What to Expect From Artificial Intelligence 03-01

To understand how advances in artificial intelligence are likely to change the workplace — and the work of managers — you need to know where AI delivers the most value.

Major technology companies such as Apple, Google, and Amazon are prominently featuring artificial intelligence (AI) in their product launches and acquiring AI-based startups. The flurry of interest in AI is triggering a variety of reactions — everything from excitement about how the capabilities will augment human labor to trepidation about how they will eliminate jobs. In our view, the best way to assess the impact of radical technological change is to ask a fundamental question: How does the technology reduce costs? Only then can we really figure out how things might change.

To appreciate how useful this framing can be, let’s review the rise of computer technology through the same lens. Moore’s law, the long-held view that the number of transistors on an integrated circuit doubles approximately every two years, dominated information technology until just a few years ago. What did the semiconductor revolution reduce the cost of? In a word: arithmetic.

This answer may seem surprising since computers have become so widespread. We use them to communicate, play games and music, design buildings, and even produce art. But deep down, computers are souped-up calculators. That they appear to do more is testament to the power of arithmetic. The link between computers and arithmetic was clear in the early days, when computers were primarily used for censuses and various military applications. Before semiconductors, “computers” were humans who were employed to do arithmetic problems. Digital computers made arithmetic inexpensive, which eventually resulted in thousands of new applications for everything from data storage to word processing to photography.

AI presents a similar opportunity: to make something that has been comparatively expensive abundant and cheap. The task that AI makes abundant and inexpensive is prediction — in other words, the ability to take information you have and generate information you didn’t previously have. In this article, we will demonstrate how improvement in AI is linked to advances in prediction. We will explore how AI can help us solve problems that were not previously prediction oriented, how the value of some human skills will rise while others fall, and what the implications are for managers. Our speculations are informed by how technological change has affected the cost of previous tasks, allowing us to anticipate how AI may affect what workers and managers do.

Machine Learning and Prediction

The recent advances in AI come under the rubric of what’s known as “machine learning,” which involves programming computers to learn from example data or past experience. Consider, for example, what it takes to identify objects in a basket of groceries. If we could describe how an apple looks, then we could program a computer to recognize apples based on their color and shape. However, there are other objects that are apple-like in both color and shape. We could continue encoding our knowledge of apples in finer detail, but in the real world, the amount of complexity increases exponentially.

Environments with a high degree of complexity are where machine learning is most useful. In one type of training, the machine is shown a set of pictures with names attached. It is then shown millions of pictures that each contain named objects, only some of which are apples. As a result, the machine notices correlations — for example, apples are often red. Using correlates such as color, shape, texture, and, most important, context, the machine references information from past images of apples to predict whether an unidentified new image it’s viewing contains an apple.

When we talk about prediction, we usually mean anticipating what will happen in the future. For example, machine learning can be used to predict whether a bank customer will default on a loan. But we can also apply it to the present by, for instance, using symptoms to develop a medical diagnosis (in effect, predicting the presence of a disease). Using data this way is not new. The mathematical ideas behind machine learning are decades old. Many of the algorithms are even older. So what has changed?

Recent advances in computational speed, data storage, data retrieval, sensors, and algorithms have combined to dramatically reduce the cost of machine learning-based predictions. And the results can be seen in the speed of image recognition and language translation, which have gone from clunky to nearly perfect. All this progress has resulted in a dramatic decrease in the cost of prediction.

The Value of Prediction

So how will improvements in machine learning impact what happens in the workplace? How will they affect one’s ability to complete a task, which might be anything from driving a car to establishing the price for a new product? Once actions are taken, they generate outcomes. (See “The Anatomy of a Task.”) But actions don’t occur in a vacuum. Rather, they are shaped by underlying conditions. For example, a driver’s decision to turn right or left is influenced by predictions about what other drivers will do and what the best course of action may be in light of those predictions.

The Anatomy of a Task

Actions are shaped by the underlying conditions and the resolution of uncertainty to lead to outcomes. Drivers, for example, need to observe the immediate environment and make adjustments to minimize the risk of accidents and avoid bottlenecks. In doing so, they use judgment in combination with prediction.

Seen in this way, it’s useful to distinguish between the cost versus the value of prediction. As we have noted, advances in AI have reduced the cost of prediction. Just as important is what has happened to the value. Prediction becomes more valuable when data is more widely available and more accessible. The computer revolution has enabled huge increases in both the amount and variety of data. As data availability expands, prediction becomes increasingly possible in a wider variety of tasks.

Autonomous driving offers a good example. The technology required for a car to accelerate, turn, and brake without a driver is decades old. Engineers initially focused on directing the car with what computer scientists call “if then else” algorithms, such as “If an object is in front of the car, then brake.” But progress was slow; there were too many possibilities to codify everything. Then, in the early 2000s, several research groups pursued a useful insight: A vehicle could drive autonomously by predicting what a human driver would do in response to a set of inputs (inputs that, in the vehicle’s case, could come from camera images, information using the laser-based measurement method known as LIDAR, and mapping data). The recognition that autonomous driving was a prediction problem solvable with machine learning meant that autonomous vehicles could start to become a reality in the marketplace years earlier than had been anticipated.

Who Judges?

Judgment is the ability to make considered decisions — to understand the impact different actions will have on outcomes in light of predictions. Tasks where the desired outcome can be easily described and there is limited need for human judgment are generally easier to automate. For other tasks, describing a precise outcome can be more difficult, particularly when the desired outcome resides in the minds of humans and cannot be translated into something a machine can understand.
This is not to say that our understanding of human judgment won’t improve and therefore become subject to automation. New modes of machine learning may find ways to examine the relationships between actions and outcomes, and then use the information to improve predictions. We saw an example of this in 2016, when AlphaGo, Google’s DeepMind artificial intelligence program, succeeded in beating one of the top players in the world in the game of Go. AlphaGo honed its capability by analyzing thousands of human-to-human Go games and playing against itself millions of times. It then incorporated the feedback on actions and outcomes to develop more accurate predictions and new strategies.

Examples of machine learning are beginning to appear more in everyday contexts. For instance, x.ai, a New York City-based artificial intelligence startup, provides a virtual personal assistant for scheduling appointments over email and managing calendars. To train the virtual assistants, development team members had the virtual assistants study the email interactions between people as they schedule meetings with one another so that the technology could learn to anticipate the human responses and see the choices humans make. Although this training didn’t produce a formal catalog of outcomes, the idea is to help virtual assistants mimic human judgment so that over time, the feedback can turn some aspects of judgment into prediction problems.

By breaking down tasks into their constituent components, we can begin to see ways AI will affect the workplace. Although the discussion about AI is usually framed in terms of machines versus humans, we see it more in terms of understanding the level of judgment necessary to pursue actions. In cases where whole decisions can be clearly defined with an algorithm (for example, image recognition and autonomous driving), we expect to see computers replace humans. This will take longer in areas where judgment can’t be easily described, although as the cost of prediction falls, the number of such tasks will decline.

