Innovation in Public and Private Weather Services [ 1 ] Innovation in Public and Private Weather Services Photo: Xinhua/Alamy Live News —A student at the 32nd Hefei teenager science & technology innovation competition shows a smart weather condition reporting window she invented. Anhui Province, China­ Abstract Authors The weather enterprise must continue to innovate and expand to keep people safe from David P. Rogers, Lead Meteorological disruptive weather, and to keep up with the demand for weather intelligence for eco- Consultant, World Bank nomic development. Despite considerable scientific and technical advances, underin- Vladimir Tsirkunov, Lead Specialist, vestment in services, coupled with poor regulation and business practices, prevents the Head of the GFDRR Hydromet enterprise from fully protecting developing economies. New approaches are needed. Program, World Bank The lean startup focuses on development with a minimum level of investment. While the Anna-Maria Bogdanova, Disaster Risk build–measure–learn approach is used extensively in the private sector, we consider its Management Specialist, World Bank application to NMHSs and how this might be used to prototype new capabilities quickly Haleh Kootval, Senior Technical with modest investment in different components of the meteorological value chain. The Specialist, World Bank process introduces new ways to manage projects under extreme uncertainty and bene- Makoto Suwa, Senior Disaster Risk fits traditional institutions to strengthen the way they operate. Management Specialist, World Bank Introduction Technical Note from the series focusing on specific aspects of Present and growing weather extremes and the increasing dependence of econom- Public and Private Engagement ic performance on weather intelligence are challenging societies everywhere. Keeping in Hydromet Services people safe from disruptive weather, while enabling economic development, depend on inventive products and services that meet the needs of people and markets. Innovation impinges on the public, private, and academic sectors to embrace new technologies in the weather enterprise,1 exploit the benefits of open data policies, enhance the capacity of the public and private sectors, expand the capacity of users to utilize weather-related intelligence for decision making, and operate within the framework of increasing con- straints on government financing of public services. [ 2 ] Innovation in Public and Private Weather Services Organization and innovation form essential components throughout the weather enter- prise to ensure that it is well suited and equipped for its roles. Despite its fundamental im- portance to society, the weather enterprise has not yet achieved its full potential (Thorpe and Rogers 2018). Underinvestment, coupled with poor regulation and business practices, has limited the ability of the enterprise to “weather proof” developing economies (Rogers and Tsirkunov 2021). Rivalry between public and private actors—rather than competitive innovation as well as limited interactions with the academic sector—has stymied the de- velopment of the weather enterprise in many countries (Rogers et al. 2021). Open data policies, which are proven to increase economic benefit, are not routinely applied to me- teorological data (Rogers and Tsirkunov 2021). Ensuring the whole of society has access to the best available products and services is a challenge. Modern meteorological services must be in the vanguard of innovation—exploiting non-traditional observations and using data analytics to create impact-based forecasting and warning services or enabling busi- ness data integration of weather intelligence, are two examples. They require new tools to prototype rapidly and operationalize novel services and business practices that apply equally to the private and public sectors to enhance and expand the enterprise. Large public sector investments in National Meteorological and Hydrological Services (NMHSs) and disaster management agencies tend to apply proven but often outmod- ed solutions based on the experience of past projects. In general, these are scoped well in the overall results expected, but do not account for the rapid development of technology and knowhow. The most advanced NMHSs and private companies include research and development (R&D) branches, and increasingly these include units that focus on rapid prototyping of new solutions. One approach is the application of the concept of the lean startup (Reis 2011), which has been proposed as a means to create nascent private weather services (World Bank 2019) but may be equally applicable to innovate products and services in the public sector. The lean startup focuses on the development of new services with a minimum level of investment and with users of the potential services engaged from the outset to ensure the products are tailored ap- propriately. In the public sector, this approach applies equally to products and services for any government, including disaster management, water resources, and agriculture, and can be used to create new integrated products and services. In civil society, this could focus on development of tools specifically to help disadvantaged or citizens with special needs. In this technical note, we examine the application of the lean startup in the NMHSs, and consider how this might work within the context of international investment in develop- ing countries. A potential benefit of the approach is the ability to test and evaluate the coproduction of services with the private sector or other government departments and agencies or jointly with both. The Changing Landscape As countries embrace open data policies, they can