The National Weather Service (NWS) provides a comprehensive program of meteorological data gathering and analysis, regular forecasts, and special event warnings that we all rely upon. For many, the six-hour forecast update cycle from the NWS is all we need. Yet, in the United States, the private weather forecasting industry is big business.
A 2006 survey of meteorologists conducted by David B. Spiegler, a certified consulting meteorologist, and with the cooperation of the American Meteorological Society (AMS), showed that the private meteorology industry earns revenues in excess of $1.8 billion dollars. A year earlier, the AMS projected in its publication “The Outlook for U.S. Meteorological Research in a Commercializing World” that 50 percent of its members would be employed in the private sector by 2010. The NWS itself lists hundreds of private meteorology firms on its Web site (www.nws.noaa.gov/im/dirintro.htm.
Many of these private companies provide services and sell products that are not in competition with the core functions of the NWS. For example, a number of private companies design and sell software to present and package forecasts for the media. Others work as meteorology consultants serving, for example, as expert witnesses in litigation regarding past weather events or as consultants for the design of weather derivatives in the financial services industry. Some offer Web sites to the public that simply further distribute data and forecasts issued by the NWS. However, a large number of private meteorology firms gather and analyze data and issue forecasts and warnings in what would appear to be direct competition with the NWS.
The need to avoid potentially destructive competition between the NWS and the private meteorology industry has long been recognized by both parties. These private firms, which are represented by their lobbying arm the American Weather and Climate Industry Association (formerly the Commercial Weather Services Association), have, for decades, advocated prohibiting the NWS from providing services that the private firms could provide on their own. As early as 1991, the NWS issued a policy statement establishing its commitment to cooperation with the private sector. In its most recent statement on the subject, NOAA (NWS's parent agency) stated the following on its Web site (http://www.noaa.gov/partnershippolicy/):
The nation benefits from government information disseminated both by Federal agencies and by diverse nonfederal parties, including commercial and not-for-profit entities. NOAA recognizes cooperation, not competition, with private sector and academic and research entities best serves the public interest and best meets the varied needs of specific individuals, organizations, and economic entities. NOAA will take advantage of existing capabilities and services of commercial and academic sectors to support efficient performance of NOAA's mission and avoid duplication and competition in areas not related to the NOAA mission. NOAA will give due consideration to these abilities and consider the effects of its decisions on the activities of these entities, in accordance with its responsibilities as an agency of the U.S. Government, to serve the public interest and advance the nation's environmental information enterprise as a whole.
Far from attempting to outcompete private industry, the NWS even goes so far as to help publicize the existence of private firms and the services they perform. And in many ways, the NWS benefits from the existence of the private sector. For example, the NWS relies heavily on the airline industry to provide upper air data observations. And on the surface, more than 1,000 private observation stations are feeding data into NWS operational systems, with the potential to add more than 16,000 additional private stations over time.
Yet, amid all this activity, a few basic questions naturally arise. If up-to-date weather forecasts provided by the NWS are free and widely available, how do private forecasting companies—which charge for their forecasts—compete with the abundance of free data from the government? And because all meteorological data, model outputs, and other services the NWS provides are equally available to all private firms, how do these firms differentiate themselves from each other with customers? What types of private weather solutions are businesses looking for, and what innovators are on the rise? And what is on the horizon for the private weather forecasting industry?
Why Pay for Private Forecasting Services?
The private meteorology forecasting industry is composed of many small companies that coexist with a handful of very large organizations such as Accuweather and Weather Services International (WSI), which is a division of The Weather Channel. Many of these companies, both big and small, offer services other than the private forecasting services on which we are focusing in this article. They supply recorded forecasts for local radio stations, databases of historical meteorological data for litigation support, and, in the case of the larger organizations, proprietary software for generating the graphics with which we are so familiar on local television newscasts.
With regard to private forecasting, however, the smaller companies tend to serve clients such as cities and towns (and their school departments and road crews), smaller airports, and individual corporations that help to safely operate their helicopter fleets, for example. The large companies tend to serve much bigger clients such as major airlines, airports, and commodity investment firms. Yet the reasons these clients choose to engage private forecasting companies, rather than relying on publicly available forecasts, tend to be quite similar.
The most important reason that clients choose to engage private forecasting companies is accuracy. This factor was cited by virtually all the companies and clients we interviewed. Although most clients continue to consult the National Weather Service (NWS) forecasts, and forecasts available on Internet sites and through local television and radio, they regard the forecasts they receive from their private services as more accurate. Some attribute this to what they regard as differing missions and motivations of other forecast sources.