Employing Prediction Machines

Major advances in prediction may facilitate the automation of entire tasks. This will require machines that can both generate reliable predictions and rely on those predictions to determine what to do next. For example, for many business-related language translation tasks, the role of human judgment will become limited as prediction-driven translation improves (though judgment might still be important when translations are part of complex negotiations). However, in other contexts, cheaper and more readily available predictions could lead to increased value for human-led judgment tasks. For instance, Google’s Inbox by Gmail can process incoming email messages and propose several short responses, but it asks the human judge which automated response is the most appropriate. Selecting from a list of choices is faster than typing a reply, enabling the user to respond to more emails in less time.

Medicine is an area where AI will likely play a larger role — but humans will still have an important role, too. Although artificial intelligence can improve diagnosis, which is likely to lead to more effective treatments and better patient care, treatment and care will still rely on human judgment.

Different patients have different needs, which humans are better able to respond to than machines. There are many situations where machines may never be able to weigh the relevant pros and cons of doing things one way as opposed to another way in a manner that is acceptable to humans.

The Managerial Challenge

As artificial intelligence technology improves, predictions by machines will increasingly take the place of predictions by humans. As this scenario unfolds, what roles will humans play that emphasize their strengths in judgment while recognizing their limitations in prediction? Preparing for such a future requires considering three interrelated insights:

1. Prediction is not the same as automation. Prediction is an input in automation, but successful automation requires a variety of other activities. Tasks are made up of data, prediction, judgment, and action. Machine learning involves just one component: prediction. Automation also requires that machines be involved with data collection, judgment, and action. For example, autonomous driving involves vision (data); scenarios — given sensory inputs, what action would a human take? (prediction); assessment of consequences (judgment); and acceleration, braking, and steering (action). Medical care can involve information about the patient’s condition (data); diagnostics (prediction); treatment choices (judgment); bedside manner (judgment and action); and physical intervention (action). Prediction is the aspect of automation in which the technology is currently improving especially rapidly, although sensor technology (data) and robotics (action) are also advancing quickly.

2. The most valuable workforce skills involve judgment. In many work activities, prediction has been the bottleneck to automation. In some activities, such as driving, this bottleneck has meant that human workers have remained involved in prediction tasks. Going forward, such human involvement is all but certain to diminish. Instead, employers will want workers to augment the value of prediction; the future’s most valuable skills will be those that are complementary to prediction — in other words, those related to judgment. Consider this analogy: The demand for golf balls rises if the price of golf clubs falls, because golf clubs and golf balls are what economists call complementary goods. Similarly, judgment skills are complementary to prediction and will be in greater demand if the price of prediction falls due to advances in AI. For now, we can only speculate on which aspects of judgment are apt to be most vital: ethical judgment, emotional intelligence, artistic taste, the ability to define tasks well, or some other forms of judgment. However, it seems likely that organizations will have continuing demand for people who can make responsible decisions (requiring ethical judgment), engage customers and employees (requiring emotional intelligence), and identify new opportunities (requiring creativity).

Judgment-related skills will be increasingly valuable in a variety of settings. For example, if prediction leads to cheaper, faster, and earlier diagnosis of diseases, nursing skills related to physical intervention and emotional comfort may become more important. Similarly, as AI becomes better at predicting shopping behavior, skilled human greeters at stores may help differentiate retailers from their competitors. And as AI becomes better at anticipating crimes, private security guards who combine ethical judgment with policing skills may be in greater demand. The part of a task that requires human judgment may change over time, as AI learns to predict human judgment in a particular context. Thus, the judgment aspect of a task will be a moving target, requiring humans to adapt to new situations where judgment is required.

3. Managing may require a new set of talents and expertise. Today, many managerial tasks are predictive. Hiring and promoting decisions, for example, are predicated on prediction: Which job applicant is most likely to succeed in a particular role? As machines become better at prediction, managers’ prediction skills will become less valuable while their judgment skills (which include the ability to mentor, provide emotional support, and maintain ethical standards) become more valuable.
Increasingly, the role of the manager will involve determining how best to apply artificial intelligence, by asking questions such as: What are the opportunities for prediction? What should be predicted?

How should the AI agent learn in order to improve predictions over time? Managing in this context will require judgment both in identifying and applying the most useful predictions, and in being able to weigh the relative costs of different types of errors. Sometimes there will be well-acknowledged objectives (for example, identifying people from their faces). Other times, the objective will be less clear and therefore require judgment to specify the desired outcome. In such cases, managers’ judgment will become a particularly valuable complement to prediction technology.

Looking Ahead

At the dawn of the 21st century, the most common prediction problems in business were classic statistical questions such as inventory management and demand forecasting. However, over the last 10 years, researchers have learned that image recognition, driving, and translation may also be framed as prediction problems. As the range of tasks that are recast as prediction problems continues to grow, we believe the scope of new applications will be extraordinary. The key challenges for executives will be (1) shifting the training of employees from a focus on prediction-related skills to judgment-related ones; (2) assessing the rate and direction of the adoption of AI technologies in order to properly time the shifting of workforce training (not too early, yet not too late); and (3) developing management processes that build the most effective teams of judgment-focused humans and prediction-focused AI agents.

Reproduced from MIT Sloan Management Review.

How to Monetize Your Data 01-22

These days, most companies are awash in data. But figuring out how to derive a profit from the data deluge can help distinguish your company in the marketplace. 

Image credit : Shyam's Imagination Library

The possession of rich amounts of data is hardly unique in today’s world. Indeed, data itself is increasingly a commodity. But the ability to monetize data effectively — and not simply hoard it — can be a source of competitive advantage in the digital economy.

Companies can take three approaches to monetizing their data: (1) improving internal business processes and decisions, (2) wrapping information around core products and services, and (3) selling information offerings to new and existing markets. These approaches differ significantly in the types of capabilities and commitments they require, but each represents an important opportunity for a company to distinguish itself in the marketplace.

Theoretically, companies can pursue more than one approach to data monetization at the same time. In practice, adopting each approach requires management commitment to specific organizational changes and targeted technology and data management upgrades. Thus, it’s best to identify your most promising opportunity and start there. In doing so, you will enhance your data in ways that will accelerate subsequent efforts related to the other approaches. More importantly, you’ll build your company’s capacity for monetizing its data.

Improving Internal Processes

Using data to improve operational processes and boost decision-making quality may not be the most glamorous path to monetizing data, but it is the most immediate. Executives often underestimate the financial returns that can be generated by using data to create operational efficiencies. Companies see positive results when they put data and analytics in the hands of employees who are positioned to make decisions, such as those who interact with customers, oversee product development, or run production processes. With data-based insights and clear decision rules, people can deliver more meaningful services, better assess and address customer demands, and optimize production.

