectly honest, outages are extremely rare in this day and age, but when they do happen we should be prepared. Keep local backups of your data, develop pipelines and procedures to handle long-term outages, and hire consultants to help you shield your business against this kind of issue.
Outages can happen, it doesn’t matter if you are using an in-house server or if you are working with a service provider. In either case, the best solution is to always be prepared. Prepare for the worst and hope for the best.
