Social listening — analyzing what consumers say on social media — can serve as the map for a treasure hunt, allowing companies to tap into rich consumer insights in real time. Social listening is transforming how consumer product companies develop, market, and package their products to gain a competitive advantage — and we’re only just beginning to grasp the possibilities that machine learning advances will catalyze.

This week’s must-reads for managers: designing AI systems with customer privacy in mind, reshaping the role of CEO, what employees are saying about job automation, and how not to hate the holidays.

Though autonomous technology has a large and growing range of potential applications, it also may threaten users’ sense of autonomy and free will, or their belief that they can decide how to pursue their lives freely. But companies can create systems users don’t hate by protecting users’ autonomy, unpredictability, and privacy.

Forming a partnership between brands can help companies reap rewards — but what happens when there are more than two partners? The answer depends on the goals and the industry; each must find its own “sweet spot” of how many collaborators is just enough, but not too much, to match consumer expectations.

Online retailers must look to consumer behavior data to understand the evolution of e-commerce and improve their online presence.

Do consumers care enough about companies’ environmental and social practices to give them more business? Caesars Entertainment tested the question at one of its hotels, where one group of customers was told about its green efforts and the other group was told nothing. The casino company got encouraging results: The group who got the message spent 1.5% more. That group also recommended the hotel more enthusiastically.

The traditional method for selling — typically framed around asking the customers’ needs — is not enough. Why? Because often customers don’t recognize their core problem. To increase sales, managers should direct their teams on taking a problem-centric approach.

Companies often treat social media as the conversation that represents what consumers are saying about any given brand. But research shows that online and off-line conversations are different beasts. Even though they both drive sales, they need to be measured and managed separately.

Recommendation engines influence the choices we make every day — what book to read next, which song to download, which person to date. But digital recommendations are also a source of unintended consequences. Research shows that recommendations do more than just reflect consumer preferences — they actually shape them. Given that perfect prediction is not possible, retailers and managers must be aware of the potential discord from unintended side effects of their recommendations.

Digital customer service is becoming more widely adopted, but one place it falls short is in the language and phrases it uses. Many digital service platforms use words that alienate customers rather than engage them; selecting customer-centric language for chatbots and service platforms can make a significant difference in customer satisfaction.

A relatively small number of purchase and return metrics can accurately predict customer profitability — and the likelihood of policy abuse. By using analytics to identify the few customers who cost the company the most money by abusing return policies before they make their next purchase, companies can prevent abusive returns, avoid PR disasters, and boost their profitability.

Retail customers may accept different prices on different channels — but retailers need to manage new complexities to make it work. These include understanding what customers value in each channel and how that affects what they will pay, giving store employees the right language for talking about price differences, and working out operational challenges. Getting it right has a real payoff: Retailers that effectively price differently across all channels see bottom-line growth of 2 to 5%.

In this webinar, Cornell University’s Sheryl E. Kimes discusses the expectations disconnect between companies and customers over self-service technologies.

New research finds that stories about consumers’ positive experiences with a brand significantly increase users’ engagement with brand websites, and stories originating from consumers are especially powerful in shaping brand attitudes in social media. Indeed, companies that aren’t offering experiences that leverage consumer input in brand-related narratives are missing out on important opportunities to connect in a meaningful way with potential buyers.

Many innovative new products don’t succeed. One common reason: Companies don’t focus on understanding how customers make purchase decisions. But paying attention to how customers search for information about what to buy, and how they make guesses about details they can’t easily find, helps predict whether customers will embrace certain product innovations. Companies need to focus on innovations that customers will easily recognize or find ways to alert them to innovations they may not detect on their own.

Consumers are not running away from self-service options — just poorly implemented ones. Managers often underestimate customer’s need for employee interaction during a self-service experience, as well as customer desires for convenience and for transaction speed. “These three areas have a tremendous impact on the implementation of a self-service technology,” write the authors, “and might explain why some self-service applications have received a lukewarm reception.”

How much choice do people really want? Asking people to make their own choices requires time and focus — there’s all those options to consider. Harvard Law School professor Cass R. Sunstein writes that default rules, which establish starting points for everything from rental car agreements to health insurance plans, can save people time and keep them from being overwhelmed by too much choice.

The obvious risk of pay-as-you-wish pricing is that customers may be tempted to offer unreasonably low payment. But organizations as diverse as the Metropolitan Museum of Art, Wikipedia and Humble Bundle have figured out how to manage that risk and make pay-as-you-wish work.

New research shows that mobile advertising targeted to consumers based on their locations can be effective. This is particularly the case with customers who have shown a high level of interest in the type of product being shown to them. Researchers also think that some users might simply need more time to evaluate the trustworthiness of an app or offer — suggesting that marketers might see delayed responses to location targeted mobile ads.

Retail customers now readily use both online and offline retail channels. To thrive in this new environment, retailers need to reexamine their strategies for delivering information and products. Companies that are successful at navigating the omnichannel environment take a customer perspective and view the activities of the company through two core functions: information and fulfillment. They also consider hybrid online-offline approaches, including inventory-only showrooms and “buy online, pick up in store” options.