Those companies we interviewed suggested that the forecasts carried on broadcast outlets were tainted by a perceived need to sensationalize the weather to garner viewers and listeners. Others felt that the NWS was motivated mainly by public safety and was therefore less likely to deal in probabilities and likelihoods when forecasting a potentially dangerous situation. But when talking about accuracy, both companies and clients cited two aspects of accuracy above all as the value they cannot get from publicly available forecasts: “localness” and “customization.”
Especially for the smaller clients, what they cannot get from television, radio, and the Internet is a forecast for the local areas these clients service. The City of Cambridge, Massachusetts, consists of a variety of neighborhoods whose elevations and closeness to the ocean vary considerably. When Bill Dwyer, Superintendent of Streets, Sidewalks, and Sewer Maintenance for Cambridge, is deciding whether to gear up for road de-icing, line painting, snow plowing, and other activities, he cannot get neighborhood-level information from any public source. He turns to Hometown Forecast Services, of Nashua, New Hampshire,
Kevin Ward, the Public Works Director of Scotch Plains, New Jersey, agrees: “The minor details dictate our operations hour-by-hour.” He continues, “TV and radio forecasts can't be as local as we need and get—what you come to find is that the information you need must be timely and accurate. Forecasters on TV aren't specific enough—it's all geared to a broad audience. ‘Hazardous’ is a very general term!”
Similarly, when Jennisse Schule, Transportation Coordinator for the Paramus, New Jersey, public schools, is deciding whether to close the public school system in the face of an approaching snowstorm, she needs information specific to Paramus, not a general forecast for northern New Jersey. The smaller private forecasting companies usually specialize in the weather of their surrounding area and are very aware of the microclimates within each forecast area. Some of these companies even deploy private sensors on road surfaces and other data-gathering devices to assist them in forecasting for these localities.
Both Ward and Schule use The National Weather Station, of Hasbrouck Heights, New Jersey, for their forecasting needs.
In addition to “localness,” smaller forecasting companies and their clients also laud the ability of private forecasters to customize their services to directly serve the unique needs of individual clients. Forecasts are generally updated on a much more frequent basis than those that are publicly available, often hourly or more often in the case of an upcoming weather event. And updates are usually the result of fresh analysis, not simply repeats of forecasts issued earlier. Referring to his relationship with Dan Ventola, President of The National Weather Station, Ward says, “I've gotten updates at 3 a.m., and not just a repeat.” Furthermore, these companies can tailor the information they supply to their clients so that clients receive information they can actually use. For example, by exploiting private sensors, Ward receives specific road surface temperature forecasts from The National Weather Station, not just generic air temperature, relative humidity, wind speed, and direction.
Although it may not be surprising that private forecasters and their clients mention increased accuracy as a reason to engage a private forecasting company, two other reasons are mentioned just as frequently and may play an even more important role: operationalizing the forecast and personal consultation.
Operationalizing the Forecast
Clients of private forecasting companies consistently praise the companies' ability to help operationalize their forecasts. As Carl Braley, Airport Superintendent for the City of Manchester, New Hampshire (Manchester-Boston Regional Airport), describes his relationship with his private forecast company: “This is the decision I need to make. How should I decide?” Instead of supplying just data and forecasts, Rob Carolan, President of Hometown Forecast Services, knows that Braley needs to decide whether to close the Manchester airport or even bring in extra personnel if aircraft are likely to be diverted to Manchester from other storm-affected airports. Each of these decisions can involve extreme costs that far exceed the fee for an entire year's private forecasting contract. Therefore, clients need to get data and forecasts that are directly relevant to their business operations and costs. They need forecasters to be willing to explain how these data and forecasts translate into the decisions they need to make. And they need forecasters who are willing to speak in terms of probabilities, not unattainable certainties. In Braley's words, “Part of the value of the service is knowing levels of probability. In my work, de-icing the airfield may cost me $60,000-$80,000, so I need to understand the likelihood of any particular weather occurrence.
The smallest details can yield savings or headaches. Dwyer states, “It is very important to know the timeline with accuracy as it affects how I utilize my people and equipment.” He continues, “I'm in charge of salting and snow operations, so timing has a dramatic impact on how we approach things. Will the sun break in time to melt away the ice glaze on the sidewalks? Do I want all my trucks out salting the roads—with expensive overtime—or can today's core crew do the job?” Clearly the forecasters don't make these ultimate decisions, but they need to be helpful in translating their forecasts into operational inputs.