expect that many data-focused public sector agencies will need to adjust their products and services (World Bank 2017). In particular, they may be forced to retreat from various forms of commercial enterprise— especially data sales—and comply with competition laws. In many developing countries, where the NMHSs exercise a monopoly over data and may have legislated in the past against private commercial services, this will require rethinking the role of the NMHSs in Innovation in Public and Private Weather Services [ 3 ] the national economy. The prevalent understanding suggests that the weather-depen- dent economy will develop faster and more sustainability by encouraging: (i) the devel- opment of private commercial services that will innovate and improve economic efficien- cy through competition, and (ii) a public sector which focuses on delivery of public goods contributing to the economy and safeguarding people (World Bank 2019). Regulations would foster rebalancing the roles of the public and private sector and would define the respective responsibilities of the protagonists in the weather enterprise replacing often adversarial positions with cooperation, coproduction and an open market (Rogers et al. 2021, Rogers and Tsirkunov 2021). An underlying assumption of a policy to make public data free to use and re-use without restriction is that it will fuel more economic activity. While sufficient evidence supports this notion (World Bank 2019), resetting the relationship between the NMHSs and the private sector has a number of consequences that may adversely affect the public insti- tutions’ ability to fulfil their responsibilities, unless anticipated and mitigated. These in- clude replacing any lost income from the NMHSs’ data sales and non-competitive com- We are drawn to examine the mercial activities or repurposing or downsizing (World Bank 2019). In countries where entire weather enterprise when the private sector has been absent, the NMHS is a monopsony, which means the gov- ernment may set the price for wages and goods artificially low. As the economy opens, investing in the development of the market price for wages and goods is likely to be much higher and the government meteorological and hydrological may no longer be an attractive employer or consumer of goods. Such a likelihood can services. result in a transfer of human capital from the public to the nascent private sector, and put additional pressure on the NMHSs’ ability to provide services. We are drawn to examine the entire weather enterprise when investing in the develop- ment of meteorological and hydrological services. Transformation of Meteorological and Hydrological Services Modernization or upgrades to the NMHSs funded by development banks and other in- ternational development agencies have had mixed results. While the aim is to help the government and its agencies develop public services that meet the growing needs of society, the investments often fall short of this ambition. The issues have been elabo- rated at length (Rogers et al. 2019, Rogers and Tsirkunov 2013, World Bank 2019) but transformation remains a stubbornly difficult problem to solve. Rather than reiterate the problems and attempted solutions, we draw on a different line of thinking about busi- ness practices and innovation. We can address an aspect of the root cause of some of the problems facing development partners investing in the NMHS using five “why” questions: Why do large investments in the NMHS regularly fail to achieve the expected improve- ments in public services? • After the project ends, no funds are allocated to cover the additional costs of operations and maintenance of the new systems. Therefore, modernization is not sustainable. Why is the modernization not sustainable? • A lack of understanding exists about how much resources and effort are needed to support the national weather enterprise, particularly its public part. This leads to government policies that stimulate the use of fee-based services, neglect education [ 4 ] Innovation in Public and Private Weather Services and R&D, and that introduce regulations supporting the NMHS’s monopolistic role instead of incentivizing market development, and give insufficient support for basic infrastructure. Why is the government not committed? • Neither revenue nor public response demonstrates additional social and economic benefits by the end of the project. Why are the benefits not clear? • The project financed the development of new services; however, the producers (NMHSs) believed that infrastructure would somehow resolve the issues, and did not consider what users and customers wanted, or the need for education and training, and raising public awareness of users. Why did the NMHS not engage with users more effectively? • It is not in the culture of the NMHS to engage with its users in a way that they help define the products and services that they want. Historically, the NMHS’s main task was to produce high quality data and conduct research. We can surmise that user engagement is critical in any modernization project. However, it is a skillset that is often missing from the NMHSs’ engagement strategy. Consequently, it is often heavily discounted, or considered an add-on rather than the first step in de- veloping services for users and limited to a few stakeholder encounters concluding with the request for more timely and accurate weather forecasts and warnings; a statement that is axiomatic and not particularly helpful in designing specific products and services. Meteorological and hydrological Meteorological and hydrological services are going through a major transformation about services are going through a the way products are