Which of the following is characteristic of both early adopters and the early majority?

Phase 5: “Plateau of Productivity”

Finally, the real market appears, and the use of technology begins to develop itself and is eventually adopted by an “early majority”. The criteria for viability become clearer, the pertinence of the innovation is more convincing and the financial profit arrives.

These five phases have different durations and magnitudes depending on the technology and the markets in which they appear. Certain products can reach a plateau of productivity in two years, some in 10 years and some can become obsolete before even reaching it. With experience, Gartner was able to define a hundred reference curves per technological sector: e-commerce, telemedicine, transport, software, etc., and of course Wearables and smart apparel.

2.4.1 Market scaling

The main challenge for new technology firms is crossing the chasm. Geoffrey Moore (1991) argues that the gap between early adopters and early majority provides a significant challenge for companies because these groups are quite different and require completely different marketing strategies. While early adopters are willing to try something entirely new, the early and late majority wants a product solution that is error free and full-featured. The only way to cross the chasm is to put all your eggs in one basket, meaning that the strategy should be to identify a niche segment among the early majority and focus all efforts on developing the whole product solution by serving this particular segment. When this particular customer niche has adopted the solution, the firm can focus its effort on a second customer niche. The key to getting a foothold in the mass market is to use the initial customer segment as reference customers. Thereby the firm can start shaping all marketing communications to position itself as a market leader to derisk mass market adoption. The key to successfully redefining the market or create a market leader position is to choose an unoccupied space where there is a legitimate market need (Furr et al., 2014; Moore, 1991).

To attract the first customer niche among the early majority, the firm should focus on developing a minimum awesome product (MAP) (Furr et al., 2014). While the MVP is used for validating the core assumptions during the initial stages among the innovators and early adaptors, the MAP is a solution that is extraordinary on the dimensions that customers value the most (Fig. 2.1). The point is to use the MVP to improve the key attributes of the solution that can evoke positive emotions and thereby turn it into an MAP by focusing on the functional, social, and emotional dimensions of the solution. While using the MAP to get the solution adopted by the early majority, the firm can move to a second customer niche. In parallel to that, the MAP is further developed into the whole solution that could address the needs of the main market.

4.4.2 Modeling risk premiums and social and technological learning

Rogers (2003) distinguishes consumer segments along a normal distribution of adoption propensities. EAs have high initial adoption propensities and so high risk tolerance; early majority (EM), late majority (LM), and laggards (LG) are increasingly risk averse and have low initial adoption propensities. Based on this conceptualization, Pettifor et al. (2017) calculate initial RPs as a measure of adoption propensity for each of the four different adopter groups. Their RP estimates are based on discrete choice experiments that provide willingness-to-pay (WTP) estimates for new technologies, such as BEVs, for which limited market data is available. Pettifor et al. (2017) use a normal distribution of WTP point estimates from discrete choice studies to calculate a mean RP (x¯RP) with associated standard deviation (σ¯RP) for different adopter groups. Negative initial RPs indicate attraction to new technologies (risk-seeking), and high positive initial RPs indicate aversion to new technologies (risk-aversion). Following Rogers (2003), the EAs1 occupy a 16% market share; both the EM and LM account for 34% of the market; and the LG the final 16%.

Pettifor et al. (2017) also use a metaanalysis of 21 empirical studies to measure the effect of social influence on vehicle purchase propensities. They find that for every one standard deviation increase in market share, RPs decrease by 0.241 standard deviations, which increases vehicle adoption propensities [95% CI (0.157, 0.322), Z=5.505,|P|<.000]. In other words, RPs decline as market share grows, using market share as a proxy for social influence. In the vehicle choice model of IMAGE the RPs (in $/passenger km) for each consumer group have been added to the travel cost. More details on the empirical analysis and the implementation in IMAGE are provided in “Supplementary Materials” of the study by Edelenbosch et al. (2018).

Besides SL dynamics, this study focuses on TL of the battery costs and distinguish between exogenous and endogenous learning scenarios. Battery costs in electric vehicles (EVs) have declined rapidly over recent years (Nykvist and Nilsson, 2015); therefore the battery costs start from a cost estimate of 300 US$/kWh in 2014 (Nykvist and Nilsson, 2015). In the exogenous cost scenario, we assume that battery costs could reach 125 $/kWh by 2025 (Faguy, 2015) and decline further to 100 US$/kWh over the course of the century. In the endogenous cost scenario, we use a learning rate of 7.5% (uncertainty range from 6% to 9%) in line with estimates from the literature (Nykvist and Nilsson, 2015). We also assume a floor price of 50 $/kWh, affecting the purchase cost of plug-in electric vehicles (PHEVs), BEVs, and fuel cell vehicles (FCVs). More widely used components of cars such as the car frame or engine are not assumed to be influenced by learning after many years of experience and so follow the same path as in the exogenous scenario.

