Transforming your behavior into digital information

COGITO – transforming behavior into digital information


Cogito is a start-up in Boston, Massachusetts that converts behavioral data from a person’s voice into usable information. For example, in January of this year, the company was awarded a $1.8 Mn Small Business Innovation Research grant from the National Institute of Mental Health to collaborate with Brigham and Women’s Hospital on using Cogito’s technology to support detection and monitoring of mental health disease. One method involves using a mobile phone app that collects behavioral data relevant to mental health assessment and putting it through computational models. The output is information provided directly to the patient-user as well as to the clinician [1].

But Cogito is not stopping only with healthcare. There are a myriad of ways that behavioral data from voices can be used, including in customer support functions in businesses. Here, Cogito is offering software that will allow customer support phone agents to collect behavioral information from a customer on the phone line and learn useful information about the customer real-time.

The idea is that the information provided by the software will allow the customer support agent to rapidly change their style to match with the customer’s preferences. In addition, the software will provide management with metrics on customer support agent performance based on the behavioral data collected from the phone calls. Together, the real-time coaching of agents and increased the management transparency should hopefully lead to increased customer satisfaction and ultimately increased customer retention [2].

Cogito has some significant clients and partners, including Aetna, Partners, and BCBS. Beyond the traction that they clearly have already achieved, I think what really strikes me is the potential of the concept and the technology – the ability to digitize behavioral information from mobile devices. Today, all kinds of information are collected from customers – buying patterns, search patterns, personal relationships and networks to name a few. But what if our phones can start to understand what makes us nervous, what makes us happy, and what makes us afraid? Or what times of the day we are stressed and what times of the day we are relaxed? How will that affect the types of advertising provided to us? Or perhaps we can imagine a world in which the user interface that we see on our phone actually changes dynamically based on behavioral data collected from our last conversation.

Obviously, companies who want to collect this data will need to receive their customer’s approval before doing so. But when was the last time you read closely through all of the permissions required by a new app or app update? If you are like me, you skim down the list and hope that everything will be okay before hitting “accept.” I think Cogito’s behavioral data platform is a digital winner, and I’m interested to see how it and other similar technologies will change the way we interact with our service providers.

[1] January 27, 2015, Company press release, “Cogito Corp Receives NIMH Grant to Improve Outcomes in Mental Health Treatment in Primary Care”

[2] Cogito Dialog brochure, accessed September 13, 2015 and available on


Quip and the end of the paper communication protocol


The Waze model – make your users do the work!

Student comments on Transforming your behavior into digital information

  1. Interesting read Angela.

    At the outset, it seems that this technology can have a lot of positive impact for corporations. The sentence that struck me the most was “The idea is that the information provided by the software will allow the customer support agent to rapidly change their style to match with the customer’s preferences.” What worries me is that employees will need to be really well versed in the technology and in human interaction to be able to match their customers preferences on the fly. Considering the pay that customer support employees receive, I think it would be hard to motivate the employees to become experts in human interaction. I see this as more of a skill a trained interrogator or detective would have.

    I do think that this technology has some great potential, especially on the national security side.

  2. I’m always very skeptical of machine learning startups that claim to have solved hard problems in human-machine interaction such as speech recognition over imperfect connections and language-based emotion detection without supplemental facial expression data. Having worked in this field, I’ve experienced how infuriatingly difficult it is to accurately elucidate complex signals from noisy, often incomplete data. Not to say this startup hasn’t made significant progress in that area; but I’d be very surprised if they have improved the state-of-the-art NLP algorithms enough to actually add value to businesses. Oftentimes its more about style over substance, especially when the underlying tech is so complicated that nontechnical customers can’t understand how it works. I’d be concerned about Cogito’s machine misclassifying a customer’s emotional state and adjusting accordingly, leading to confusion and poorer service overall.

  3. Very interesting. I actually agree that this technology could be integrated with the activities customer support employees if the application with which they interact is well designed and user friendly. For example, call center operators usually have predefined scripts that they have to follow, that indicate what to say under which circumstances. If the application is able to classify data inputs from the customers voice inflections, tone, volume, pitch, speed, etc, into simple moods (e.g. indifferent, angry, curious, etc) this new classification variable just adds another layer to the decision tree that guides the operator speech.

    Moreover, this could be integrated with voice recognition technologies (already available) to identify which words or particular scripts frequently trigger negative reactions in the consumers (we all know customer support employees often make us angry), and thus, companies could refine their customer support strategy.

  4. Very interesting Angela!

    A friend’s start-up, Soma Analytics (, has developed an app for HR departments to monitor the wellbeing of employees. This seems like an interesting application of Cogito’s software. Similar to the application in mental health treatment you described, it could gather data during phone calls to prevent burn-out for example.

  5. Great post on a very though provoking topic. The part that is indeed quite scary is consumers being able to opt-in or out of such services and data collection, or even being aware of the possibility to do so, as you point out. The collection of voice-and behavioral based information to infer moods such as anger, fear, happiness, etc… here makes me think of other companies we can see planning ahead to monetize biometric data for “mood-based” applications. Examples would include Apple and Microsoft, which have both filed for patents (Microsoft’s can be viewed here: and Apple’s here: ) for “mood-based advertising delivery” systems that rely on collecting biometric data (think of an iWatch a few generations down the road, for instance, and what this could mean for you as a consumer if advertisers were to know your mood and deliver ads based on it…) It will be interesting to see how consumers’ rights are protected when technologies like these are used for “mood collection/inference” for such ends as customer support, or marketing.

Leave a comment