No Jitter is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Turning Emotion into Call Center Data

Anyone who has ever driven to an event only to waste lots of time circling to find parking will surely relate to the level of stress and emotion that can come when looking for someplace to park.


Struggles related to finding parking is exactly the problem that ParkWhiz, an online marketplace and aggregator for parking spots, aims to solve. Through the ParkWhiz app, drivers in more than 150 cities can reserve spots ahead of time -- days of circling the lot be gone! But even with a service like this, things can still go wrong, Meagan McKinnon, manager of customer experience at ParkWhiz, told me in a recent interview.

To serve customers, ParkWhiz operates a contact center with about 35 local and remote agents who handle inquiries to the company's three main channels -- phone, text, and email -- and monitor social network interactions. On average, ParkWhiz fields roughly 1,500 calls and 4,700 combined email and text interactions weekly.

For voice calls, ParkWhiz agents rely on cloud contact software from TalkDesk, McKinnon said. "Most of the calls we receive deal with user understanding or a user error. We're very fortunate to not have many calls about bugs or issues with the system, but there's always going to be problems in the field -- that's just how any business works."

Sometimes ParkWhiz will get a call from a customer who is upset because the purchased spot didn't work out as expected. "Parking is just the tiniest thing, but it can really have an impact on your day, so we really have to be there with empathy and apologize when things don't go right -- issue refunds, book a new pass for them, etc. You might be surprised how high emotions can run when you're dealing with parking and something goes wrong, and we really have to help them cope," McKinnon said.



When emotions run high, keeping tabs on customer satisfaction is especially important. But getting good customer feedback via traditional measures, such as optional IVR surveys, can be challenging. Companies typically only get a one to two percent completion rate, as Talkdesk COO Gadi Shamia told me in a recent briefing. This lack of customer satisfaction data is, in part, what led Talkdesk to create a new feedback system, called Sentiment, aimed at helping customer support teams better assess the customer experience and the level of service received.


Sentiment consists of two elements: an SMS mechanism for measuring a customer satisfaction score, or CSAT, and Mood. As you can see in the image to the right, after a call center interaction has concluded, the customer receives an SMS that asks him or her to rate quality of the service on a scale of one to five.

ParkWhiz has been one of 25 early beta users and a test case for Sentiment, using it for about the past six months, McKinnon told me. "It's really innovative," she said. "Imagine without it, if you as a customer contact a business by phone, the only way for that company to get feedback is to have the customer agree to stay on the line and listen to surveys. Think about it, by the time you get to the end of the call, you're really impatient and you've got other things to get back to. Why would you stay on the line? But with Sentiment, a text message is sent to the customer immediately when the call is over. It just requires one reply -- just type a number, hit send, and call it a day."

Before Sentiment, ParkWhiz aimed for about 10 to 12% response rates for customer satisfaction surveys. Since deploying Sentiment, ParkWhiz has far surpassed this goal, now seeing upwards of 40% completion rate, McKinnon said.



Currently, ParkWhiz is leveraging Sentiment mainly to assess the quality of customer care that its agents are providing, but McKinnon said she sees room to add more functionality moving forward. With Mood, the second piece of Sentiment, an agent would be able to input his or her perspective of the customer's mood, easily designated with an emoticon (see image, at right). The combination of the customer's reported satisfaction score and the agent's assessment of the customer's emotional state enables a broader view of the service experience, Shamia explained.


In comparing these two scores, a contact center manager would be able to tell whether agents demonstrate empathy. An agent consistently predicting how customers rate their experiences will be designated as having a high degree of empathy, for example. On the flip side, an agent who frequently misjudges caller experiences might receive empathy training to improve how he or she engages with customers, Shamia said.

Another capability Sentiment and Mood enables is the ability to program Talkdesk to take automated action upon receiving certain CSAT scores. For example, low CSAT could trigger Talkdesk to dip into Salesforce and create a high-priority callback queue for problem resolution and to head off lingering customer dissatisfaction.

"We want the agent to think about customer satisfaction throughout the call; customer feedback is no longer something you discuss with a manager once a week, it's something you need to do every call," Shamia said. "Our ultimate goal with Sentiment is not only to measure, but to help supervisors develop policy. On a case level, Sentiment can trigger interactions with customers, but in aggregate, it can help identify areas for training and improvement."

Sentiment and Mood are available in open beta for all Talkdesk customers. To start, SMS CSAT is available for inbound calls, while Mood is available for both inbound and outbound.