4.4.2.1 Model setup

Consumer heterogeneity, TL, SL, and policy measures, can all influence vehicle choice. Fig. 4.2 demonstrates schematically how these processes are related in the model setup within the IMAGE transport vehicle choice module, a global integrated assessment model used for this analysis (Girod et al., 2012). Increased market share affects SL and TL for different adopter groups: EA, EM, LM, and LG.

Figure 4.2. Schematic overview of the dynamic relationship between technological learning, social learning, and market deployment of new technologies. Four adopter groups are distinguished: EA, EM, LM, and LG. At a given time point, all four groups face the same technology cost but different monetized risk premiums. Net perceived costs therefore differ per group, with the lowest perceived cost vehicle selected by the cost-minimizing decision algorithm, resulting in changes to market share, which in turn stimulates further technological and social learning. EAs, Early adopters; EM, early majority; LG, laggards; LM, late majority.

Source: From Edelenbosch et al. (2018).

5 Simulation

Each agent in the model executes its behaviors once every simulation tick. A simulation tick represents one day.

Before performing the simulation, instances of OPA are created following the distribution of the Rogers categories: innovator (2.5%), early adopter (13.5%), early majority (34.0%), late majority (34.0%), and laggard (16.0%). Advertisement of the new oil-recovery scheme is set to happen until the end of the simulation. The maximum capacity of all collection boxes is assumed is equally set. Currently, only one OCA per region is considered with a fixed capacity and collection frequency. When visiting a collection site, if collecting the oil of that site results in an excess of capacity the collection is not performed but after discharging the oil at the biodiesel processing plant.

Know your tools

My involvement in embedded systems has always been on the debugging tool side of the industry, so I am understandably focused on this dimension of the process. One of the biggest issues that I saw and experienced many times over was the inability of our customer to understand how to properly use our tools for maximum advantage.

I’ve seen it with really senior engineers and with my students. Of course, we the tool vendors have to share some of the blame because we’re the ones who created the tool and then did not provide adequate documentation to make it easy to understand how to use the tool to its fullest extent.

In the introduction to the book, I mentioned Hansen’s Law. However, because most people never read the introduction, let me summarize it again. John Hansen was a brilliant HP engineer whom I had the privilege to work with in the Logic Systems Division in Colorado Springs. He said,

If a customer doesn’t know how to use a feature, the feature doesn’t exist.

It’s a very simple yet extremely insightful statement about designing complex products and the need to be able to simply convey their usefulness to an end user.

In his seminal book, Crossing the Chasm [6], Geoffrey Moore looks at the marketing of high-technology products. Moore identifies a fallacy in the traditional way that high-technology markets are modeled. Consider the traditional life cycle model for the adoption of a new product in the marketplace, shown in Fig. 4.4A. We see each segment of the market occupying a portion of the area under the bell curve. The area in their segment represents the potential sales volume for that market. I think we can easily identify the characteristics of each segment.

However, for the successful marketing and sales of new technology-based products, Moore argues that this model is wrong. He argues that there is a fundamental gap, or chasm, that exists between the early adopters and the early majority. Referring to Fig. 4.4B, we see that the segments comprised of the early and late majorities are the bulk of the market. Therefore, while initial sales to the “techies” might be very gratifying, those sales can’t sustain a successful product for very long.

Moore says that the early majority are the gatekeepers for the rest of the market. If they embrace the product, then the product can continue to grow in sales and market impact. If they reject it, then it will die.

In order to be accepted by the early majority, there are several key factors that must be in play, but I’ll just zero in on two factors that I think are germane to the point I’m trying to make.

The early majority tends to seek validation of a product’s value by seeking the recommendation of other like-minded members of the early majority whose judgment they trust.

There must be a “complete solution” available for the product.

The second bullet is the one that is relevant here.

