How to use ChatOps to boost business engagement across teams

Anyone who has worked in a large enterprise can tell you that collaboration across silos is difficult. That’s almost to be expected. The fact that silos exist in the first place points to functional boundaries between teams—where communications and trust break down, and where reaching out to help other teams is not so likely.

When something goes wrong and no one team is clearly to blame, finger-pointing is more likely to follow than is collaboration. Meanwhile, whatever went wrong is still an unresolved, potentially business-critical problem. When employees who need to collaborate instead stay cocooned within their areas of experience and expertise, they have become disengaged from the overall concerns of the enterprise.


When something goes wrong in a siloed organization, different teams often point to each other as the source of trouble rather than working together to solve the problem.

Here's how to use chatbots to boost business engagement across your technology teams.

Multicloud Monitoring: How to Ensure Success the First Time

Put chatbots to work

Gallup study found that just 32% of workers in the US are engaged with their day-to-day work. This means most employees are not as productive for the overall business as they could be. A lack of engagement contributes to poor collaboration, and non-engaged employees are more likely to leave the company, the study argues.

ChatOps and chatbots offer one way to solve this problem. These and related tools can facilitate engagement across silos, including development, operations, stakeholders, IT security, and whoever else might be involved in a tech project.

Cross-silo collaboration is difficult 

To understand the effects of employee disengagement with a business, it helps to understand cross-silo collaboration—when multiple teams work with each other across boundaries. Consider how enterprise IT teams typically cope with disruptions (incidents in their production environments):

An alert pops up in an IT operations center saying that customers are experiencing low performance on the site. Since this is a critical issue, Jim starts an incident war-room drill by inviting experts into a conference call.

Jim (ops): Hey, folks, our customer site is sluggish. Does anyone know why?

[Silence. Phone line crackles are heard but no answer.]

Jim: Anna, how does the server look?

Anna (infrastructure): The server’s doing great. CPU and memory are low. It’s not the server. Maybe it’s a database query issue?

Li (DBA): DB is just fine. No one has changed any DB query in the last month. It’s probably the network. Tom?

[Phone line crackles are heard again for a while, but no answer.]

Jim: Hey, does anyone know where Tom is? He was supposed to be on-call for network issues.

[Meanwhile, Tom was spotted dodging the call and pretending to be immersed in a webinar.]

Tom (network): C’mon, guys, this is the third time you are blaming network today! Our network is top-notch. Have you asked the app team?

The above is just one example of mission-critical collaborations in cross-silo settings. There are other examples. Consider the collaboration between the security team and the ops team when handling an attack or tracing a vulnerability. How do they work together? How does collaboration work among the customer success manager, customer support, and developer teams when working on a customer escalation? Is there a weak link that needs to be strengthened?

What's not working here?

Several factors can contribute to a breakdown in communication. Among them:

  • The medium isn’t efficient. Solving a major production issue often requires collaboration between 20-plus people over a single communications bridge—a challenging prospect.
  • Visibility is impaired. It is difficult for experts who participate in the collaboration to see the same picture and data as their peers. They are siloed not only in their roles and teams, but also in the information they see.
  • No shared context. With visibility impaired, the experts do not share the same context for the problem. When they lack the context, they do not have the same sense of urgency.
  • Limited recognition. The process and medium don’t expose all the hard work that experts put into resolving the issue. The experts then miss out on formal recognition from their peers—and from management.

  • Not much remains for retrospection. When the issue is resolved and the call is over, little remains for future retrospection and learning. At best, a coordinator might have taken a few notes and attached them to the ticket, but these often miss many details about the issue and the solution.

In sum, the medium and tooling currently in place do not allow for efficient collaboration. And beyond that, these elements do not foster any higher level of engagement in resolving the matter.

How people become engaged

Disengagement is a likely cause for the behavioral patterns exhibited in the example above, such as people bouncing issues from one department to another and dodging rather than jumping into action. These patterns become evident when siloed organizations try to collaborate.

Consider these engagement drivers:

  • Context and purpose. People need to understand why and how their work matters, and feel involved in the decision making.

  • Communication and recognition. People appreciate understanding what's going on, being able to voice their opinion, and being recognized for their contribution.

  • Personal growth. People want to grow personally, intellectually, and in their careers. Albert Einstein once said, “Intellectual growth should commence at birth and cease only at death.”

Consider how these drivers affect the cross-silo collaboration example described above.

In that example, IT experts try to solve a major production issue, but the communication medium (a phone call and email) inhibits the engagement drivers. A conference call with 20-plus people doesn’t allow an open communication and discussion among all participants.

Any individual contributor’s efforts may be missed. It is hard for the participants to have full visibility into everything that is discussed over an hour-long call, so they may miss the context and purpose of the requests they’re getting. And when the call finally ends, there’s not much room for personal or organizational growth among the 20-plus participants.

Fortunately, bots can help enhance employee engagement and resolve some of these issues.