made, how services are delivered (Rogers et al. 2021) and who provides those services (Thorpe and Rogers 2018, World Bank 2019). A pivotal change major transformation about the occurred in 1994 with the creation of the public weather services program within World way products are made, how Meteorological Organization (WMO) (WMO 1996, 2000) when the focus shifted from services are delivered and who the NMHSs providing products and services which they considered useful to engag- ing with customers to determine what services were required for these customers. The provides those services. WMO (1996) states, “Public weather services should only be undertaken in response to actual and real needs and requirements expressed by the user community and not as an end in itself.” This assertion is prescient of modern business practices that emphasize engagement with customers to determine what they want, and therefore, the right thing to build. It may seem alien to many NMHSs which have mandates to provide weather forecasts and warnings, but in the end, it is actions of the users of these services that determine their utility. The Power of Partnership: Public and Private Engagement in Hydromet Services (World Bank 2019) poses two questions: (i) what are the characteristics, advantages, challenges and potential risks of the various public-private engagement models in the hydromet sector?; and (ii) what are the available options for developing countries that wish to streamline and augment public and private engagement to strengthen the provision of hydromet services, and, accordingly, the conditions necessary to succeed? Among many important conclusions, the study (World Bank 2019) highlights the need for development projects to focus on improving the entire hydromet value chain, rather than only on the NMHS infrastructure and capacity that has hitherto been the main motivation for investment. Innovation in Public and Private Weather Services [ 5 ] The new approach means supporting activities that will enable both the public and pri- vate sectors weather services as well as the academic sector develop cooperative and competitive relationships that sustain the entire weather enterprise. This includes early adoption of advances in science, and using business practices that support consumer-fo- cused development and implementation of innovative solutions across the enterprise. The Lean Startup Rogers et al. (2019, 2021) highlight different business models that can be applied to We pose the question: can the NMHSs ranging from outsourcing activities to the private sector to the coproduc- the lean startup methodology tion of services with the private sector. However, in many developing countries, the private sector is nascent, and its growth depends on developing a commercial market for be used to develop tailored services. This stage of development aligns with a public–private engagement scenario, products and services within the which emphasises strengthening and focusing on the NMHSs providing public services, NMHSs that would otherwise and lays the foundation for the private sector to provide non-public services (World have limited or no access to Bank 2019).2 The lean startup concept is proposed as a means of rapidly developing and bringing to market new products and services in the private meteorological services R&D? sector (World Bank 2019), but is equally applicable to the production of government ser- vices (Ries 2011). In the private sector, an example of the application of the lean startup methodology, relevant to meteorology, is the company Symphony Creative Solutions Private Limited3 founded by Nippon Yusen Kabushiki Kaisha Group (NYK), Weathernews (WNI) and Kozo Keikaku. The company fuses NYK’s shipping and logistics knowledge with WNI’s innovative infrastructure network and weather forecasting technology and Kozo Keikaku Engineering’s advanced operations research techniques and data–analysis technology. The lean startup methodology is used in product development that employs an entrepreneurial business work style to create next generation shipping and logistics solutions. While the most advanced NMHSs can rely on their own internal R&D departments and on academic research, many developing NMHSs do not have this capacity (Rogers et al. 2019). Instead, they tend to depend on products created through external R&D and technology transfer from more advanced NMHSs, the private sector or both. Infusion of new technologies is often sporadic and not necessarily fully fit for purpose. Consequent- ly, we pose the question: can the lean startup methodology be used to develop tailored products and services within the NMHSs that would otherwise have limited or no access to R&D? Ries (2011) defines a startup as “a human institution designed to create a new prod- uct or service under conditions of extreme uncertainty.” This would apply equally to a new multi-hazard, impact-based warning service for a biological threat created by an NMHS, health service and disaster management agency (Rogers et al. 2020a), and a weather service to optimize train operations developed by a private firm, for example.4 The definition highlights several important points: (i) the activity is essentially a human enterprise involving creativity, coordination and a culture to deliver results; (ii) it aims to create something that is a source of value for the people who become its customers or users and how they interact with the company or agency; and (iii) it is undertaken in an environment with a lot of unknowns that will influence the viability of the product or service. Innovation is at the heart of success in this enterprise. It can take many different [ 6 ] Innovation in