6 strategy and innovation insights from MIT Sloan Management Review

by Brian Eastwood

The 5 conditions that drive rapid innovation in a crisis
COVID-19 vaccines progressed from testing to development in two months. The National Health Service built a 4,000-bed hospital in four days. How did this innovation happen so quickly? MIT Sloan senior lecturerElsbeth Johnson and associate deanFiona Murray identify the five conditions of a crisis that boost innovation — and note how leaders can try to replicate these conditions in normal times.

Urgency. Few things trigger action better than close proximity to acute and imminent danger. In the absence of a crisis, people stick to the status quo. Leaders can counteract this tendency by assessing future threats and identifying the opportunities they present.
Prioritization. In a crisis, there’s a single problem to solve. Normally, organizations face a multitude of competing priorities. Narrowing this list to a handful of high-priority goals is a difficult but necessary exercise for narrowing a company’s focus.

Collaboration. Effective crisis response brings together diverse perspectives to tackle a specific problem. To keep these groups on task outside of a crisis, leaders should set clear constraints to ensure that all solutions are developed with the same shared objective in mind.

Experimentation. In a crisis, the risk of not finding a solution exceeds the risk of pursuing an unsuccessful solution. Under normal operations, leaders must distinguish between waste and slack. The latter enables the use of spare resources for learning and development, which will benefit a company in challenging times.

Intensity. Crisis response is all hands on deck. It’s also temporary. The best innovation efforts mimic this by combining a clear goal with a deadline. This forces teams to set a series of short-term milestones that, when reached, help everyone keep their eyes on the long-term prize.

Employees give firms high marks for communication, low marks for agility amid COVID-19
In theory, the one-two punch of a widespread shift to remote work and an economic downturn should have a negative impact on corporate culture. In fact, the opposite happened, according to MIT Sloan senior lecturer Donald Sull, who reviewed 1.4 million Glassdoor reviews written between April and August 2020.

Culture and values ratings for top companies rose during the COVID-19 era, with employees giving their companies high marks for honest and transparent executive communication, coupled with clarity about corporate strategy and overall integrity of leadership. These were the highest marks since January 2015, the beginning point of Sull’s review. Communication doesn’t typically rank as a core value for companies in normal circumstances, Sull writes, but it’s central to how a firm responds in a crisis.

At the other end of the spectrum, employees spoke most negatively about their firms’ lack of agility amid COVID-19 and political unrest. Glassdoor reviews expressed concerns about levels of bureaucracy as well as stability, risk management, and maintaining a consistent strategy. This suggests that many firms struggled to respond to the changes brought on by COVID-19, whether those changes impacted day-to-day work or the well-being of employees and their families.  

The new elements of digital transformation capabilities
More than six years after writing “The Nine Elements of Digital Transformation,” MIT Sloan principal research scientistGeorge Westerman and Capgemini Invent executive vice president Didier Bonnet revisit the topic to assess the impact of digital technologies that have emerged since 2014. The five core elements of digital capabilities for today’s enterprises take on added importance as COVID-19 continues to accelerate the shift to digital activity in order to protect workers as well as customers.

1. Customer experience improves through experience design efforts such as journey mapping and customer persona creation, customer intelligence data powering personalized interactions, and emotional engagement across a company’s value chain.
2. Operations is enhanced through the automation of core processes, the use of sensor data to inform manufacturing and maintenance processes, and the integration of operational and strategic decision-making to minimize the impact of disruption. 
3. Employee experience gets better by augmenting employee productivity, giving employees the skills they need to match the pace of change, and building agility into sourcing talent for both multi-skilled and gig workers.
4. Business models evolve through incremental enhancements, information-based “as-a-service” product offerings, and multisided platforms that bring buyers and sellers together in new markets.
5. Digital platforms serve as the foundation for digital transformation by supporting core back-office systems, external-facing applications, and data analytics.

How recommendation engines influence enterprise decision-making
Recommendation engines spurred the growth of companies such as Netflix and Spotify. Companies that are able to help customers make better choices in turn make those customers more loyal and more profitable. According to MIT Sloan visiting scholar Michael Schrage, the value of recommendation engines is the ability to influence choice rather than force a choice, and to make users confident that the recommenders have their best interests in mind.

These tools have uses beyond finding movies or songs you may like. The combination of data analytics and dynamic visualizations can offer high-potential leads to sales teams, provide prototype imagery and wireframes to user interface designers, set milestones and goals for project managers, and even help executives change the tone and direction of a company-wide email.

As with the nudges that consumer apps provide, the goal of enterprise recommendations isn’t to explicitly dictate the right answer or the only solution. Instead, Schrage writes, the aim is to uncover a wider array of choices, including some users didn’t know they had. And also like consumer apps, both the recommendations themselves and the way they’re delivered become more beneficial the more they are used.

The 6 characteristics of minimum viability for AI products
Enterprises and venture capital firms alike need to understand the unique requirements of artificial intelligence products in order to make prudent investments. Thomas Davenport, a visiting fellow at the MIT Initiative on the Digital Economy, and Glasswing Ventures managing partner Rudina Seseri identify six key characteristics of a minimum viable product in the AI market.

1. Proprietary data. Because machine learning algorithms have become commoditized, an AI product that uses a broadly available database is unlikely to provide a competitive advantage.
2. Hybrid models. When one type of data is sparse, it may be more practical to apply deep learning to that dataset and combine it with existing knowledge modeling and rules-based reasoning.
3. Integration potential. AI applications should plug into existing data records and combine with transactional systems, likely through an application programming interface, commonly known as an API. That way, an organization isn’t forced to use a separate system.
4. Domain knowledge. It’s critical for an AI application to not only solve a particular business problem but to present that solution within an industry’s commonly accepted workflows.
5. Immediate value. While it’s true that AI applications improve over time, a product’s first customers will expect value as soon as they have the product.
6. Initial performance. A product doesn’t need to beat the world; it just needs to beat the status quo. The best bet is to start with a business process that represents low-hanging fruit for potential customers.

How to identify and nurture data and analytics talent
Job titles such as data scientist or quantitative analyst mean different things to different people, and it’s likely to take years for standard job descriptions to emerge. Until then, large companies need to understand the types of analytics jobs they have and will need, Davenport writes. Projects are likely to require a range of skills — such as statistical modeling, coding, or solution development — that are unlikely to be found in a single employee or job candidate. 

- Identified seven “families” of data-focused jobs: advanced analytics, business application management, information management, insights, business intelligence and reporting, data governance, and visualization.

- Defined different roles within each job family, paying attention to purpose, accountability, and required education and experience. 

- Mapped existing employees to the appropriate role and job family.

- Forged partnerships with leading universities and research programs across Canada.

- Through in-person and virtual events as well as volunteer efforts, created a community that enables knowledge-sharing among data and analytics professionals.

Link:https://mitsloan.mit.edu/ideas-made-to-matter/6-strategy-and-innovation-insights-mit-sloan-management-review?fbclid=IwAR0pElMfbfPVa91f6KfzsZbCqyo3ZfZRHjySGBm_G2WzR9T8BKGYaq7XC6M