Exploring Cognitive Technology’s Value for Gathering Emotional Data

Change is happening—and fast

Previously, emotional data was hard to measure, but that's no longer the case. Getty Images
Headshot of Alicia Hatch

In June, Forrester revealed in its Customer Experience Index that satisfaction has fallen for a second year in a row, despite the fact that satisfaction is known as one of the biggest predictors of brand preference and differentiation. What’s missing? We’re not considering emotion and context, meaning we’re not thinking of customers as people.

As we move into a cognitive age, this is a big deal. Artificial intelligence is only as good as the information it’s provided. If we’re ignoring major swaths of customer data, like how they feel and what they’re trying to do, then technology can’t reach its full potential.

As marketers, we’re familiar with demographic and demand generation data, and for years have used it to make assumptions about our target audiences. Emotional data at scale has largely been out of reach. However, cognitive technologies that can help us analyze emotional data at scale are maturing—and fast.

AI has the ability to extrapolate key personality traits by scanning eye movements and can decipher subliminal facial expressions to predict a person’s honesty or passion. Imagine if you could extrapolate personality traits to forecast prospects’ emotional reactions to pitches, advertising or branding decisions. AI could take a lot of guesswork out of the process, empowering marketers to deliver the most relevant messages to the most responsive people.

Today, we aren’t setting ourselves up for success. EMarketer’s recent CX report found 22 percent of surveyed marketers used psychographic data for segmentation while only 11 percent used intent analysis. That means most aren’t paying attention to whether campaigns are delivered to angry or happy customers. Imagine the difference it would make if you as a customer were treated based on the context of your relationship?

Cognitive technologies that can help us analyze emotional data at scale are maturing—and fast.

To get better we must consider making some shifts.

Move from customer-centric to human-centric

Stop thinking of your audience as just customers and start thinking of them as complex human beings with thoughts, feelings and values. This shift will change how you categorize campaigns. In addition to the standard funnel, you should add a layer of information that speaks specifically to frustrated, angry or curious customers. This perspective might also change the services you provide around products or broaden your view to include opportunities for new products.

Think about human and machine workflow

Companies often architect human systems around technology capabilities. Instead, think about the moments in workflows when people need better information to perform. Architect your technology stack around critical data flows and build workflows to match. This approach gives you better workflows to supercharge customer experience.

Consider a customer support workflow. Bots can often handle simple support calls on their own. But when they can’t, they can provide information collected along the way to the human support agent. This optimizes the process by smoothly handing customers from one type of engagement to the next, providing seamless quality of service.

This model shakes up marketing’s old-school linear content production workflows. Emotional data allows you to measure responsiveness mid-campaign and adjust to better represent the speed of the marketplace and fickleness of customers. The more you test, the more you know which metrics matter and the more successful strategies become.

This human-to-machine interchange can work well. One-third of respondents in an Econsultancy-IBM study said they saw best results when pairing data collection with the right technology. Rewiring tactics and workflow around AI-based data can ensure strategy success. Machine-supported processes empower you to make better decisions today, and position your company to outperform competitors.

Workflows are the new organizational chart

The biggest obstacle to data-centric workflows isn’t outdated technology, but an outdated approach toward organizational structures and people management. Companies that operate in silos restrict flow of customer intelligence, thus restricting customer experience potential.

Bank of America understands customers don’t care about internal org charts and instead visualizes the whole customer journey when implementing new marketing strategies. Its virtual assistant Erica allows customers to engage based on preference to solve issues as efficiently as possible. As Erica collects more data through customer interactions, Bank of America can optimize customer experience.

Underpinning this machine-assisted workflow design is a simple principle: Look at how customer data flows into the company, then make it easy for that data to reach everyone who needs it, regardless of their place in the organization. Cross-departmental collaboration is a challenge for operations managers, and interconnecting different functions with a common thread of data is a results-driven means of doing so.

Emotional data at scale

Many brands are struggling to use AI effectively. While some understand the role it plays in data processing, few realize its potential to garner emotional data at scale. Companies that incorporate shifts in mindset—from customer to human—and workflow—from linear to nonlinear, siloed to integrated—will be able to use emotional data to leapfrog the competition.

This story first appeared in the November 26, 2018, issue of Adweek magazine. Click here to subscribe.

@aliciahatch Alicia Hatch is CMO of Deloitte Digital and a member of the Adweek Advisory Board.
{"taxonomy":"","sortby":"","label":"","shouldShow":""}