Tools that can help

Imagine introducing two tool changes into the analog collaboration captured above.

First, let’s switch from team collaboration done over the phone, or with emails, or by shouting down the hallway to a digital team-chat tool such as Slack, or Microsoft Teams, or MatterMost. Such tools can improve online collaboration among people and teams. 

These tools have a chat interface but organize communication around persistent channels and topics. They provide rich functionality around the way people share, highlight, organize, and control information. So each service or application that IT or the business provides can have its own dedicated channel, and each new production issue can have its own created channel that can be archived once the issue has been resolved. This allows all participants to share the same view of the discussion and lets them go back to see the history and status of prior or ongoing discussions. It also helps by logging the discussion for future retrospectives.

Once you add chatbots to the mix, you can begin practicing ChatOps. Chatbots are services that present a chat-based interface to systems so that experts can work with them. Using this interface, experts can share data that they see with others. With ChatOps, everyone shares the same context and can act as one virtual team to solve an issue. 

Greater efficiencies

With ChatOps, collaboration becomes more efficient because:

  • A bot can start a conversation by automatically creating a new channel as an incident emerges. It can automatically invite the right people and establish the context by sharing data related to the incident.

  • Each expert can bring additional bots into the conversation. In this way, the experts can share additional information, such as monitoring status of the environment or data captured from the ticketing system or the defect management system. 

  • Experts can also take actions upon the environment by asking automation bots to perform specified tasks, and everyone else is aware of the steps taken.

  • Anyone who joins the conversation midway can go back and review the discussion to quickly get up to speed.

  • Managers who want a status update can simply open the channel to discover the severity of the issue and the estimated time to resolve it.

  • Once the issue is resolved, the entire collaboration—with all of the data used and actions taken—gets logged, where it's available for future review.

Charl Joubert, the ITSM Solution Owner of University of Pretoria explains that following the implementation of their own ChatOps practice they are now solving issues about 20% faster, and that they are capturing about 60% more valuable information that is used for learning and improvement.

Increased engagement, better 

Okay, ChatOps is clearly a modern and helpful way of doing collaboration. But does it really help improve the engagement of people? Can chatbots really transform the team’s culture?

When I worked for Hewlett Packard Enterprise, our IT organization underwent a transformation to embrace enterprise DevOps practices. As part of that process, the team contemplated restructuring and reorganizing from function-based teams (such as IT operations versus developers) to application-centric teams that brought together both IT developers and IT operation managers who worked on the same application. We also started practicing ChatOps with a bot we'd created based on Github's Hubot automation tool.

Within the collaboration tool, each application had its own dedicated channel. And in this space the bot would detect any new defects that were created in the defect management tool. The team could discuss the defect and update it from the channel, and it could use the bot to fetch monitoring information for the application to further troubleshoot the defect collaboratively.

Olivier Jacques, the DevOps strategist at the time, said that after a few months of doing ChatOps, the group no longer had to reorganize in order to improve collaboration between the dev and ops teams. They could now practice DevOps effectively. The movement of cross-silo collaborations into a digital medium and the addition of a bot provided the necessary bridge across the silos, allowing for the teams to effectively work together.

Better engagement promotes a healthy culture

The University of Pretoria has achieved significant measurable improvements to the efficiency of its IT incident war-room collaboration using chatbots. But team members were far more excited by the cultural transformation that they experienced. Teams that were previously slinging mud at each other, bouncing issues, and avoiding taking responsibility were now not just working together smoothly, but also congratulating each other on the progress that they were making toward the resolution of issues.

This is one benefit of engagement in action. Experts who once tended to avoid jumping into action now participate more actively. This occurred because their contribution (or lack thereof) became significantly more visible to the entire virtual team. At some point, even the CIO started peeking into collaborations on major issues and praising the team for its hard work.

The introduction of a chatbot into the collaboration, along with the practice of ChatOps, significantly affected the engagement of employees.

You can see the results, as measured against the engagement drivers I described above:

  • Context and purpose. The move to a digital collaboration medium allowed for an open and visible collaboration among all participants, so that everyone fully understood the mission at hand and the purpose of the activities. The addition of bots allowed the expert to communicate their pieces of the puzzle and share those to achieve a common context.
  • Communication and recognition. The highly visible collaboration carried out in ChatOps helped to make everyone fully aware of the situation and what actions had been taken, and anyone could voice an opinion at any time.
  • Personal growth. The collaboration allowed more junior employees to observe and learn from the team. People could more easily learn how to leverage the available tools by observing how others interacted with the bots. With all collaborations logged, the team could conduct retrospective assessments.

Learn more about ChatOps for collaboration

Want to know more about using ChatOps for better enterprise collaboration? Read my mini-series of blog posts on the subject. And share your own questions about experiences with ChatOps and chatbots in the comments below.

For more on ChatOps and employee engagement, drop in on my panel discussion at the Chatbot Summit in Tel Aviv, Israel, January 30-31.