When Satya Nadella became CEO of Microsoft Corp. in February 2014, he urged employees to find ways to improve the company’s processes with data. Within sales, executives believed that, with the right tools and systems, they could improve the productivity of their salespeople by 30%. To do so, Microsoft’s sales leaders sought to deploy tools that would help salespeople spend more of their time engaging with customers — and in more effective ways — by arming them with key computed insights such as how likely a sale is to close and when.

To deliver actionable insights, sales executives first had to define shared concepts (for example, what is meant by “a lead”). They then needed to locate data sources that could be used to calculate performance. They quickly learned that sales data was located in too many different systems to easily create a comprehensive snapshot of a salesperson’s business. Within a year, they created a new, integrated customer system that could produce 360-degree views of Microsoft’s relationships with corporate customers, including what those customers bought, what issues they encountered, and how the company engaged with them.

The new system saved 10 to 15 minutes per sales opportunity by eliminating the need for Microsoft salespeople to manually search for and prepare data. The system also helped sales executives more accurately manage their pipelines; it used predictive analytics and machine learning to compute the likelihood of a successful sales engagement based on data that the salesperson provided about an opportunity. For example, buying and deploying enterprise software is complex and often requires a partner’s involvement, so the system may calculate a higher likelihood for success when customers already have partners involved. Information about an opportunity’s likelihood of success, along with suggestions on how to advance engagements along the sales pipeline, helped salespeople prioritize their leads and act in ways most likely to achieve their goals. Over time, Microsoft salespeople learned how to forecast more accurately (for example, the accuracy of forecasts regarding global accounts has risen from 55% to 70%), which has led to better sales-pipeline data and, in turn, improved pipeline management.

Wrapping Information Around Products

Most companies have opportunities — often quite significant ones — to enrich their products, services, and customer experiences using data and analytics, a phenomenon that we call “wrapping.” Companies are wrapping their offerings with data to escape commoditization and satisfy increasingly hard-to-please customers — with the goals of generating sales increases, higher prices, and deeper customer loyalty. FedEx Corp. was an early exemplar of wrapping when it introduced online package tracking as a free service in the 1990s. Now examples abound as companies bundle reporting, alerts, and other information to add value to products ranging from credit cards to health monitors.

Wrapping is a creative exercise in which companies identify what problems their customers have and then find ways to solve those problems using data and analytics. For example, Capital One Financial Corp., a diversified bank based in McLean, Virginia, learned that many of its credit card holders are concerned about fraudulent transactions but find the task of examining every charge to be tedious. So the company helps customers identify fraud more easily and more quickly by displaying merchant logos and maps with each transaction in online statements. The visual cues jog cardholders’ memories about whether they made a purchase or not. As a result, customers are more satisfied with the credit card and more likely to use it more often.

Johnson & Johnson has discovered the value of providing pattern identification to users of its health-monitoring products, including those for diabetics. The company offers its OneTouch Verio Sync Meter customers historical reporting on their blood glucose levels along with tools to help them understand patterns of changes. The reporting is intended to help customers identify the possible causes for the glucose level variations and thus identify behavioral changes that can result in healthier living.

Wrapping activities are best viewed as extensions of a company’s product management processes. This means offering data and analytics to customers at the same level of quality as the core product. Doing so requires comparable levels of scrutiny and control. Most companies don’t manage and cannot deliver data and analytics in this way. In fact, exposing data to customers could reveal quality problems and a lack of analytical sophistication. Thus, in most cases, wrapping requires companies to “up their game” in their information capabilities so that wrapping doesn’t damage their reputation or undermine their value proposition. This effort may entail heavy investment in data-quality programs, advanced computing platforms (for instance, Hadoop), or data-science talent.

Selling Data

Many executives are eager to sell their company’s data, convinced that it has inherent value and can generate important new revenues for the company. We caution that selling represents the hardest way to monetize data, mainly because it requires a unique business model that most companies are not set up to execute. Yet it can be done to potentially great effect under the right circumstances.
State Street Corp. is a Boston, Massachusetts–based financial services company that reported $10.4 billion in 2015 revenue. It provides products and services to institutional investors such as mutual funds, corporate and public retirement plans, and insurance companies.

In 2013, State Street announced a new information-business division called State Street Global Exchange that would combine existing State Street data and analytics capabilities with new research to develop information-based solutions that clients would be willing to buy independently of the company’s core services. State Street established a new division for the information business in recognition of its unique business model needs — something the company had not done in 30 years.

Even though it started out as a discrete unit, State Street Global Exchange focused on developing products that were tightly associated with State Street’s core business. For example, State Street is one of the largest administrators of private equity assets, which means that it collects data about the financial capital that is not noted on a public exchange; this kind of data is of great value to markets that require an accurate representation of the private equity industry. State Street Global Exchange appreciated that the data was not automatically monetizable. Executives secured permission from 3,000 private equity clients to aggregate and anonymize that data — and then created an index that conveyed the financial performance of the private equity industry.

State Street leaders realized that they would need an entirely new operating model to support the information business. For one, sales processes had to change because, although State Street Global Exchange often sold to State Street clients, a buyer of Global Exchange products was frequently a different person or cost center than the kind of buyer traditional State Street products attract. In addition, the information business required salespeople with different selling experience and skills in selling stand-alone data and analytics-based products.

State Street understood that establishing an information business is hard and takes time. State Street Global Exchange had to learn to achieve balance between maintaining key ties with State Street (to create benefits from being a part of the larger organization) and responding quickly to new markets and new needs. Executives believe that State Street Global Exchange is gaining significant traction with its clients — and that their commitment will pay off. But we caution that such a model is not easy to replicate. Other companies should think carefully about the operational capabilities, investment, and commitment required to successfully sell data.

The Importance of Accountability

Chances are you have two major obstacles to monetizing your data. The first is the accessibility and quality of your data. Our research has found that only about a quarter of companies offer employees and customers easy access to the data they most need. You can’t monetize data no one can use.
The second obstacle is lack of accountability. All three approaches to data monetization require committed leaders who can redirect the behaviors of employees to deliver an important new value proposition.

Your inclination may be to solve the data quality issue first with big investments in new infrastructure. We propose that addressing the second issue of accountability will create urgency and commitment to addressing data quality issues — and so we recommend starting there.
Data monetization through process improvement requires strong process leaders. These leaders systematically use data to analyze the outcomes of existing processes and test hypotheses about proposed improvements. At Microsoft, for example, sales managers designated specific people to reshape and institutionalize new ways of selling. Process leaders are ultimately responsible for the design of best practices, the capture of the right data, the availability of tools, and the training of all staff regarding how to use data to do their jobs.