The third factor, and perhaps most important in deciding why clients hire and retain private forecasting companies, is the opportunity to have personal, one-on-one conversations with a meteorologist who is on call whenever there is a need. “We need someone we can have dialogue with,” says Braley. Clients repeatedly cite the value of having a person who is familiar with the client's location, business, and forecast needs throughout a weather event, by taking their calls and answering their e-mails. In the most satisfying relationships, the same forecaster is consistently assigned to the client so that an interpersonal relationship is created. This leads the best companies to be proactive in their relationships, such as calling the client with updates and warnings before even being asked. Not unlike many other professions such as legal, accounting, medical, etc., private meteorological forecasting turns out to be, at its core, a relationships business. “I feel like they know me and I feel like I know them,” says Dwyer, referring to his relationship with Hometown Forecast Services.
Large Forecasting Companies
The largest private forecasting companies operate on a much grander scale than their smaller competitors, servicing much larger organizations. Yet, the basic reasons that large clients hire private forecasters are much the same as in their smaller counterparts: accuracy, operationalizing the forecast, and the availability of one-on-one consultation.
Peter Neilley, Vice President of Global Forecasting Services for WSI, describes the two largest industries served by his company as aviation (specifically large airlines, with some smaller airports and flight planning for clients with fleets of aircraft) and energy (specifically decision support for investment decisions in the energy commodities markets). Of course, WSI is also a major player in the media industry, providing forecasts, graphics, and supporting software to television stations in most major markets.
With regard to increased accuracy in comparison to generally available forecasts, all the factors that contribute to greater accuracy in smaller forecasting companies (such as knowledge of local topography and microclimates, mastery of particular industry niches, ability to customize information for a client's needs) are equally applicable to WSI and its large counterparts. But the vast scale of WSI's operations allows it to enhance these advantages in ways that are not available to smaller competitors. For example, WSI uses proprietary computer forecasting models not available elsewhere, software that it developed for cleaner data input and analysis, and proprietary data sets.
WSI's ability to customize and operationalize its services for its clients' needs is also significantly enhanced by its larger size, scale, and capabilities. “We deliver content in the way it is most useful and integrated into our clients' operations,” says Neilley. In the aviation industry, for example, WSI's software is integrated into the fleet management systems of the airlines, feeding meteorological information and forecasts directly into the decision-making software that supports scheduling, fueling, and dispatching decisions. WSI software is often installed directly into the cockpits of individual planes for use by pilots in flight.
For energy trading companies, WSI recognizes that its clients need an edge in their ability to predict price movements due to the weather. They cannot wait for generally available forecasts. Thus, WSI has developed proprietary prediction models that essentially “forecast the forecast,” giving their clients advance knowledge of what the next generally available model outputs are likely to say and how that will translate into energy demand.
And despite their breadth and the size of their clients, the larger private forecasting companies are at least as aware of the value of interpersonal communication as their smaller counterparts. For example, WSI has “embedded” forecasters at their airline clients' sites, sitting side by side with airline employees as they make dispatch and fueling decisions. And, just like local public works departments and school districts served by smaller forecasting companies, energy traders have dedicated forecasters assigned to their companies with whom they can have direct communication and conversations whenever necessary.
Overall, clients of private forecasting companies see tremendous value in their relationships with their forecasters, well beyond what they can get from publicly available sources. From a cost-benefit point of view, clients repeatedly emphasize that one mistaken decision to deploy snowplow staff and equipment or line painting crews, one mistaken decision to send home a company's workforce ahead of a storm that doesn't arrive, or one unnecessary closing of an airport can easily cost more than an entire year's fee for private forecasting services. “The value for your dollar is a great benefit,” says Ward.
Competition Within the Industry
Although accuracy is critical in private forecasting, this is also a relationship business. Private forecasters are chosen by their clients based on their consultative skills and how well they can relate their meteorological knowledge to costly or critical operational decisions. Because of these factors, price plays a relatively minor role in choosing a private forecasting service.
Moreover, product lines—the various ways that weather information can be received—do not vary widely, or appear to be the main method of differentiation, especially among smaller weather information providers. Storm alerts and warnings are core services offered by all forecasters. Weather information is generally available in a wide array of formats (72-hour forecasts broken out in three-hour segments, 10-day outlooks sent daily, etc.). Generally, with the exception of the larger companies, forecasters are relying on identical computer models, although some forecasters also consider proprietary sensing and data.
