How to Use AI at Your Service Desk

Throughout the year, I have been writing articles for this audience on the use of AI in the service desk space. Why? Because AI is not going away. With all of the worldwide investment, capability and untapped future potential, if you’re not seriously looking at how it can help your service desk, you will soon fall behind.

The biggest question I’m asked is, “Where should I start?”  Let’s answer that!

Start with what your company might already be doing or considering. First, understand where your company is on their overall AI journey. Have they purchased tools?  If so, can they integrate into your ITSM? If they haven’t purchased an enterprise tool yet, are they looking at tools? Do they have a governance policy? Does leadership see a potential use case on your desk? Understanding this first (and articulating your particular needs) will help you to see how long the road could be to get going.

Next, build your business case as to how AI can help on the service desk to get people’s attention. What’s your end game? How do you think that AI can help your operation? How can you quantify that? In order to get money, you need to have a solid business case as to what AI can accomplish and how that tangibly benefits your company.

There’s lots of resources available from HDI and other organizations as to different use cases for AI on the service desk to guide you. Each desk is unique so you have to think about how these various use cases might work in your particular situation.

OK, now onto a quick sanity check…if you build it, will they come? As you’re considering various use cases, seek out some honest answers from those who will be using your AI to validate that it’s helpful to them. Nothing is worse than spending lots of money on a really cool system that no one uses.

As you work through this, you may have employees raise concerns about AI taking their jobs. Be transparent with them as to your goals and expectations and how they could be potentially impacted. In my company, we’re growing. By using AI, we don’t need as many new hires to keep up with our growth. Maybe in your case, you may decide not to backfill a position once the business case is proven in practice. The important thing here is to be honest and upfront with your team. If you’re not, you may not see a big AI adoption rate, which hinders success.

Armed with a business case and an understanding as to what your company is doing, you are now in a position to either work with your leadership to satisfy your business case or maybe strike out on your own to meet with vendors, if you’re allowed to do so. 

If you start meeting with vendors, beware of the hype! There are vendors out there that will tell you that AI will automatically resolve 90% of your incidents or more. That may be true, but that assumes that all of your systems have open APIs, that your cybersecurity and legal teams are OK with AI accessing them and that you have the resources available to integrate them to the tool.

When you’re talking to a vendor, challenge what you hear. And don’t forget to talk to their customers, too.

If you were able to strike out on your own and meet with vendors, get some sample boilerplate agreements to review with your legal and cybersecurity teams. If they’re good to proceed, pilot each of them for as long as possible, negotiating set-up cost and what happens if you were to walk away after the pilot. You don’t want to get locked into a system that does not work as you expect.

In my company, we’re coming up on two years of using an AI product for my service desk.  Overall, I’m happy. However, I’m also continuing to stay on top of trends, review potential new use cases and look for even more ways to become efficient leveraging this new technology.

Good luck on your AI journey!