Data monetization through wrapping requires strong product leaders. These leaders treat the data that accompanies a core product or service much like any other product innovation — they hold it to the same quality standards. At Capital One, product leaders know the value of adding a data or analytics feature to a credit card because they predict — and then track — the lift in revenue from the information as well as the cost of providing it. Product leaders assemble teams to design experiments and methodologies that help analyze the impacts of information features and make appropriate adjustments.

Monetizing data by selling it requires a strong business-unit leader. That leader, in turn, must assemble a team that can launch and grow what is for most companies a new line of business. The head of that business will start by ensuring the value of the data and related services to potential customers. But the business head and his or her team must also design data, analytics, and dashboards to monitor the business and enable rapid response to new business opportunities.
Each of the data-monetization strategies requires new processes, new skills, and new cultures to generate maximum returns. Companies with data-monetization experience have learned that it is insufficient to simply put data and tools into the hands of employees. Microsoft refined goals, cleaned up data, honed reports and algorithms, grew talent, and changed habits. Capital One and Johnson & Johnson reshaped product-management talent, platforms, and capabilities. State Street redesigned its organization and created a new profit formula that would generate stand-alone revenues from information.

Impressive results from data monetization do not transpire from single “aha” moments. Instead, they stem from a clear data-monetization strategy, combined with investment and commitment.

View at the original source

The Next Wave of Business Models in Asia 01-03

There’s a new generation of sophisticated entrepreneurial growth companies in Asia — and they’re competing by  reconfiguring business models. 

Image credit : Shyam's Innovation Library

THE SOUTH KOREAN television drama “My Love From the Star” features a dashing, 400-yearold alien who falls in love with an actress. The plot isn’t difficult to grasp. It’s essentially a boy-meets-girl story with an interstellar twist. The global appetite for such Korean entertainment — movies, TV shows, and music videos — has exploded in recent years. For non-Korean-speaking viewers, subtitles are crucial to the experience. Enter Viki Inc., a company that hosts content for streaming and provides subtitles and closed captions. Viki both eliminates language barriers and introduces the content to an otherwise unserved audience. Traditionally, subtitles are created by a bilingual translator hired by the producer or broadcaster. But the process is expensive and slow to scale. To overcome these challenges, Viki developed a business model leveraging a community of more than 150,000 volunteers.

This model allows Viki to crowdsource subtitles for Asian content in numerous languages.1 Viki rewards volunteers with gamified badges, the ability to view videos not otherwise available in their region, early access to new shows, and an advertising-free, high-definition experience of the content. As it happens, the market is ripe for services like Viki’s. In fact, the combination of rapidly increasing internet video adoption rates and a greater appetite for foreign content — both in Asia and globally — has become a big opportunity for Viki, which was acquired by Tokyo-based Rakuten Inc. for a reported price of $200 million in 2013.2

Two Business Model Waves From our perspective, as a consultancy that analyzes business model innovation across the globe, Viki’s story exemplifies a larger trend playing out in Asia. We see Viki as an archetype of a new generation of companies emerging in Asia and leveraging business model innovation to drive growth in the region. But to understand this type of business model innovation in its proper context, it’s important to understand Viki’s forerunners. Our research into business model innovation in Asia uncovered two distinct, yet overlapping, waves of innovation: one decades old and still going, and one that includes Viki and is evolving now. (See “About the Research.”) The first wave, as we call it, primarily exploited differences in labor and other input costs between developed and developing markets. By contrast, 

the second wave is driven primarily by business model innovation and typically leverages new technology. These companies are characterized by extensive and often radical reconfigurations of the profit formula, resources, processes, and relationships within a broader stakeholder ecosystem. They may have a sophisticated global orientation from the start; for example, in Viki’s case, the company was “born global,” beginning as a class project by graduate students who were studying in the United States but who later moved the company to Singapore.3

The First Wave The first wave of contemporary business innovation emerged in Asia during the post-World War II era. It became a tidal wave from China following Deng Xiaoping’s 1978 “open door” policy, which changed the competitive landscape of global manufacturing. Another sea change involved opening and deregulation in India in the 1990s, which transformed the global services industry. These changes have been explored before. For example, C.K. Prahalad and Stuart L. Hart pioneered research on the opportunities at the “bottom of the pyramid” in emerging markets.4 John Seely Brown and John Hagel III also investigated product and process innovation practices from Asia.5 In a similar vein, Vijay Govindarajan and Chris Trimble’s

ABOUT THE RESEARCH To chart the emerging wave of business model innovation in Asia, we applied a four-step process. First,  we reviewed 27 lists of the most innovative global companies, compiled during the past year in publications such as Forbes, MIT Technology Review, and Fast Company. These lists emphasize measures such as  patents filed, revenues, and spending on research and development. To complement the lists, we sought input from our venture capital arm, Innosight Ventures, on relevant startups. In addition, we reached out through our networks to identify other innovative companies.

All told, this resulted in a list of roughly 200 companies operating in Asia. In step two, we filtered this list based on business models. We evaluated the business models on several parameters, including the extent to which they address important and unmet customer jobs to be done; the complexity of resource and process configuration; the novelty of the profit model; the leveraging of technology; the reduction of barriers to adoption through simplicity, convenience, accessibility, and affordability; and the applicability to emerging market circumstances.

We also gauged the business models by the companies’ demonstrated growth so far and by their potential, in our view, to become globally disruptive. In step three, we synthesized these findings and categorized the companies based on their business models. In step four, we selected representative companies for both waves, and we interviewed company executives to develop further insights into their business models. An overwhelming majority of the companies we reviewed were incumbents or low-cost innovators that we categorized as first-wave innovators. However, we saw evidence of a distinct category that we subsequently labeled second-wave innovators. The success of these companies is built upon the reconfiguration of their business model components. From the second-wave companies, we selected two illustrative examples out of a handful.

What Successful Project Managers Do 06-09

What Successful Project Managers Do

Traditional approaches to project management emphasize long-term planning and a focus on stability to manage risk. But today, managers leading complex projects often combine traditional and “agile” methods to give them more flexibility — and better results.

Coping with frequent unexpected events requires an organizational culture that allows the project manager to exercise a great amount of flexibility. Here are two examples of advanced organizations that took steps to modify their cultures accordingly.In today’s dynamic and competitive world, a project manager’s key challenge is coping with frequent unexpected events. Despite meticulous planning and risk-management processes, a project manager may encounter, on a near-daily basis, such events as the failure of workers to show up at a site, the bankruptcy of a key vendor, a contradiction in the guidelines provided by two engineering consultants or changes in customers’ requirements.

Such events can be classified according to their level of predictability as follows: events that were anticipated but whose impacts were much stronger than expected; events that could not have been predicted; and events that could have been predicted but were not. All three types of events can become problems that need to be addressed by the project manager. The objective of this article is to describe how successful project managers cope with this challenge.

