Crushbank Insights uses Watson to glean emotion and sentiment analysis from help desk tickets, to provide information about how the tickets are being handled, for specific clients, and by individual engineers.
ORLANDO – Crushbank, which exhibited at the recent ConnectWise IT Connect event here, used the event to demonstrate a new addition to their platform. Crushbank Insights is an add-on module which gives additional information about how user experience is being handled, including the ability to determine longer-term sentiment and emotion analysis for specific clients or even for specific techs.
“We look on this as putting a data scientist on your help desk,” said Evan Leonard, Crushbank’s President.
Leonard and David Tan, Crushbank’s Chief Technology Officer, worked together at the same New York City-area MSP for a quarter of a century. After IBM opened up Watson to a partner ecosystem in 2015, they submitted the idea of utilizing Watson on help desks through a Shark Tank concept, which led to Crushbank. Their product is an application that leverages the analytics of Watson to leverage cognitive computing and make Help Desks more effective, as well as make them more interesting places to work.
Crushbank Insights is the result of back-end changes that IBM made to their platform. They end-of-lifed the Retrieve and Rank functionality that they had been using and replaced it with Discovery, a suite of APIs designed to make it easier for companies to ingest and analyze their data. Configuring Discovery allows a user to gain useful insights by enriching their own data and then delivering it in a query-able form. Enrichment refers specifically to adding cognitive metadata to the text field.
“Discovery will enrich data by tagging it for emotion, sentiment analysis, concepts and entities,” said Brian Mullaney, Crushbank’s CRO. “When we re-ingested all the data with these new enrichments, we realized there was a lot of additional value in the unstructured data that we could now present. Our basic value proposition was always around unstructured data. Using these new enrichments, Watson goes through an engineer’s notes on tickets and scrapes them. It can determine root cause from the engineer’s notes, which may be things that the engineer missed.”
“It’s hard to know exactly how a client is doing around user experience,” Leonard said. “This allows an understanding of that, and gives insight into how engineers are putting together notes and how much time they are spending on tickets. It also compares them to other engineer reports. More information is generally always better. This will help service desk managers in their reviews.”
Crushbank demoed Insights at the Connect event, and they said the results were hugely positive.
“There are a lot of dashboards out there, and they typically present structured data to save MSPs a lot of time,” Mullaney stated. “The distinction here is that this is unstructured information. It takes information buried within a ticket note, and presents it in a way that lets it be assessed and prioritized. It also allows for triage if something has been misclassified.”
“It is also really valuable for showing trends,” Leonard said. “You can see anger or negativity grow or fade over time, at the level of the client, and also for each specific engineer.”
Crushbank Insights is available now in limited release. It is slated for general availability in January 2020.
“It will be sold as an add-on module, but people who purchase before the end of the year will still get the show discount,” Leonard added.