As developers of new technology, we are constantly coming out with newer and better solutions to meet the needs of our customers, who are themselves developing new and innovative products. These customers, the early majority, don’t have the time or desire to put up with the glaring omissions of a new product’s support infrastructure. Manuals with errors, lack of technical support, training are all unacceptable show-stoppers for the early majority.

So, what’s this got to do with “know your tools?” We, the R&D engineers, can provide all the features in the world to make our in-circuit emulator or logic analyzer more compelling, but if a customer can’t take advantage of the feature set because it is too difficult to learn, or they don’t have the time to learn, then the tool is not what they need.

Yes, the tool manufacturer needs to bring every technology transfer best practices to bear to make the features of their products accessible and easy to understand. In their defense, I will say that the tools today are much better than when I was working in the field. This is primarily due to the additional processing power built into the tools and the amount of memory that even the most modest of instruments carries within. Rather than trying to find it in the manual, I can press a context-sensitive help button and see the manual page that I need to access.

But… the onus is still on me to devote enough time to learn the tool. If I’m always too busy to learn how to take advantage of my tools, then I have no one to blame but me for not surviving the next round of layoffs. Just remember the parable about the wood cutter who was always too busy chopping down trees to sharpen his axe and then couldn’t understand why he was not able to cut as much wood as he needed to.

Understanding your tools extends beyond a knowledge of how to use the feature set. It also includes an understanding of how the tool interacts with the environment that it is being used in.

In my Introduction to electrical engineering class (Circuits I),we cover the topic of the D’Arsonval meter.j For those readers who never used an analog multimeter, the basic meter consisted of a meter movement that could deflect to the full scale of its range with only microamperes of current flowing through the meter coil. So, for example, if you have an analog meter that will deflect full scale with 10 μA of current and you place that meter in series with a 10 MΩ resistor, you now have a voltmeter that can measure 10 V full scale with a 10 MΩ input resistance. Of course, the meter’s windings also have a resistance that usually is part of the circuit calculation. The point of this lesson is to sensitize the student to the interplay between the circuit under observation and the tool being used to observe the circuit.

Another exercise that we go through in class is the difference between accuracy and resolution. Your digital multimeter may be able to resolve the voltage down to ± 1 mV, but the accuracy of the meter may only be trustworthy down to ± 15 mV on the 10 V range. As experienced engineers, we know this, but students constantly fall into the trap of accepting without question the reading on the meter.

Oscilloscopes can also be a significant perturbation to a circuit as well as a source of error in their own right due to the differences in bandwidth among the probes. I recall a student asking me why the signal amplitude of a pulse train was so low. I poked around and sure enough, the pulses should have had an amplitude around 5 V, but the scope registered below 500 mV.

I then started looking at the scope setup and I noticed two things:

The scope input was set for 50 Ω.

The probe was set at 1 × attenuation.

In effect, when the student probed the circuit node, he was hanging a 50 Ω resistor to ground on the node.

This is all part of the educational process and it is why labs are such important parts of an engineering student’s education, even though they tend to complain about the time they have to spend in the labs. I often wonder if metrology should be a required course in an EE’s curriculum rather than an elective course. In lecture, the students could learn the theory of measurement and their lab experiments could demonstrate the practical side of measurement instruments, such as the proper way to use them and how to understand and interpret the results.

On the digital side, the logic analyzer has always been the premier measurement tool, although that dominance may be waning (more on that in a later chapter). However, the logic analyzer (LA) can be a very intimidating instrument to learn the basics on and even more intimidating to learn how to use it to solve the really tough problems.

Trying to observe 100 or more signals going in and out of a surface-mount integrated circuit with pin spacings of 0.5 mm and clock rates of 500 MHz is not something that you can do by clipping 100 flying leads to the IC. In these situations, the measurement tool is an integral part of the system and the system must be designed from the start with the logic analyzer interface designed into the hardware. Typically, this will require that the first PC board be a “throwaway” and only used for development. Fig. 4.5 [7] illustrates the necessity of planning ahead.

This circuit adapter provides 16 input channels and 100 kΩ isolation (see equivalent load diagram in the lower right). This probing solution is recommended for normal density applications (parts on 0.1 in. centers) and where speed is not a significant issue. Keysight offers additional probing solutions and detailed application literature for probing high-speed and high-density circuits. The Keysight probes are also designed to mate to high-density surface mount connectors made by Mictor and Samtec.