A group of 23 project managers who had come from all over NASA to participate in an advanced project management course declared mutiny. They left the class in the middle of the course, claiming that the course text, based on NASA’s standard procedures, was too restrictive for their projects and that they needed more flexibility. With the blessing of NASA’s top leadership, the class members then spent four months conducting interviews at companies outside of NASA. This led to a rewriting of numerous NASA procedures. 

Among other things, NASA headquarters accepted the group’s recommendation to give NASA project managers the freedom to tailor NASA’s standard procedures to the unique needs of their projects. A similar movement to enhance project managers’ flexibility occurred at Procter & Gamble, where the number of procedures for capital projects was reduced from 18 technical standards and 32 standard operating procedures to four technical standards and four standard operating procedures.

Concurrent with these changes at NASA and P&G, a heated debate emerged within the wider project management profession regarding the need for flexibility, as opposed to the traditional approach, which emphasizes that project success depends on stability. According to the traditional approach, project success can be achieved by focusing on planning and on controlling and managing risks. Although the popularity of this approach has sharply increased across industries, research covering a wide variety of projects consistently reveals poor performance. A large percentage of projects run significantly over budget and behind schedule and deliver only a fraction of their original requirements.

The other side in this debate is best represented by a newer project management approach popular within the software industry. Called the agile method, it asserts that project success requires enormous flexibility throughout the project’s life. However, even proponents of the agile approach acknowledge that this approach is best suited to small projects and teams.

Our studies, employing experiential data collected from more than 150 successful project managers affiliated with more than 20 organizations, indicate that today’s successful project managers cope with unexpected events by a combination of the traditional and agile approaches, assuming four roles. (See “About the Research.”) Two of the roles are intention-driven and two are event-driven, with each role assumed on its own time schedule throughout the life of the project. The first role, developing collaboration, is performed early on during the project. The second role, integrating planning and review with learning, is performed periodically. The third role, preventing major disruptions, is performed occasionally. The fourth role, maintaining forward momentum, is performed continuously.5(See “The Four Roles of the Project Manager.”)

About the Research

In recent years, many researchers have concluded that one reason for the widespread poor statistics about project results is the wide gap between research and practice.i The overall objective of our research was to develop a practice-based theory of project management.ii To this end, we used three complementary approaches to collect firsthand data on the practices of successful project managers. Believing that management is best learned by emulating exemplary role models, we focused our studies on a selective sample of the best practitioners in their respective organizations.

Our first approach consisted of field studies and structured research tools, particularly 40 interviews (two to four hours each) and 20 observations (four hours to a week each) of practitioners in the following organizations: AT&T, Bechtel (the San Francisco-based construction and civil engineering company), DuPont, General Motors, IBM, Motorola, PPL Electric Utilities (an electric utility company based in Allentown, Pennsylvania), Procter & Gamble and Turner Construction Company (a construction services company headquartered in New York City).

For our second approach, we convened project teams and facilitated reflective dialogues in which participants shared their stories and practices from recent projects. We collected most of the cases, stories and practices through our role as the facilitators of the project management knowledge-development and -sharing communities in three organizations. In this capacity, Laufer and Hoffman worked for five years with NASA, Laufer and Cameron worked for three years with P&G and Laufer and Russell worked for two years with Boldt (a construction services company based in Appleton, Wisconsin). Project managers from the following organizations participated in these community of practice meetings: AeroVironment (a technology company based in Monrovia, California), Boldt, The Johns Hopkins University Applied Physics Laboratory, Lockheed Martin, NASA, Procter & Gamble, Raytheon and the U.S. Air Force.

To make sure that the principles we developed were a valid interpretation of the stories we had collected, we adopted a third approach — testing our interim results in real-life situations. Through consulting engagements with four project-based organizations — Boldt, Parsons Brinckerhoff (the multinational engineering and design firm headquartered in New York City), Skanska (the Scandinavian construction and property development group) and Turner Construction — we validated and refined our understanding and developed the four-role framework presented in the current article. 

We then tested and refined this framework in our work with the Boldt project management knowledge-development and -sharing community. The model presented in this article is the result of a final refinement process, which included a series of interviews with 10 project managers and 10 senior managers. We held these interviews (two to three hours long) with a carefully selected group of practitioners from companies that represented a variety of industries, including Cedars-Sinai Medical Center, NASA, PricewaterhouseCoopers, P&G and the U.S. Air Force.

1. Develop Collaboration

Since project progress depends on the contribution of individuals who represent different disciplines and are affiliated with different parties, collaboration is crucial for the early detection of problems as well as the quick development and smooth implementation of solutions. The importance of collaboration can be demonstrated by the following two examples in which projects failed.

Tim Flores analyzed the causes for the different outcomes of three Mars exploration missions initiated by NASA’s Jet Propulsion Laboratory: Pathfinder, Climate Orbiter and Polar Lander. Although all three projects were conducted under the same guiding principles, were of comparable scope and shared many elements (even some of the same team members), Pathfinder was a success, whereas the other two missions failed. Flores expected to find that the Pathfinder project differed from the other projects in a variety of factors, such as resources, constraints and personnel. 

Although this was true to some extent, he found that the primary factor distinguishing the successful mission from the failed missions was the level of collaboration. The Pathfinder team developed trusting relationships within a culture of openness. Managers felt free to make the best decisions they could, and they knew that they weren’t going to be harshly punished for mistakes. That trust never developed in the other two projects.

A different NASA project, the Wide-Field Infrared Explorer (WIRE) mission, was designed to study the formation and evolution of galaxies. Its telescope was so delicate it had to be sealed inside a solid hydrogen cryostat. When, shortly after launch, a digital error ejected the cryostat’s cover prematurely, hydrogen was discharged with a force that sent the Explorer craft tumbling wildly through space, and the mission was lost.

Jim Watzin, a project manager at NASA and a member of the WIRE project team, had this to say regarding the official report that NASA issued following the WIRE failure: “WIRE failed because people could not or would not communicate well with each other. … Individuals ... simply were uncomfortable allowing others to see their work.” Watzin added: “The real [lesson] from this loss is that any team member that does not participate as a true team player should be excused [from the project].”

In the next two examples, project success can be attributed to the project manager’s deliberate attempt to develop collaboration. (Note that in the discussions that follow, we use only the project managers’ first names.)

Allan, the payload manager for NASA’s Advanced Composition Explorer project at the Jet Propulsion Laboratory, has described how he developed trust between his team and the 20 groups of scientists developing instruments for the project, who were based at universities throughout the United States and Europe. Allan devised a three-stage plan. First, he selected team members who could operate in a university environment — people who knew when to bend or even break the rules. Second, he relocated his JPL team to a university environment (California Institute of Technology), recognizing that it might be difficult to develop an open, flexible culture at JPL. Third, he came up with an uncommon process for interacting with the scientists.

