DialogTech launches machine learning to automatically categorize phone calls

DialogTech makes its living by automatically identifying the advertisement that drove a phone call to a business, as well as by routing, transcribing and storing calls.

Today, the Chicago-based company is adding Dialog Analytics to its Conversation Insights call attribution platform. It says this machine learning-based analysis is the first use of that technology outside of call centers to automatically categorize calls.

CEO Irv Shapiro told me he wasn’t certain to what extent machine learning-driven categorization is used in call centers, but he added that it isn’t otherwise used in businesses where incoming calls go to local retailers or franchises.

DialogTech specializes in industries in which similar kinds of calls are generated over and over again by centralized marketing operations but received by distributed sales operations that do not utilize call centers, such as auto makers, life insurance companies and apartment rental services.

Once a client company determines the kinds of phone call categories it wants, the DialogTech data science team develops specialized algorithms for TensorFlow, the machine learning engine that Google has open-sourced.

The tuned algorithms are trained by analyzing the text transcriptions and metadata of previous calls that fall into the chosen categories, such as calls that resulted in sales leads. Shapiro said that the tone of the incoming caller’s voice will eventually be added as another factor for the machine learning analysis.

A model is developed, based on the previous in-category calls, and that is used to categorize new calls. Product Marketing Manager Jonathon Byrd told me that Dialog Analytics had a 80- to 89-percent accuracy during the beta phase, in terms of accurately and automatically predicting the correct category.

The automated categorization is generally processed overnight. This means, for instance, that an auto dealer could see the automatic categorization of, say, the hundred calls that came in yesterday and which had been attributed to two digital ads.

Because the categorization might indicate which calls appear to be leads, the auto dealer can determine which ad, channel, campaign, keywords or website content is generating more leads, as well as direct the sales team to prioritize getting back to those callers.

The categorization can also be employed to indicate the trends of the phone calls, without listening to the calls or reading the transcripts.

About The Author

Barry Levine

Barry Levine covers marketing technology for Third Door Media. Previously, he covered this space as a Senior Writer for VentureBeat, and he has written about these and other tech subjects for such publications as CMSWire and NewsFactor. He founded and led the web site/unit at PBS station Thirteen/WNET; worked as an online Senior Producer/writer for Viacom; created a successful interactive game, PLAY IT BY EAR: The First CD Game; founded and led an independent film showcase, CENTER SCREEN, based at Harvard and M.I.T.; and served over five years as a consultant to the M.I.T. Media Lab. You can find him at LinkedIn, and on Twitter at xBarryLevine.

Popular Stories

How Autodesk gave its chatbot the smarts needed to increase engagement, conversions

Amazon wants to make Alexa the 'UI for the enterprise'

Autopilot debuts real-time collaborative customer journey map

Building an Amazon Alexa Skill is so easy, Grandma can do it

Related Topics

Channel: Martech: Analytics & DataMachine Learning & Artificial IntelligenceMarketing ToolsMarketing Tools: Analytics