However, in most cases, these solutions also require that the probing adapter be built into the PCB and be part of the planning process from the outset. Once the circuit is thoroughly debugged and characterized, the probes can be removed in the next revision of the board.

14.3.2.1 About the theory

The diffusion of innovation (DOI) theory explains how a new behavior or product spreads through a social system, such that people in that social system start to adopt and accept this new behavior or product (Rogers, 1982). The prerequisite to diffusion is that the new behavior or product has to be perceived as something innovative. The DOI outlines five adopter categories, named according to the timing in which people start to adopt the innovation: (1) innovators, (2) early adopters, (3) early majority, (4) late majority, and (5) laggards. Similar to the TTM, these categories are associated with certain characteristics, but the DOI and TTM are different in that the DOI’s categories are distinct, whereas people cycle through the TTM stages.

Innovators are the earliest adopters of innovation, characterized by people who are interested and open to trying something new. Not much is required to get the innovators on board with an innovation. Early adopters are those who understand that change is necessary and are willing to embrace innovations. As this group of people is already motivated to change, they only require instructions on the implementation portion (i.e., a how-to guide) to get started on trying out the new behavior or product. The early majority are willing to adopt an innovation only when convinced that the innovation is effective in achieving a desired outcome. Hence, appealing to this group requires empirical evidence to prove that adopting the innovation is indeed advantageous. The late majority have their doubts about innovation, preferring to wait and observe the previous adopters of innovation before evaluating whether to jump onto the bandwagon. This group requires the success stories of many prior adopters to accept the innovation. The laggards are the last group to adopt innovation. This group is resistant to change and is difficult to persuade. Strategies to appeal to this group would include normative influences, fear, and evidence of effectiveness.

15.4.1 Diffusion of Innovation

The underlying theory of much of modern marketing is built upon the work of Everett Rogers, which spanned the 1950s to 1980s and is detailed in the book Diffusion of Innovations. Rogers studied the factors that influence how new innovations (e.g., products, ideas, movements) spread throughout a population. Rogers’ work has many dimensions, but one has become the cornerstone of marketing a new product; the diffusion of innovation curve, as shown in Figure 15.7. The blue bell-shaped curve is meant to represent the percentage of customer archetypes for a particular innovation. Rogers hypothesized that there would be a small number of Innovators, many of whom were involved in the creation of the innovation. There are a few more Early Adopters (a term coined by Rogers) who would readily adopt a new innovation. The bulk of a population, however, would be made up of an Early Majority and Late Majority. Finally, there are the Laggards who will only accept a new innovation after it becomes difficult to resist any longer. In the context of medical devices, these potential adopters might include patients, caretakers, healthcare providers, or purchasing agents.

Rogers documented the evolution of an innovation spreading from the Early Adopters to the Early Majority to Late Majority and Laggards until all capturable parts of the population had adopted. This progressive adoption is captured in the yellow S-shaped curve, as more and more people adopt the innovation. Although the bottom axis is not time, it does roughly mirror time as more and more people adopt the innovation. The maximum market share per cent (the y-axis) is 100%, but this does not mean 100% of the overall population has adopted; most new innovative medical devices are not intended to diffuse to an entire population. Rather, the goal is to capture 100% of the available market, which is usually a small subset of the overall population. In the healthcare ecosystem, the reachable market often includes a complex and varied range of healthcare providers.

Rogers used the diffusion curves to derive several additional theoretical frameworks that have greatly influenced modern marketing. We only cover one; how an individual moves throughout the decision to adopt. The model is shown in Figure 15.8 and includes five steps. First, an individual becomes aware of the existence of the innovation. Second, more is learned about the relative advantages of adoption. Third, a cognitive decision is made to adopt, but the adoption has not yet taken place. Fourth, the innovation is tested. Last, after a trial or demonstration, a decision is made whether to continue to use the innovation (to become an advocate) or discontinue use (to become a skeptic). Rogers imagined a potential adaptor to progress through these steps in stops and starts. Furthermore, a potential adopter can exit the adoption process at any point. As a potential adaptor progresses from stage to stage, she is reevaluating the value of adoption.

Where marketers can intervene is in the type and timing of information that is presented to individuals. Rogers hypothesized that certain channels for information, shown on the left side of Figure 15.8, would be most effective and efficient at particular points in the decision-making process. To build awareness and persuasion, mass media (e.g., radio, television, billboards) could be used. However, as an individual progresses, she requires more and more detailed and trusted information and therefore will turn to local networks, peers, and mentors.