The challenge, with regard to interaction, was getting the scientists to regard his JPL team as partners. Having dealt with NASA before, they tended to believe that someone coming from JPL would demand a lot of paperwork, lay out sets of rules to be followed and expect things to be done a certain way. In fact, many of the scientists weren’t sure they should share with Allan’s team the problems they were encountering along the way — problems that could slow down the project’s progress.

When unexpected events affect one task, many other interdependent tasks may also be quickly impacted. Thus, solving problems as soon as they emerge is vital for maintaining work progress.

The primary role of Allan’s team was to review the development of the instruments, and Allan believed that the best way to do this was by focusing on trust and convincing the scientists that his team was there to help them solve their problems. To facilitate this, Allan and his team of five to eight members traveled to each university and stayed on site for an extended period of time. By spending days and nights with the scientists and helping them solve their problems — not as auditors but as colleagues — the JPL team gradually became accepted as partners.

Most projects are characterized by an inherent incompatibility: The various parties to the project are loosely coupled, whereas the tasks themselves are tightly coupled. When unexpected events affect one task, many other interdependent tasks are quickly affected. Yet the direct responsibility for these tasks is distributed among various loosely coupled parties, who are unable to coordinate their actions and provide a timely response. Project success, therefore, requires both interdependence and trust among the various parties.

However, if one of the parties believes that project planning and contractual documents provide sufficient protection from unexpected problems, developing collaboration among all the parties may require creative and bold practices.

This was the case in a large construction project that P&G launched at one of its European plants. After the contractor’s project manager, Karl, brushed off numerous team-building efforts, Pierre, the P&G project manager, finally found an opportunity to change Karl’s attitude. Three months into construction, the contractor accidentally placed a set of foundations 10 inches inside the planned periphery and poured about 600 lineal feet of striped foundation in the wrong place. Instead of forcing the contractor to fix his mistake and start over — a solution that would have damaged the contractor’s reputation and ego — Pierre chose a different approach. 

Through several intensive days of meetings and negotiations with the project’s users and designers, he was able to modify the interior layout of the plant, thereby minimizing damage to the users without having to tear down the misplaced foundations and hurt the project’s schedule. The financial cost of making the changes incurred by the contractor’s mistake was significant, but the loss in reputation was minimal. As a result, Karl gradually embraced Pierre’s working philosophy — namely, “If they fail, we fail.” The realization that the organizations involved in the project are all interdependent led to the development of a collaborative relationship.

2. Integrate Planning and Review With Learning

Project managers faced with unexpected events employ a “rolling wave” approach to planning. Recognizing that firm commitments cannot be made on the basis of volatile information, they develop plans in waves as the project unfolds and information becomes more reliable. With their teams, they develop detailed short-term plans with firm commitments while also preparing tentative long-term plans with fewer details. To ensure that project milestones and objectives are met, these long-term plans include redundancies, such as backup systems or human resources.11

One key difference between the traditional planning approach, in which both short- and long-term plans are prepared in great detail, and the rolling wave approach becomes evident when implementation deviates from the plan. In the traditional planning approach, the project team attempts to answer the question: Why didn’t our performance yesterday conform to the original plan? In the rolling wave approach, project managers also attempt to answer the question: What can we learn from the performance data to improve the next cycle of planning? In particular, they attempt to learn from their mistakes — to prevent an unexpected event from recurring.

Successful project managers do not limit the learning process to the planning phase but also use it for project reviews. For example, after a review session in the midst of a project at NASA’s Goddard Space Flight Center, Marty was a frustrated project manager. The existing review process may have fulfilled upper management’s need to control its operations, but Marty felt it did not fulfill his team’s need to learn. Therefore, he modified the process to give his team the best input for identifying problems and the best advice for solving them. This meant doing away with the usual “trial court” atmosphere at NASA review sessions, where team members’ presentations were often interrupted by review board members’ skeptical comments and “probing the truth” questions. In its place, Marty developed a review process that provided feedback from independent, supportive experts and encouraged joint problem solving rather than just reporting.

The first thing Marty did was unilaterally specify the composition of the review panel to fit the unique needs of his project, making sure that the panel members agreed with his concept of an effective review process. The second thing he did was change the structure of the sessions, devoting the first day to his team’s presentations and the second day to one-on-one, in-depth discussions between the panel and the team members to come up with possible solutions to the problems identified on the first day. This modified process enabled Marty to create a working climate based on trust and respect, in which his team members could safely share their doubts and concerns. At the end of the second day, the entire panel held a summary meeting. It was agreed that the review session had been a big success. In fact, other NASA project managers quickly adopted Marty’s process, including it in their managerial tool kits.

Successful managers of more traditional projects, such as designing and building manufacturing facilities, also practice learning-based project reviews. P&G has replaced review panels composed of external experts or senior managers with peer-review panels. These last four to eight hours and follow a simple protocol: First, the project team concisely communicates its technical and execution strategies, and then the floor is opened to all the invited peers for comments, critique and clarifying questions. Out of the numerous notes documented throughout the review process, five to 10 “nuggets” usually emerge that the project team uses to improve the technical, cost and scheduling aspects of the project. Sometimes, the invited peers even take one or two of the “nuggets” back to their own projects.

3. Prevent Major Disruptions

In their book Great by Choice, Jim Collins and Morten T. Hansen describe one of the core behaviors of great leaders as “productive paranoia.” Even in calm periods, these leaders are considering the possibility that events could turn against them at any moment and are preparing to react. Similarly, successful project managers never stop expecting surprises, even though they may effect major remedial changes only a few times during a project. They’re constantly anticipating disruptions and maintaining the flexibility to respond proactively. The following two examples illustrate that, when convinced that a change is unavoidable, a successful project manager acts as early as possible, since it is easier to tackle a threat before it reaches a full-blown state.

NASA’s Advanced Composition Explorer project, discussed earlier, was plagued from the start with severe financial problems arising from internal and external sources. Internally, the development of the nine scientific instruments led very quickly to a $22 million cost overrun. Externally, the project, which was part of a larger NASA program, inherited part of a budget overrun in an earlier project. As a result of these internal and external factors, the ACE project experienced frequent work stoppages, forcing the manager to constantly change his contractors’ and scientists’ work priorities.

Don, the project manager, believed that without immediate changes the project would continue down the same bumpy road, with the likely result that cost and time objectives would not be met. To prevent this, he made an extremely unpopular decision: He stopped the development of the instruments, calling on every science team to revisit its original technical requirements to see how they could be reduced. In every area — instruments, spacecraft, ground operation, integration and testing — scientists had to go back and ask such questions as: How much can I save if I take out a circuit board — and how much performance will I lose if I do take it out?

At the same time, Don negotiated a new agreement with NASA headquarters to secure stable funding. To seal the agreement, he assured them that, by using descoping tactics, the project would not go over budget. With the newly stable budget and the project team’s willingness to rethink its technical requirements, the ACE project gradually overcame its technical and organizational problems. Completed early and below budget, the spacecraft has provided excellent scientific data ever since.