Public and Private Weather Services forms from the application of a new scientific discovery or technique to simply bringing a product or service to a new location or to a previously underserved set of users. An important tenet of a lean startup is experimentation to test its strategy to build a sustainable business around a vision. The aim is to trial many different assumptions by initially building a simple product that enables the agency or firm to observe the real be- havior of customers and users. Let us consider, for example, a service for government-is- sued, impact-based warnings that aims to provide people at risk with information about what to do in response to a severe weather hazard. The service would then consist of creating warning messages that people understand and are willing to act on. The mete- orological agency and the disaster management agency need to interact with users to understand how they want to receive information, and how they will use the information received. Is a short messaging service (SMS) sufficient? How should the message take ac- tion to be reinforced? Do people respond on their own, or only when the information is validated by other sources? How long will people mill around before making a decision? How long does it take to reunite a family with children at school and family members at work before collective action is possible? These and many other aspects can only be informed through direct contact with the potential users of these services. So, even if no intention exists to build a business, the approach helps agencies engage with their users. Interacting with potential users and customers of the service helps build a minimum viable product that allows the team building the product to collect the maximum amount of validat- ed learning to test various assumptions with the least effort. In many countries, it is common to find that, while relatively good meteorological forecasts and warnings are issued, it is just as likely to find that these warnings are not heeded (WMO 2015). Many reasons support why In both private and public this may be the case, but often these are not well known to the service providers because sectors, we need to address two they have minimal or sometimes no interaction with the users of their services, contrary to hypotheses—value creation the WMO guidelines for public weather services (WMO 1996, 2000). Professional standards often lead the NMHSs to strive to create a quality product, but without knowledge of their and growth. It is fundamental specific clientele behaviors and needs, it is not possible to know what quality is, and there- to determine if the product or fore, risk overengineering a solution. Adherence to International Standards Organization service is value creating or value (ISO) standards that focus on meeting customer demands and needs would help address this challenge. The NMHSs often claim that their observations are of a higher quality than those destroying, and whether this made by the private sector, but this does not mean that the products produced by the public leads to growth or not. sector are superior to those produced at probably lower cost by a private firm; moreover, the private sector will produce what is needed and paid for. Quality is directly linked to what the customer or user cares about. Discovering that a product or service does not meet expecta- tions drains resources in money and effort, and ultimately is a risk to reputation and belies the authoritative voice to which most NMHSs aspire. In both private and public sectors, we need to address two hypotheses—value creation and growth. It is fundamental to determine if the product or service is value creating or value destroying, and whether this leads to growth or not. An increasing number of peo- ple who use a public warning service or more clients who buy a weather product from a private weather firm, would be measures of growth. The value creation of a product or service should be a primary driver of any public sector entity, or else its sustainability will depend on continuous infusion of funds from development partners. In practice, this is a common scenario where huge capital investments are made, the systems they support work for a while and then fail, and the process cycle begins again. It is highly Innovation in Public and Private Weather Services [ 7 ] likely that the supposed users of these services hardly noticed the period when services functioned as expected. While not a guarantee of sustainability, services are more likely to be publicly financed and supportable if they create value for a significant number of customers or users. Ideally users would provide continuous feedback that enables quan- titative measurement of the value and helps minimize a waste of resources. A core component of the lean startup methodology is the build–measure–learn feed- back loop (Reis 2011). A goal of the process is to minimize the amount of total time Figure 1. Build–measure–learn through the feedback loop (Figure 1). In the process, the focus should be on what needs feedback loop. to be learned, what needs to be measured to gain validated learning, and then to figure Ideas out products to be built that run the experiment to get that measurement. The initial loop leads to the minimum viable product, and thereafter the validated learning based on customer or user experience for further iterations and product or service improve- Measure Build ments. Maximum speed in its implementation is critically important to minimize wasting time, effort and money developing something that nobody wants, and to minimize the time reaching the decision point to persevere with the product or to pivot—that