This Company Accidentally Deleted Its Clients' Data. Here's How It Won Them Back.
Grow Your Business, Not Your Inbox
Bad things happen even to good companies. No matter what the reasons for a crisis are, it can affect and even destroy a brand’s reputation. In such a situation, a company must react.
There are two possible ways for a company to deal with a crisis situation. The first is to hide and keep silent. Do not admit the problem, do not comment, do not say anything. Just wait and hope your clients or the media don’t notice what’s happened. That might work, but be careful. If the problem catches the attention of social media and people see you’ve yet to react, any future comments might prove useless.
The second way is to face and attempt to repair the issue. This is tough. It requires split-second reactions to comments and mentions throughout the web, especially on social media. But if you handle it properly, you’ll save your brand’s reputation and show the public you’re committed to helping them.
GitLab’s recent handling of a crisis is a perfect example of the latter approach.
January 31 was a nightmare for GitLab. Some of its clients’ data from a primary database service was accidentally removed. In addition, the backup didn't work and GitLab lost access to a huge amount of data. As a result, GitLab.com service was not available for 18 hours. The brand’s reputation was at serious risk.
Just imagine: You’re a social media manager for a company that has to store developers’ code. Your clients trust you with the most valuable thing -- their intellectual property. And your company lost it. In GitLab’s case, this accident affected more than 5,000 clients’ projects worth billions of dollars. What’s more, the list of GitLab’s clients include such organizations as IBM, Sony, NASA, Alibaba, Invincea, Jülich Research Center, O’Reilly Media, Leibniz-Rechenzentrum (LRZ) and CERN.
Can you imagine dealing with thousands of furious customers? Can you imagine how quickly the negative information was distributed? What would you do in this situation?
Let’s look at how GitLab handled this crisis, step by step.
Step 1: Take immediate action.
Thinking carefully and analyzing the situation is dramatically important in any crisis situation. It’s all about making the right decisions within the right timeframe. Remember, you can hide or you can admit the problem publicly. If you choose the second way, go to social media first, as Gitlab did.
Gitlab posted on Twitter and admitted that it was performing emergency maintenance in order to get the service back up and running.
We are performing emergency database maintenance, https://t.co/r11UmmDLDE will be taken offline— GitLab.com Status (@gitlabstatus) January 31, 2017
Step 2: Keep your audience updated.
It’s a common mistake to wait until the problem is resolved to communicate with your audience. If you decide to go public with the problem, make sure everyone believes that it’s under control and that you’re diligently working to rectify the problem. Update your customers about the status of the problem, including what’s going on and how long it will take to fix.
The day after the incident, GitLab provided an explanation and informed everyone what went wrong and how it planned to resolve it. The company published an explanation on its website and posted it on Twitter.
So, what happened yesterday? Here's what went wrong and what we're doing to fix it: https://t.co/vWqs0TmrkL— GitLab (@gitlab) February 1, 2017
From that moment on, the company kept its audience updated regularly, communicating and answering all questions. It even created the hashtag #HugOps so it wouldn’t miss important comments and mentions. This move gave GitLab’s audience the opportunity to find all the answers they needed. Instead of talking to each other and gossiping, the audience conversed with GitLab and participated in solving the problem.
Step 3: Monitor social media and forums.
In crisis situations, it’s also important to keep up with how your customers are talking about the issue across the web. Track the conversations and react with information and updates if your number of complaints grows. There are plenty of tools that will help you spot all important mentions or comments on blogs and news sites.
Gather all your brand mentions in one place and sort resources by authority and number of comments per resource. Monitor the situation on social media to discover, follow and collaborate with people and companies who mention you or your brand. You can sort them by number of followers and decide which requires your immediate attention.
Prioritize all resources and people that mentioned your brand, figure out who’s creating the biggest buzz and connect with the authors of mentions as soon as possible.
Remember, reacting quickly is the most important thing in such situations, so use the tools that can help you react immediately.
Step 4: Be transparent.
Sometimes the problem can’t be fixed in one day. You’ll need some time. While you are fixing the situation, do not disappear. Stay with your audience. They want to know what’s going on and if you don’t give them answers, they will draw their own conclusions.
From a marketing point of view, GitLab did a great thing -- it stayed in touch with its audience throughout the crisis situation.
GitLab maintained an online presence by:
- Creating a Google document with live notes
- Creating a special hashtag, #HugOps, to monitor the entire situation
- Streaming on YouTube to discuss the problem-solving process
- Encouraging anyone to join and help -- the problem became essential for the entire professional community. A lot of those professionals were glad to help.
After the problem was fixed, GitLab thanked everyone for their help and created a special logo to express its appreciation.
This smart move transformed the situation entirely. The audience thanked GitLab for being so transparent and keeping customers informed.
Step 5: Make it work for you.
Ten days after the incident, GitLab published a post in which it described in detail what it had learned from this situation. Why did it do that? The audience was ready to understand the problem. Give the audience an explanation!
Looking at an engagement chart, we can see that there were two posts that got the most attention during the crisis period. Both were explanations. The first was posted just after the incident and the last one was posted after everything was fixed.
GitLab’s employees wrote the second post so that their colleagues in the industry could learn from this occurrence. They admitted their mistakes, provided their action plan, and explained how they will prevent these mistakes from happening again. This move transformed a negative experience into a positive one.
And, in the end:
- GitLab’s reputation was saved.
- Clients were happy.
- The issue was resolved.
- And they created an excellent example of crisis management for the entire marketing community.
Step 6: Humor always helps.
No matter how bad the situation is, never lose your sense of humor (situation depending). It usually works to diffuse hot tempers. Yorick Peterse was the guy who pressed the wrong button and deleted all the data. No doubt he was terrified when he realized what he did, but still he found the courage to laugh about it.
If you have problems with your database not scaling due to too much data, I can probably help you :troll:— Yorick Peterse (@YorickPeterse) February 1, 2017
Crisis situations can happen at any company. No one is safe. The thing is, audiences will focus on your timing, your actions and how you resolved the particular situation.
Why was GitLab’s case so remarkable?
- It admitted the mistake publicly and asked for help.
- It was transparent and responded effectively, and pointed out that it was doing everything possible to solve the problem.
- It took it as a professional challenge, and that’s why it got help from the entire community.
- It was professional, accepted responsibility and solved the problem.