To keep costs under control, Terry decided to have two contractors compete for the final contract. Terry quickly realized that both contractors were approaching the development too conservatively and that unless he took a more radical approach, the project would be canceled again. Therefore, he told the contractors to completely disregard the military standards and adhere to only three key performance parameters. One of the contractors, Lockheed Martin, took this directive seriously and changed its approach dramatically. It decided to build the missile fuselage not out of metal but out of composites. And to accomplish this, it found a company that made baseball bats and golf club shafts. 

The company had never built a military product, but it knew how to weave carbon fiber and was open-minded. Following trials with several prototypes, this company was able to manufacture a product of the highest quality. Lockheed Martin transformed this small company from a baseball bat provider to a cruise missile supplier, which led to Lockheed Martin winning the contract — as well as to remarkable cost reductions.The second example of preventing a major disruption from occurring took place during the Joint Air-to-Surface Standoff Missile, or JASSM, project. In this case, the Pentagon had decided to make another attempt to develop JASSM after the first attempt was aborted due to a cost overrun of more than $2 billion. The original project manager for the second attempt was dismissed in midcourse due to poor performance, and a new project manager, Terry, replaced him.

4. Maintain Forward Momentum

As noted earlier, when unexpected events affect one task, many other interdependent tasks may also be quickly impacted. Thus, solving problems as soon as they emerge is vital for maintaining work progress. As Leonard R. Sayles and Margaret K. Chandler wrote in their 1971 book Managing Large Systems, “In working to maintain a forward momentum, the manager seeks to avoid stalemates. Another penalty for waiting is that in a good many situations, corrective action is possible only during a brief ‘window.’ … The heart of the matter is quickness of response.” In a study of project managers on construction sites, it was found that they addressed (not necessarily solved) 95 percent of the problems during the first seven minutes following problem detection.

In a recent knowledge development meeting, a group of 20 project managers at The Boldt Company, a construction services company based in Appleton, Wisconsin, focused on how best to cope with unexpected events. It became evident that most of the managers employed three complementary practices: hands-on engagement; frequent face-to-face communication; and frequent moving about.

Regarding hands-on engagement, one project manager, Charlie, said that to solve problems he often engaged in activities such as making phone calls, convening urgent meetings and taking trips to local retail stores to purchase missing parts. Documenting the time it took him to resolve 10 recent problems, Charlie reported that three were resolved within 30 minutes, three within 60 minutes, and three in less than one day; one problem took two days until it was resolved. Charlie also said that, because of his quick responses, he made one mistake. However, he was able to quickly repair its damage the following day. The entire group at Boldt agreed that maintaining forward momentum was more important than always being right.

The second practice, frequent face-to-face communication, was described by Matt, one of the project managers, in terms of “daily 10-minute huddles” with all the on-site team members (the superintendent, field engineers, project coordinator and safety officer). Matt used these informal morning meetings to share the latest instructions from the client and to ensure that team members understood one another’s current workloads and constraints and understood how they could help one another. Very often, the meetings enabled the team to identify and resolve conflicting priorities before they became problems. Matt noted that, while the primary purpose of the huddle was to update everyone, it also reinforced a spirit of camaraderie and a sense of shared purpose. As a result, these meetings turned out to be very valuable for sustaining teamwork.

As for the third practice, frequent moving about, one project manager, Tony, described the three primary outcomes of spending 30 minutes a day roaming around the project site. First, he was able to develop rich and open communication with his team members. Tony explained that while many workers did not feel safe asking him questions during various formal meetings, they felt very comfortable interacting with him freely during his on-site visits, which had a great impact on their motivation. Second, receiving immediate information, and in particular a greater range of information, enabled him to identify problems early on. At times, he was able to detect conflicts before they actually became an issue. Third, Tony developed a much better understanding of where the project was with respect to the schedule, rather than having to take someone’s word for it. He found that coming to the weekly and monthly planning and scheduling meetings equipped with firsthand, undistorted information allowed him to address questions and solve problems much better. The Boldt project managers did not agree on the preferred timing for moving about and, in particular, whether one should schedule the visits, as Tony did, or leave their timing flexible. However, they all agreed that moving about is a most effective practice that should be applied as often as possible.

These three practices are not limited to construction projects. For example, in the previously mentioned JASSM project, which was geographically dispersed, all three practices necessary to maintain forward momentum were employed by the various project managers at each production site. Additionally, Terry, the customer’s project manager, spent much of his time moving about between all the different production sites.

Implications for Senior Managers

Although every project manager tries to minimize the frequency and negative impact of unexpected events, in today’s dynamic environment such events will still occur. Acknowledging the emergence of a problem is a necessary first step, allowing the project manager to respond quickly and effectively. Some organizations assume that almost all problems can be prevented if the project manager is competent enough — resulting in project managers who are hesitant to admit that they are facing an emerging problem. In fact, a recent study indicates that project managers submit biased reports as often as 60 percent of the time. When upper management fosters an organizational climate that embraces problems as an inherent part of a project’s progression, project managers are able to detect and resolve problems more successfully.

Management scholar Henry Mintzberg argues that today’s managers must be people-oriented, information-oriented and action-oriented. In contrast, the two prevailing project management approaches, the traditional approach and the agile approach, do not require project managers to encompass all three orientations. The traditional approach (primarily intention-driven) stresses information, whereas the agile approach (primarily event-driven) stresses people and action.

By assuming the four roles discussed in this article, the successful project managers we studied are both intention- and event-driven and embrace all three orientations. Developing collaboration requires them to be people-oriented. Integrating planning and review with learning requires them to be information-oriented. Preventing major disruptions requires them to be action-oriented. Finally, maintaining forward momentum, which is pursued throughout a project, requires them to adopt all three orientations. Senior managers must ensure that all three orientations are considered when selecting project managers and developing project management methodologies.

View at the original source

The Upside to Large Competitors 08-20

The Upside to Large Competitors

New research suggests that a smaller company can benefit by making consumers aware that it competes against bigger corporations.

Large competitors are often viewed as a major threat for startups and small companies; big companies have more financial resources and greater scale, market power and brand awareness than smaller ones. However, our research finds that a smaller brand can actually benefit if consumers can see the competitive threat it faces from a larger organization.

When Cold Stone Creamery, a U.S.-based ice cream chain with about 1,400 stores, moved within 50 steps of a J.P. Licks ice cream store in Newton, Massachusetts, some people expected that J.P. Licks, a small, locally owned company, would be beaten out of the Newton market. But consumers rallied around J.P. Licks, and Cold Stone later closed its nearby location. When the owner of the Los Angeles-based coffee store chain The Coffee Bean & Tea Leaf could not stop a Starbucks coffee shop from moving in next door, he was surprised to see his sales shoot up — so much so that he started proactively colocating new stores next to Starbucks ones.