is, make a structured course correction to test a new fundamental hypothesis about the prod- uct, strategy and sustainability based on feedback. Returning to the example of an im- pact-based warning service, the team might learn that while users understand the prod- Data Product uct—a warning of the likelihood of flooding of their homes—they do not have sufficient lead time to evacuate. The startup needs to pivot to provide earlier warnings, possibly with a reduction in the accuracy of the warnings. The earlier alerting may be more useful Learn to those at risk than the precise timing and intensity of the event. Source: Reis 2011. Prototyping in the NMHSs Previously, we have argued that investments in the NMHSs should be transformative, and that small investments are insufficient to change a weak organization into a mod- erate or high performing service (Rogers and Tsirkunov 2013). While this argument re- mains true, the transformation structure needs further investigation. Modernization efforts have and continue to focus on institutional strengthening, up- grades to infrastructure, and service delivery (Rogers et al. 2019). While the latter has become more prominent in recent years, most NMHSs want to focus on infrastructure— more observations, more computers, access to turn-key products, ability to run limit- ed-area models, and development of applications. Any training emphasises the use of these tools. This consumes most of the project’s money and time. As we noted in our preliminary five-question assessment, the most important element in the process—the customer, user, consumer—is not really part of the process despite results frameworks promising actionable weather warnings for an entire population. Certainly, investment in infrastructure is important and should proceed, but it should not outpace the development of new products and services and the institutional changes needed to support these developments. Creating an environment where the lean startup methodology can be applied could address this problem; and would require restructuring the service delivery aspects of modernization projects as a collection of lean startups in parallel with some traditional business methods aimed at improving existing services. Such measures would increase opportunities to engage actively with different sets of users or customers as a part of the product or service development process. [ 8 ] Innovation in Public and Private Weather Services The prototyping that is synonymous with lean startups would allow new products and services to be developed while also helping build new infrastructure. An element of a no-regrets strategy has to be applied to the latter, but the build–measure–learn model used to develop new services would incorporate assumptions about the existing and future capabilities of the infrastructure. It should be noted that the lean startup method- ology could also be applied to internal products, such as using data analytics or ensemble predictions systems or both in the forecasting process, for example, where the custom- ers are the forecasters in the NMHS, providing validated learning to the design team. At this stage, we consider how we might develop an integrated new multi hazard im- pact-based early warning service and a decision support system. The first consideration is to create the team to undertake the venture. The team could be a small group from the ex- isting public organizations—disaster managers, meteorologists, hydrologists, Information and Communication Technology (ICT) experts—or it could be a firm contracted for the pur- pose, or a combination of a firm and internal experts. A lean startup is a human institution; therefore, it requires management geared to its context of extreme uncertainty. It requires creating space for experimentation and allowing continuous innovation through testing and interaction with users, where the measure of success initially is in validated learning that is used to improve the service. While the basic building blocks of warning and decision support systems are well established, how to adapt them to the specific needs of individu- als, communities and local authorities are at the outset unknown and untested. Traditional warnings focus on what the weather will be; in contrast, the impact-based system empha- sises the impact of the weather on people, and therefore needs considerable data on the vulnerability and exposure of people. The initial minimum viable product will likely have While the basic building blocks very limited functionality and may only work for a small group of people, but will be able to iterate quickly. It might develop ways to acquire vulnerability and exposure data directly of warning and decision support from users to improve the specificity of the warnings through mobile applications or SMS. systems are well established, how The expectation is that by incorporating users in the process, warning messages will be to adapt them to the specific relevant to their needs, the quality of the service will match users’ expectations, and the needs of individuals, communities service will be scalable migrating from a prototype to a mainstream operational product. and local authorities are at the outset unknown and untested. Conclusions Our technical note has examined how innovative products and services can be devel- oped in public and private weather services. The lean startup methodology described in detail by Reis (2011) is an essential tool that supports an entrepreneurial business, applicable to any sector—government, private, academic, non-profit. It can be used for internal development or external products. The approach can help the weather enter- prise expand