Chatbot technology brings 'self-help IT' to life at LogMeIn

In this SearchCIO Q&A, LogMeIn CIO Ian Pitt describes how chatbot technology and other cutting-edge tools help him run a lean IT department and deliver services more efficiently to the growing company.

When Ian Pitt started as CIO at LogMeIn Inc., he didn't imagine that, overnight, his user base would double in size. But that's precisely what happened as Pitt came aboard and LogMeIn, a SaaS company for remote connectivity services, finalized its merger with Citrix's GetGo Communications.

Two years later, Pitt and his 250-person team find themselves at the forefront of transforming IT and IT service management (ITSM) with AI and machine learning -- and for good reason. "We can't continue to scale a linear rate," Pitt said. "If we look at the workload we have, there are pockets of opportunity for machine learning across the board from self-help IT … to looking at how we monitor and maintain and augment our production systems."

Here, Pitt talks about how he and his team are using cutting-edge tools including chatbot technology to not only deliver self-help IT to LogMeIn's 3,500 employees, but also to refine how his IT team manages security.

Editor's note: The following has been edited for clarity and brevity.

What do you mean by 'self-help IT?'

Ian PittIan Pitt

Ian Pitt: About this time last year, we bought a company called Nanorep, an Israeli chatbot company that we use as part of our Bold360 product line. We've integrated the chatbot [technology] into our standard messaging tool, Slack. And we're using the chatbot to deliver IT help desk self-help to the user population for the people who are working for the company. So, if an employee has an issue with one of the product lines or if they just want to understand how to get things done, they don't have to open a ticket or talk to IT; they simply chat to the bot using Slack and get actual responses.

How did you get started on self-help IT?

Pitt: It began as automating the help desk, and we went down the self-help route. Clearly, this is around service management, and that leads us into some other areas of machine learning that we've got waiting in the wings, which is the whole idea of self-fixing or what we call 'level zero.'

It would be an ideal situation if a user asks for help with a problem and the system actually kicks in and fixes that problem for them. You could then take that to the next level and truly automate ITSM or fix the problem before the user knows about it. That's the true level zero that we're aiming for and hoping to achieve in the near future.

How is what you're doing different from a traditional ITSM approach?

Pitt: If you look at some of the frameworks on the market or how companies deal with customer requests and make sure that IT actually delivers service to the [organization], it tends to be very ticket-based, very rote, very people-based. We're trying to automate that as much as possible to make sure that we're able to scale the org without having to double or triple the size of the company.

We keep an eye on industry benchmarks, and LogMeIn typically operates on the lean side. I'd like to keep it that way because it allows us to deliver services but keep the costs down.

How else is IT using AI and machine learning?

Pitt: One of most exciting things we're working on at the moment is what's known in the industry as AIOps or artificial intelligence within the operations world. As you can imagine, for a company of our scale with the number of products that we operate, keeping an eye on how the system is doing through traditional tools can take a lot of effort. We've got lots of dashboards, lots of monitoring, lots of alerts.

We started to deploy machine learning against a couple of areas within the production world. One, security -- to make sure that my security team can focus on real events rather than false alarms by filtering through the information feeds and looking for bad actor patterns or getting ahead of anything that could be happening in the system. We're also looking at turning machine learning onto production operations to get a head's up of any potential failure and taking proactive action before the systems come down.

Dig Deeper on IT applications, infrastructure and operations

Cloud Computing
Mobile Computing
Data Center
and ESG