Employee Self-Service is Not Enough!

atSpoke’s mission is to empower internal support teams to help every employee be their most productive. We’re the front door to IT, HR, and Operations teams for employees to get the services and support they need, fast.

One of the levers we use to accomplish our mission is machine learning. atSpoke automates repetitive tasks with machine learning (ML), giving support teams more time to focus on more meaningful work. By “more time” for teams, we’re not talking 1.5x more time; we’re talking 5x more time. And we do this by applying machine learning to self-service and team-assisted requests.

Machine Learning in Two Ways

atSpoke enables employee self-service through best-in-class ML models and an NLU engine pre-trained to understand employee requests and give them the right answer and action. With an integrated knowledge base and built-in self-service actions, atSpoke automatically resolves 40% of employee requests!

But that’s less than half of the problem. Employee self-service requests represent around 40% of ticket volume, comprising only 32% of how teams spend their time. What sets atSpoke apart is how we also apply ML to drive efficiencies for the other 68% of support team time.

Unlike other systems that require someone to look at every ticket and manually route it to the right destination –– atSpoke’s ML automatically triages requests to the right team or individual with over 90% accuracy. We also automatically capture the answers to follow-up questions, and kick off an automated workflow to the right team member for ticket resolution. And now, with the launch of our Integrations Command Center (ICC), where support teams can access third-party workplace tools directly within atSpoke and Slack without context-switching, we’re delivering unmatched service desk efficiencies. This is how we enable teams to resolve tickets 5x faster.

Data Methodology

To understand the effects of machine learning on team time allocation, we analyzed millions of tickets from our anonymized customer data set. We looked at the volume percentages between repetitive vs non-repetitive tickets, then determined the proportion of team time allocated per ticket segment. Finally, we calculated the changes in team time allocation when atSpoke’s machine learning is applied.

Here’s an overview of our findings.

  • As the graph indicates, support teams spend 32% of their time on employee self-service requests (pink section).
  • True efficiency for teams requires also tackling the 68% of time spent on non-self-service tickets (blue and gray sections).
  • atSpoke does this by automatically triaging tickets and follow-ups, classifying them, and now with the ICC, eliminating the wasted time that teams spend toggling between applications to resolve tickets.
  • As a result, atSpoke reduces team time on a) repetitive agent tickets from 51% to 21% and b) non-repetitive tickets from 17% to 9%.

With machine learning applied to self-service AND team assistance, atSpoke reduces team ticket time by 70%.

And this is precisely why atSpoke will continue to innovate on both end-user self-service and team efficiency.

So if you’re evaluating service desk solutions, be sure to ask vendors if they automate self-service and speed up team-assisted tickets.

Are you an existing atSpoke user?