We know that consumer data sets are a valuable tool for businesses to understand, serve and target consumers. 

When accurate, consumer data can reveal information about behaviors, messaging preferences and purchasing habits, and increase precision targeting and ROI. But recent U.S. state privacy laws — while well-meaning — are starting to prohibit the collection of “sensitive” information, specifically around details like race and ethnicity, that can actually drive a more positive and relevant experience for every consumer.

It’s argued that removing ethnicity data is necessary to protect individual privacy and prevent discrimination. That’s a valid (and warranted) critique, but there’s also another side of the coin: Race and ethnicity data is essential to accurately represent diverse audiences.

For brands, agencies and publishers, race and ethnicity data is often used to target specific groups with marketing messages and products tailored to their interests and needs. Losing this information, certain communities may be inadvertently excluded or overlooked. That could mean both lost revenues for businesses and an overall lack of representation for various marginalized audience groups.

For instance, a beauty brand may want to target a variety of consumers — men and women, darker and lighter skin tones, etc. — with specific messaging. Losing ethnicity data means you’re potentially delivering the same message to a Hispanic man as you are to an Asian-American woman. Or perhaps a hair product is designed specifically for Black consumers’ needs. 

There’s a concentrated audience to speak to there, and without ethnicity data, it’s turning intentionally helpful targeting and customization into an unnecessary array of guesswork.

This is without even digging into the existing issues around race and ethnicity data that ultimately gets worse by simply cutting it out of data sets. At Truthset, we’ve found that 74% of ethnicity data for African-Americans consumers was accurate in Q4 2022. Just 78% was accurate for Asian-Americans and 89% was accurate for Hispanic Americans. Those numbers get almost twice as bad as source data passes through onboarding and modeling processes common in programmatic advertising. 

This informs us that there’s already an issue serving these audiences. So why are we willing to exacerbate this issue further?

Having access to race and ethnicity data can actually allow advertisers, agencies and publishers to address systemic marketing discrimination, identifying where blind spots and inequality have existed in the past to try and actively avoid it going forward. 

While we can put safeguards in place to prevent data from being used improperly, you could contend there’s a greater chance of addressing societal concerns around race (and other sensitive information) by quantifying proper representation of these populations.

This is growing increasingly crucial across many demographic and psychographic lines today as the monoculture largely breaks down. Knowing what is culturally relevant on a large scale could now represent just a quarter of the U.S. population — and perhaps a small subsection of a brand or publisher’s intended audience. Access to accurate ethnicity data helps fight tired ideas, opening doors for brands and consumers that may not have had a chance to thrive in the past.

There is more work to be done to ensure that ethnicity data is collected and utilized accurately and ethically. But the idea of that hard work shouldn’t preclude the advertising industry from doing so.

Working with state and federal regulators, the entire industry can plot a way forward that benefits all parties involved and makes sure that future privacy initiatives are both consumer-safe and act in the interest of serving consumers better experiences as well.

Cable, satellite and telco operators have been gathering data about their customers for decades, but results often fall short of expectations. These days, though, so-called big data analysis seems at least poised to create real operational efficiencies and improved customer care, big-data experts and cable operators said.

Up until recently, objections to using big data analytics seemed reasonable. Concerns over potential privacy infringement were among the biggest objections.

CONSUMERS LESS WARY

Making money on big data meant selling customer information to third parties, a practice that made subscribers even more leery of providing information. At the same time, overinflated promises and expectations dashed some operators’ enthusiasm to dive into the big-data pool.

But the technology has improved, and consumers are more willing to share information. Multichannel distributors, too, are becoming more adept at using big data to enhance customer care, improve operating efficiencies and add to the bottom line.

One of Canada’s largest multichannel pay distributors is using big data to determine where it can have the biggest impact on a customer’s experience, according to the director of consumer digital services at the unnamed MSO. The amount of data any company collects can be overwhelming, so this Canadian MSO has chosen to focus on specific customer-care areas. That means a lot data may remain in storage to be used later. But it also means present-day solutions are being crafted for present-day problems.

