Harvard Business Review September - October 2017

When Jeff Immelt announced that he was stepping down as chief executive of GE, the Wall Street view of his tenure was tepid. Analysts acknowledged his leadership through 9/11 and the Great Recession. But some also hammered him for the 30% decline in GE’s share price and noted that the company’s stock was the worst-performing component in the Dow Jones Industrial Average. Those points, while fair, obscure a bigger one: Immelt utterly remade the organization he inherited from Jack Welch. In “How I Remade GE,” which anchors our Spotlight package (page 41), Immelt calls the organization’s revamping “the most consequential makeover in its history.” The company Welch handed him was a productivity machine, a colossus of disparate businesses. But, Immelt says, “I believed that the company couldn’t simultaneously be good at…

Ron Kohavi traces his interest in online controlled experiments— the subject of his new article with Stefan Thomke—back to his days studying machine learning as a Stanford PhD candidate. That interest became a passion when he worked at Amazon, where he was the director of data mining and personalization. In 2005 he moved to Microsoft, where he’s now a distinguished engineer and the leader of the Analysis & Experimentation team, which helps the company run about 15,000 experiments annually. Annie McKee believes it’s time to finally blow up the myth that happiness doesn’t matter at the office. In her work as an executive coach and as the director of the PennCLO executive doctoral program, she’s known too many people who have sacrificed their emotional well-being to their careers, and she contends…

INTERACT WITH US The best way to comment on any article is on HBR.ORG. You can also reach us via E-MAIL hbr letters@hbr.org FACEBOOK facebook.com/HBR TWITTER twitter.com/HarvardBiz Correspondence may be edited for space and style. Unmanaged innovation frequently leads to excessive business complexity— in supply chain, sales and marketing, product development, IT, and administrative processes—resulting in higher expenses and difficulties for customers and employees. Every time customers are asked to enter the same data twice, or have to contact multiple people to get something done, it hurts the company. Every time employees can’t access important information or their decisions are derailed by silos, it hurts the company. Too much innovation can even destroy a business. This article is a healthy contrast to the notion that today’s companies need to innovate at all costs to avoid becoming…

When choosing a CEO, boards typically take into account the particular circumstances the company faces: Is it in need of a turnaround, say, or will it be scaling for growth? For a CFO position, they might ask, Are we about to do an initial public offering, or are we planning to grow by acquisition? In such cases, boards generally favor candidates with direct experience leading organizations through the situation at hand. But when hiring for and promoting people into lower-level leadership jobs, companies typically don’t pay much attention to the contextual challenges specific to the role. They tend to prefer jack-of-all-trades candidates with varied backgrounds—a tack some in HR dub the “best athlete” approach. A broad new quantitative study from the Washington-based research and advisory firm CEB (recently acquired by Gartner)…

Three years ago the Adecco Group, a Zurich-based workforce solutions company, began a pilot project in North America that uses CEB’s research to fill executive positions. Courtney Abraham, Adecco’s global head of talent strategy and development, explained the company’s motivation and results in a recent conversation with HBR. Edited excerpts follow. Why did Adecco begin using this research? Like most other companies, we had a formal talent review process in which we looked at people’s capabilities, strengths, and gaps. But it was a paper exercise, and we did it only for part of the organization. When it came to actually choosing someone for a position, it became totally subjective; it was based on the leader’s intuition. I was hired three years ago to revamp the process. Where did you start? We looked…

Sites such as Yelp, TripAdvisor, and Amazon rely on user reviews to help guide purchases, but crowdsourced reviews are open to manipulation: Restaurants sometimes offer discounts to people who will write a positive Yelp review, and some companies have offered small payments in return for reviews. New research tests the effectiveness of such incentives. Researchers studied what happened when a Chinese company offered a 25-cent credit for a review. To its surprise, the number of reviews dropped—by 30%—in the month after the payment program began. The researchers found that people with many online connections were the least likely to participate, because they feared social disapproval or having their motives questioned. And these users had previously written more reviews than other people, so the change in their behavior caused a disproportionate…