A brand new ebook from the MIT Initiative on the Digital Economy provides insights into the 2022 MIT Chief Marketing Officer Summit held this spring. The primary message for marketing professionals: Use data, algorithms, and analytics to reach more socially connected customers.(Furthermore, PNM Group is a Digital Marketing Agency Toronto that provides Digital Marketing services! Contact us today to transform your Digital Marketing Goals.)

Social media users in the broad social and digital media networks

Consumers today choose brands by relying on a broad collection of connected social networks, including Facebook, WhatsApp, and more, and the mix changes constantly.

Aral studied 71 different items purchased by more than 30 million people using WeChat and found significant positive outcomes from adding social proof into ads, but the effect was not the same. For instance, Heineken had a 271 percent click-through rate, whereas Disney's interaction increased by 21 percent. There weren't any brands where social proof slowed the effectiveness of ads, Aral said.

You can find video analysis on TikTok, YouTube, and other social media platforms.

TikTok influencers are a big deal, especially for Gen Z. The question is whether influential videos that are viral result in sales

Research has shown that engagement and appearance of the product aren't the primary aspects -- it's more focused on whether it's in a good way or well-integrated with the video advertisement. This is especially evident when it comes to "product purchases that tend to be more impulsive, hedonic, and lower-priced," according to studies done by Harvard Business School assistant professor Jeremy Yang. At the same time, he was an undergraduate student at MIT.

Monitoring consumer engagement using machine learning

It's"the "chip and dips" challenge Marketers have struggled for years with the issue of combining goods and finding the best consumer goods to co-purchases from a vast array. With billions of choices, it is a lot of work and extended scope, and data analysis isn't easy.

Madhav Kumar, a Ph.D. student studying at MIT Sloan, developed a machine learning-based system that runs through hundreds of field scenarios to determine the most more and less effective products.

"The optimized bundling policy is expected to increase revenue by 35%," the official added.

Machine learning is used to predict results.

Most marketers are concerned with retention and revenue; however, without accurate forecasts, choices about the most effective marketing tactics are often arbitrary, said Deana Eckles, the social and digital experimentation group leader at IDE. Instead, you should update your targeting of customers using AI and machine learning to predict outcomes faster and more accurately.

Together with Boston Globe, IDE researchers used a statistical machine learning method to study a discount promotion's effects on customer behavior for the initial 90 days. The short-term forecast was as precise as the prediction made over 18 months.

"There's a lot of value to applying statistical machine learning to predict long-term and hard-to-measure outcomes," Eckles stated.

Add "good friction" to reduce AI bias.

Digital marketers frequently discuss decreasing the number of "friction" points using AI and automation to enhance customer experience. However, many marketers aren't aware that bias is a significant issue with AI, according to Renee Gosline, the lead of the Human/AI Interface Research Group at IDE. Instead of becoming caught by "frictionless fever," marketers must consider when and how friction could be beneficial.

"Use friction to interrupt the automatic and potentially uncritical use of algorithms," Gosline stated. "Using AI in a way that's human-centered as opposed to exploitative will be a true strategic advantage" for marketing.