We want to help remote workers. Extended 90 Day Full-Feature Free Trial: Start Here.
Communication and transparency across all DevOps and IT workflows – from continuous testing to incident management – are more important than ever. Hence, the use of ChatOps. ChatOps offers the ability to execute commands and scripts, automate chatbots and communicate in the same places you collaborate with teammates. Fully remote engineering teams and globally distributed companies are using chat tools and real-time collaboration techniques such as video calls, email, phone calls and text messages to change the way engineering and IT teams work together.
And, nearly every engineer and IT practitioner is dealing with the fact that underlying applications and infrastructure are changing and becoming more complex. The rise in microservices, containers, cloud infrastructure, cloud monitoring and modern CI/CD practices are forcing teams to find ways to reduce feedback loops, shorten the testing and deployment lifecycles and get value into customer’s hands faster. ChatOps alone doesn’t create a culture of DevOps and continuous improvement, but it can certainly help get the ball rolling.
So, in this article, we’ll look at some of the value brought to the table by ChatOps today and what the future of ChatOps might look like.
ChatOps provides value in a few ways – but the top two benefits of ChatOps come in the form of speed and transparency. ChatOps isn’t a certain form of DevOps or a specific tool – it’s a way to think about using your chat tools and collaboration software to improve the way you build and maintain applications and infrastructure. ChatOps is a way to integrate and automate developer and operations workflows into your communication platforms.
Tracking the value of ChatOps will be a bit abstract. You need to look at specific ChatOps elements you implement and track delivery chain and incident management KPIs related to your new ChatOps processes. For instance, can you track a reduced number of support tickets due to chatbot automation in your knowledge base or in your website’s chat tool (e.g. Drift or Intercom)? Can you see that MTTR was reduced due to the ability to execute or open runbooks through Slack or MS Teams channels?
ChatOps can show specific value for customers and internal engineering teams - but you need to first know what you’re trying to measure and why. Unlike DevOps, ChatOps won’t be a complete change in the mindset of your engineering and IT operations teams but it will be a supplemental tool for driving better collaboration, transparency and speed.
So, what should ChatOps mean to you and your teammates? If you’re a CIO or CTO, what does it matter if you can run commands or interact with apps directly through Slack? You’re not even going to be the person on-call for these issues. But, you still need to know what’s going on with application health and customer sentiment. So, ChatOps can help drive visibility from developer to sysadmin to CIO to customer. Developers, IT managers and support engineers can work together in a unified way, communicating with teammates and running scripts in real-time, sharing visibility consistently throughout incident management workflows.
ChatOps helps you create detailed documentation without the hassle too. Now, you can easily aggregate communication streams through email, SMS, Slack, etc. and scroll through chat history to learn exactly what happened during a firefight. You can share all the data from post-incident reviews and post-mortems with executive teams (or even customers if you publicize your post-incident reviews), helping you drive continuous improvement across engineering and IT.
You can automate customer-facing chat tools to provide more valuable technical guides and API documentation for end-users while automating the pass-through of critical information to managers and executives.
ChatOps is already being utilized by numerous teams in different capacities and volumes. Some organizations place an emphasis on using ChatOps internally to speed up CI/CD pipelines, testing and bug fixes. While other teams are focused on the front-facing benefits of ChatOps for customer experience. And, there’s no right or wrong answer for how to use ChatOps alongside DevOps today – it’s all about what’s most important to you.
If transparency and ease-of-use for customers is generally more important than your ability to mobilize teams and execute incident remediation playbooks through chat, then it’s probably better you focus on chat automation in your knowledge base and technical documentation. ChatOps is all about using automation in the right way to improve human experiences across all aspects of your software. So, the best part about ChatOps is that you can use it as much or as little as you think helpful.
The future of ChatOps is interesting. Humans don’t want to talk directly to chatbots. Humans don’t look at chatbots or ChatOps processes as a way to eliminate people from their workflows. People look at ChatOps as a way to use automation to make collaboration easier and to make information more accessible. The future of ChatOps will be more of the same but enhanced with machine learning and AI. Chatbots and chat tools have been impressive at using natural language processing (NLP) and other machine learning capabilities to help engineering teams glean insights from chat and automate responses. However, there’s still a lot of evolution to come around AI and ML, which will inevitably lead to even more powerful ChatOps workflows.
For instance, in VictorOps today, we’re using machine learning and using information from your incident response history and chat history to automatically suggest incident responders or correlate similar incidents with current issues. The benefits of machine learning and AI in your ChatOps solutions can eventually lead to negative MTTA/MTTR – allowing you to identify, act and respond to incidents before they even happen.
ChatOps enhances the work you’re already doing to tighten the relationships between customers, engineering teams and executive leadership. ChatOps benefits everyone from end-user to CIO. Automation in your chat tools allows you to share information faster while giving you an easy way to take action on that information.
ChatOps allows on-call engineers, in the middle of the night, to get the resources and tools they need to communicate with their teammates and fix problems. Customers can use ChatOps processes to easily find critical information they need to setup or install your applications or services, as well as helping your business close more deals and make more money. Don’t look at ChatOps as an end-all-be-all solution for IT operations and DevOps – but think critically about how ChatOps could add value to your customer’s experience.
See ChatOps at work in a holistic incident response solution for on-call engineers and IT practitioners. Sign up for a 14-day, free trial of VictorOps to start making on-call suck less.