In virtually all cases, clients cite the importance of getting a live meteorologist on the phone, particularly during severe weather events. Client/forecaster trust is something that is only built up—and lost—over time. As such, those competing for new private forecasting contracts, as in most other categories of professional services, run uphill against established relationships and rhythms. Again, clients see the overall cost of weather services as low compared to the potential economic impact of the weather on their business, so they are disinclined to take on a new weather service simply for a lower price.
Private forecasters have access to and exploit the same basic weather data from the National Weather Service (NWS). According to Carolan, “We are all generally looking at the same data produced by [the models].”
Smaller firms generally compete through being geographically focused and parsing this data out. Localness—having people physically in the geography of the service—is a differentiator. Ventola says, “Our knowledge of local topography is important because we know how the major models work in our micro-climates.” He continues, “Distant weather providers—those possibly hundreds of miles away—lose all credibility by asking their client the kiss-of-death question, ‘well what's it doing there now?’” Geographic proximity, not the underlying data, is often critical to winning an account.
While relationships and trust are key, there is still healthy competition in the industry, which puts a ceiling on prices charged for services. Forecasters aim to attract new clients through referrals from existing satisfied customers. New clients might hear about a particular forecaster at an industry conference, such as a Public Works and Municipalities conference where attendees commiserate about how to improve operations or streamline costs. It is not uncommon for a buyer to run parallel private forecasts coming from two weather services companies in a “horse race” to compare and evaluate performance for their specific operational needs.
Having a string of negative experiences with one's current forecaster is commonly cited as the reason to shop for a new forecast provider. But there is also a healthy appreciation for the quality of work that private forecasters do. Ward states, “I will look out my window as a weather pattern changes within minutes of the prediction and say—that's very impressive.” He continues, “To me the cost is a pittance compared to the benefit.”
To a lesser degree, companies compete based on their ability to distribute weather information across a client's operations. For example, Braley has designated almost 30 parties to receive severe weather warnings across his extended operations at the Manchester-Boston Regional Airport. That way he can spend more time focused on airport operations and less wasteful time communicating about weather events with partners and employees.
The “800-Pound Gorillas”
Opposite to localness, large weather companies such as WSI, Accu-Weather, and EarthSat compete through economies of scope, both in geographies served but also in large-scale proprietary modeling.
Dr. Peter McNeely of WSI works to exploit the company's capabilities across business lines that focus not only on media (television news across the country), but also on the aviation, energy, and investment sectors. Large private forecasting companies such as WSI develop proprietary weather models to address their clients' unique and often far-reaching needs. Proprietary weather modeling is important in the investment industry because it can give an edge in commodities investing, where weather patterns can shape assessments of commodity values at different points in time. McNeely states, “Our clients come to us for proprietary data to augment their decision-making systems,” adding, “we think of our content as being at least a generation ahead of the National Weather Service.”
In this frame, each generation and specialization of innovation in the private forecasting industry can create new investment opportunities. There is a real economic value, especially to clients, in research and development to improve forecast modeling. In the investment industry, for example, to the degree that investors believe the weather affects commodity price movements, there is incremental asymmetric information to be unearthed and exploited. McNeely points out, “We offer interpretive value-added services in the language of the user.”
While the NWS provides a critical chassis for private weather information, it is not a major driver of innovation for private aims. To offer a historic analogy, we can think of NASA's early efforts in space exploration as critical to the emergence of the private space flight industry, where customer needs, markets, and organizational capabilities are all now forming in a developing ecosystem.
The Future of the Industry
Looking forward, the private forecasting industry should continue to grow to the degree that it can provide real economic benefits to its clients. An interesting question is, how much of this value will the weather companies themselves will be able to appropriate, or will their clients continue to capture the lion's share of this value (comparing client savings to the fees they pay)? There is the likelihood that budget cuts may ultimately cause the NWS to concentrate almost entirely on collecting and compiling foundational data sets and to de-emphasize providing specialized forecasts and warnings of increased future extreme weather events. If this happens, then private weather forecasters will be poised to take advantage of new opportunities. Yet there is little doubt that even while exploring these new opportunities, the industry will continue to be driven by the trusted relationship between forecaster and client.
RICHARD MANDEL is an Associate Professor of Law at Babson College. He earned a bachelor's degree in Meteorology at Cornell University, has studied Meteorology at the University of Massachusetts-Lowell, and teaches Meteorology at Babson.
ERIK NOYES is an Assistant Professor of Entrepreneurship at Babson College. He researches the entrepreneurial behavior of small and large companies in fast-changing industries.