to meet the demands of users and customers, particularly where the na- tional institutions need significant strengthening, and where the private sector weather services are nascent. Incorporating the lean startup methodology into major development partners’ invest- ments in the NMHSs and disaster risk reduction may be a way to significantly enhance the delivery of services without requiring a complete overhaul of the investment pro- cess. How would this work within a development project cycle? To complement and improve the design of the development project with the help of innovations such as the lean startup would most likely require grant-funding instruments. Potential innovations Innovation in Public and Private Weather Services [ 9 ] should start prior to final project design so that promising activities are included in it. The need for flexibility of procurement is another constraint. A biproduct of the process is the introduction of a new approach to managing projects under conditions of extreme uncertainty that may have wider benefits to the overall func- tioning of traditional institutions that are trying to reinvigorate the way they operate and respond to societies’ growing needs for weather, climate and hydrological information. Sustainability of the NMHSs and disaster management agencies remains a critical issue, and the approach described is not a panacea; however, by putting users first, a greater chance emerges that the cost of services can be well justified to compete effective- ly with the many demands on public finances. Actively engaging with largest possible number of users and cohorts, and for those users to be able to demonstrate the value of the services to their lives and livelihoods will be the most effective way of defending increased public investment. In the private sector, the lean startup approach is a way to bring new local actors to mar- ket. Their involvement is critical to provide weather-sensitive economic sectors the tools to manage their businesses’ exposure to meteorological and hydrological risks. Notes 1. The Global Weather Enterprise is the value chain of activities of the public, private and academic sectors providing accurate, reliable, and timely weather and climate related information. It contributes to the safety of life and property, poverty reduction, and the promotion of economic development. 2. World Bank (2019) defined public services as “Services which are of public interest and which show public good characteristics or natural monopoly features.” 3. https://www.nyk.com/english/news/2016/004280.html 4. Tuukka Järvenpää (2014) describes the lean startup methodology with numerous examples of its application. The Master’s thesis case study focuses on a Finnish startup company working in the intelligent traffic systems domain by connecting and improving different areas related to total traffic including vehicles, roads and railways and weather information. He concludes that regardless of the industry, companies should aim to quickly build minimum viable products and test them in practice in the marketplace before committing to significant investments. https://www.theseus.fi/ bitstream/handle/10024/80505/Jarvenpaa_Tuukka.pdf?sequence=1&isAllowed=y References Ries, E. The Lean Startup: How constant innovation creates radically successful businesses. Crown Publishing Group, 2011. ISBN 0307887898. Rogers, D. and V. Tsirkunov. Weather and Climate Resilience: Effective Preparedness through National Meteorological and Hydrological Services. Directions in Development. Washington, DC: World Bank, 2013. Rogers, D. and V. Tsirkunov. Open data: A path to climate resilience and economic development in South Asia? World Bank, Washington, D.C., 2021. Rogers, D.P., V.V. Tsirkunov, H. Kootval, A. Soares, D. Kull, A.-M. Bogdanova and M. Suwa. Weathering the Change: How to Improve Hydromet Services in Developing Countries. Washington, DC: World Bank, 2019. [ 10 ] Innovation in Public and Private Weather Services Rogers, D., Tsirkunov, V., Bogdanova, A.-M., Suwa, M. Public and Private Engagement in Hydromet Services: From Rivalry to Coproduction in Meteorological and Hydrological Service Delivery. World Bank, Washington, D.C., 2021. Rogers, D.P., Anderson-Berry, L., Bogdanova, A.-M., Fleming, G., Gitay, H., Kahandawa, S., Kootval, H., Staudinger, M., Suwa, M., Tsirkunov, V. and Wang, W. COVID-19 and lessons from multi-hazard early warning systems. Adv. Sci. Res., 17, 129-141, 2020a. Thorpe, A. and Rogers, D.: The Future of the Global Weather Enterprise: Opportunities and Risks. Bull. Amer. Meteor. Soc., 99, 2003–2008, 2018. World Bank: World Bank Support for Open Data: 2012-2017. World Bank, Washington, D.C., 2017. World Bank: The Power of Partnership: Public and Private Engagement in Hydromet Services. World Bank, Washington, DC., 2019. WMO: Guide to Public Weather Services. Secretariat of the World Meteorological Organization, WMO- No. 834, Geneva, Switzerland, 1996. WMO: WMO Guidelines on multi-hazard impact-based forecast and warning services. World Meteorological Organization, WMO-1150, Geneva, Switzerland, 2015. Credit: NiseriN. The Global Facility for Disaster Reduction and Recovery (GFDRR) is a global partnership that helps developing countries better understand and reduce their vulnerabilities to natural hazards and adapt to climate change. Working with over 400 local, national, regional, and international partners, GFDRR provides grant financing, technical assistance, training and knowledge sharing activities to mainstream disaster and climate risk management in policies and strategies. Managed by the World Bank, GFDRR is supported by 34 countries and 9 international organizations. www.gfdrr.org