These examples are not anomalies. In six lab and field studies, we explored the effects of having a large, dominant competitor and found that highlighting a large competitor’s size and close proximity can help smaller brands, instead of harming them. (Detailed results of our findings can be found in “Positioning Brands Against Large Competitors to Increase Sales,” forthcoming in the Journal of Marketing Research. See “Related Research.”) Compared to when they are in competition with brands that are similar to them in size or when consumers view them outside of a competitive context, small brands see consumer support go up when they are faced with a competitive threat from large brands. This support translates into higher purchase intention, more purchases and more favorable online reviews.

As part of our research, we conducted a field study at an independent bookstore in Cambridge, Massachusetts. Upon entering the bookstore, 163 prospective shoppers were exposed to one of three versions of an in-store ad, emphasizing either the store’s large competitors, small competitors or no competition. Shoppers who read the “large competitors” version were told that the store’s main competitors are large corporations that have the ability to put small businesses such as this bookstore out of business. The “small competitors” version indicated the store’s main competitors are other locally owned small bookstores in Cambridge. In the “no competition” version, participants were given no information about the competitive environment. 

Shoppers were then given a $5 coupon, coded with the in-store ad version they read. Analyzing shoppers’ sales receipts and the number of redeemed coupons, we found that shoppers were significantly more likely to make a purchase after reading the “large competitors” version of the in-store ad, compared to the “small competitors” version or the no competition version. They also purchased more items and spent more money at the store, compared to shoppers reading the “small competitors” or “no competition” versions. These results suggest that framing the competitive game and emphasizing a competitive narrative against a larger company can help a small establishment — and spur consumers to make a purchase that supports the smaller competitor.

In subsequent studies, we tested this “framing-the-game” effect in various contexts and product categories and further found that support for a large brand decreases when consumers view it as being in competition with a smaller brand. In one study, we asked participants to assess two hypothetical rival tire shops, “Tire World” and “Tire Planet,” under three conditions — small vs. large, small vs. small or large vs. large competitors. While participants indicated no preference for the small or large shop when it was competing against a competitor of similar size, the small vs. large competitive context elicited a strong preference for the small rather than large shop. Participants indicated they were significantly more favorable to the small “Tire World” shop in the small vs. large setting than in the small vs. small, and significantly more adverse to the large “Tire Planet” shop in the small vs. large setting than in the large vs. large.

We concluded that framing the game as a competition changes the way consumers view both competing brands and motivates them to express their views and to have an impact in the marketplace through their purchasing. When a brand is presented within a competitive context, consumers consider not only each brand’s attributes, but also which player they want to support and how they perceive their own purchasing actions will make a difference in the marketplace.

Related Research

N. Paharia, J. Avery and A. Keinan, “Positioning Brands Against Large Competitors to Increase Sales,” Journal of Marketing Research, in press, http://dx.doi.org/10.1509/jmr.13.0438.

Across the studies we found that, when they are perceived as competing against a smaller brand, large brands do not have to do anything explicitly wrong to trigger consumer boycotting behavior; they are, in effect, assumed guilty until proven innocent. Furthermore, large brands should be wary of overtly intense competitive maneuvers. While these strategies may not hurt large brands when they compete against other large brands, they may inspire consumer rejection when the competition is a small brand. Rather than moving in across the street from a small local café, Starbucks Corp., for example, should consider locations where it does not appear to be competing directly against smaller players.

We also analyzed more than 10,000 reviews from Yelp.com to further test whether a perceived competitive threat from a larger company elicited support for the smaller competitor. We used star ratings from Yelp as a proxy for brand evaluation. In an initial exploration, we found that in a large U.S. city, small, local establishments in the “coffee & tea” category had better average Yelp ratings when they were in close proximity to a Starbucks coffee shop than if they were located further away. However, we wished to control for a number of alternative explanations, including multiple Starbucks locations and varying quality levels of smaller establishments.

To control for quality variance, we focused on a relatively smaller chain of coffee shops, Peet’s Coffee & Tea, based in Emeryville, California. To control for the presence of multiple Starbucks locations, we used the number of times “Starbucks” was explicitly mentioned in the text of a Peet’s Yelp review as a proxy for competitive salience. We reasoned that if there were one or many locations of Starbucks nearby, it would be more often mentioned in the reviews. We predicted that if a higher percentage of reviews for a specific Peet’s location mentioned Starbucks, ratings of that particular Peet’s location would also be higher. For each Peet’s location, we recorded the total number of Yelp reviews, the number of reviews that mentioned Starbucks, and the average star rating. Overall, across the 201 Peet’s locations listed on peets.com, 24.5% of the 10,445 reviews explicitly mentioned Starbucks.

We found a significant positive relationship between the percentage of Peet’s reviews that mention Starbucks and the average star rating for a given Peet’s location. Furthermore, the effect was robust when looking at average ratings for the reviews that mentioned, or did not mention, Starbucks. If a location had 50 reviews and 15 mentioned Starbucks, the effect held whether we looked at only those 15 reviews or at the other 35 reviews that did not mention Starbucks.

These results show a positive relationship between the perceived salience of a large competitor and customers’ ratings; in other words, customers liked Peet’s better when they perceived it in competition with Starbucks. To further rule out alternative explanations, we replicated these results in a controlled lab setting, where we described a hypothetical coffee shop and manipulated its proximity to a large competitor. Again, we found that competitors near each other elicited preference for the small rather than large competitor, while competitors located further away did not.

Our research demonstrates the importance of considering a brand’s competitive context and illuminates how small brands can benefit from the real or perceived presence of a large competitor. Many smaller brands shy away from mentioning their competition in their marketing communications, especially at the point of purchase. Research has found that when Wal-Mart Stores Inc. enters a market, the most common reaction for incumbent smaller retail chains is to do nothing, at least in terms of advertising and the marketing mix. Our research suggests that this strategy may be unwise. For the framing-the-game effects outlined here to work to the advantage of small companies, competitive narratives should highlight the battle between small and large competitors, and this narrative must be made salient to consumers at the time of purchase.

Brands don’t have to be tiny businesses to benefit from such competitive narratives. Jim Koch, the founder and chairman of the Boston Beer Company, makers of Samuel Adams beer, has over the years often compared his independent brewery to the behemoth Anheuser-Busch Companies, LLC, framing the game to Boston Beer’s advantage with claims such as “Anheuser-Busch spills more beer than we make.” A Sam Adams advertising campaign entitled “Growing Up Small” attempted to remind consumers of its diminutive competitive position. Despite the objective reality that Sam Adams is by now a well-known brand, Koch deftly reframes the competitive game to benefit the Sam Adams brand. What drives the framing-the-game effect is not absolute size but consumers’ perceptions that a brand is smaller than its larger competitor.

Reproduced from MIT SLOAN Management Review