Most phone calls into care centers are still billing-related, so using big-data analytics to reduce those calls will not only make customers happier, it can also have a direct impact on the bottom line, Nibha Aggarwal, senior director of product marketing for Amdocs, said. In one case study conducted for Amdocs last year, an undisclosed wireless provider had a chronic problem of callers repeatedly asking for credits whenever their calls were dropped.

“The carrier found they could pre-emptively fix the problem by giving small credits to customers before they called in to the call center,” Aggarwal said. “That wireless provider saw a 40% drop in its call rates simply by communicating with those customers and saying, ‘Your call was dropped and you will not be charged.’

“We are also seeing increases in customer satisfaction with some multichannel distributors who are sending step-by-step videos and instructions via text or email to customers who are trying to self-install equipment,” Aggarwal added. “It is an example of helping educate the customer before they call into the call center to ask questions.”

In another case study involving an undisclosed operator, call volumes dropped 14% when the operator sent pre-emptive emails and text messages to customers who experienced onetime price hikes for such items as late charges, Aggrarwal said.

“By sending the emails before they call in, customers know what to expect and why. This company was spending millions of dollars every year giving out credits to customers,” Aggarwal said. “When they implemented this procedure, their credits dropped 12.5%. Using big data in this manner not only improves a customer’s experience, it reduces operating costs.”

SILOS A ROADBLOCK

One big hurdle that keeps big data from reaching its full potential is the hefty task of breaking down operational silos, experts said. It’s neither easy nor expensive.

But getting rid of those silos makes it easier to be proactive and helpful. For example, in the case of a TV service outage, it would be beneficial if customers in the affected area automatically received an email or text notification. That would reduce angry calls to the call center and reassure customers that their provider is on the case and working to fi x the problem. Advances in big-data technology are making this scenario easier to accomplish, but breaking down internal operational barriers and enabling communications between those business units is necessary to take full advantage of what big data has to offer.

Companies across multiple industries, including most telecommunications companies, are moving to break down those operational silos. Terms like “single customer view” or “customer 360” have been gaining traction in marketing circles for some time. Gartner Inc., an information technology research and consulting company, presents a annual summit on how to create a customer-centric organizational culture based on a companywide CRM strategy. And vendors such as Amdocs are working with operators to integrate those operational silos to improve efficiencies and customer satisfaction.

Meanwhile, balancing the benefits of big data with concerns over privacy is an issue that continues to require operators’ attention attention, the Canadian cable executive said. Companies like Google and Facebook have paved the way to help consumers feel less threatened by sharing their personal information. But minefields still exist, and operators must tread lightly to make sure consumers aren’t uncomfortable with the information their telecom providers have about them.

“When you use that big-data information in an impersonal way, it can have a negative impact,” the Canadian MSO executive said. “When a customer calls in to complain about his or her Internet speeds, it might not be the best time to upsell them at that moment even if it makes sense on paper to sell that bigger package to them.”

EARNING THE ‘RIGHT TO SELL’

He continued: “We have to earn the right to upsell or cross-sell our customers. We want to understand and fi x their problem first. We always bring it back to the human experience. Data collection is an issue people are increasingly concerned about and we have to earn the right to use that data.”

As multichannel distributors continue to build trust with their customers, they will use big data more to sell more services, the cable executive said. Operators can learn a lot from the fumbles, foibles and successes of companies like Google and Facebook. It’s one thing to learn from companies like Google when it comes to using big data effectively. It’s another to actually put those practices to use. To be sure, analyzing big data requires manpower and the skill sets needed to take advantage of those analytics are in high demand these days. Many colleges, including the University of California at Berkeley and Queen’s University in Kingston, Ontario, are creating courses specializing in big data.

“It’s often hard to find the right people in this field,” the operator said. “Most people who deal with data are used tostructured analytics, and big data is not structured and it takes a different approach. We do a lot of internal training in the classroom and in the field, and we’re working with other companies to get ramped up in this space.”

“We are working in uncharted territory,” he said.