Over the past few decades, the internet has changed the way some elements of marketing occur. For example, the distribution of helpful materials (e.g., user training, repair manuals), advertising (e.g., pop-up ads, targeted emails), customer service (e.g., chat lines, online FAQs), and customer feedback (e.g., unfiltered customer comments) all have a significant online component. These online elements have added several new channels that can help (or in some cases hinder) how a customer navigates the decision-making pathway toward adoption.

Several features of modern marketing can be derived from this decision-making framework. For example, when an individual adopts a product, it would be most helpful if he could help recruit others. One way to accomplish this is to imprint a catchy logo, name, or phrase (the essence of branding) on a product as an outward display of adoption. On a daily basis, we encounter hundreds (if not thousands) of branded products, which builds our awareness of the adoption of others.

11.6 Conclusion

This chapter has shown that expecting sustainable construction technology adoption to happen serendipitously in a complex and fragmented construction industry is fraught with uptake problems unless situational context is provided. Adoption can be piecemeal and may have limited overall impact. This chapter has served to test, refine, and apply situational context concerning the adoption of sustainable construction technology surrounding massive timber construction. Whilst it clearly has strong sustainability credentials that resonate across the full breadth of the supply chain, its uptake varies as seen through the lens of Rogers’ s-curve. For instance, central Europe is probably best described as being in the “early majority” category with a well-developed supply chain and export capability. Australia is still in the “first movers” category and this implicates the need for technology supply from Europe and reliance on early risk takers who create demand for adoption of the technology in Australia. For these companies, interest in massive timber construction revolves around more holistic business models that not only incorporate sustainability but an emphasis on design driven prefabrication as well. Related features include use of digital technologies as well as compressed and integrated supply chains.

To date, the gestation period for massive timber construction in Australian has occurred over some 4–5 years. Clearly, it takes time to develop local expertise, networks, regulatory standardization, and client interest given the risks in balancing sustainability benefits against traditional construction methods. Landmark buildings create market momentum but it is still difficult to make use of this momentum where there is insufficient local supply chain development for the client base to fully commit to the new technology. In some ways, this is the danger period for technology adoption as an unwanted lull can stall uptake momentum. There is a need to create traction quickly; otherwise, the cost benefits expected of economies of scale cannot be realized.

The next step in moving massive timber technology adoption forward is likely to be stronger development of typical or standardized construction solutions that makes client decision-making simpler and more automatic in the face of competing with traditional construction methods. Economies of scale will then follow. It would seem that a simplified compliance pathway is a good sign in supporting this cause. Even so, the role of landmark buildings will be ongoing in drawing attention to the new technology, which will itself serve to increase adoption.

3.4.2 Future market segments

According to several analyses, the early markets or market niches for electric vehicles will be commercial applications in fleets or car pools (Plötz et al., 2013). The reasons are the high mileage per year, appropriate driving range, frequent and predictable routes, as well as the possibility to charge at the depot of the company. On a TCO basis the economic competitiveness of electric vehicles will be achieved at an early stage. Furthermore, according to the analyses of McKinsey (2010b) cities will be the regions where BEVs will be used at first. The reasons are the early adopters for small city cars, who prefer image and environmental friendliness to range and costs. In some cities, such as in China, the limited access for fossil fuel cars and two-wheelers or, in London, the free access to the city (no congestion charge) drives the development of electric vehicles.

Private demand for the vehicles will be spurred by early adopters in the beginning as well. But the development path and the speed to reach an early majority will be determined by the dynamics of the drivers and how barriers can be overcome, which both are country-specific. Pure electric vehicles might profit from changes in mobility behavior, personal values, or the cross-linking of services (inter-modality, web applications, etc.) which change customer requirements.

Based on the customer requirements, today’s plug-in hybrids and range extender vehicles have major advantages. Nevertheless, the role of PHEVs and REEVs is very ambiguous, whether the vehicles will be only a transitory technology or remain holding a higher or lower market share. One major competition for this technology could come along with the successful market introduction of fuel cell vehicles. We don’t want to construct a competition between both technologies because being open to new technologies is a key to innovation, but both address similar customer requirements and market segments, and a coexistence with BEV is more realistic in this context. Fuel cell vehicles can be defined as REEV too (see Chapter 1) and have similar advantages, primarily being able to offer longer driving distances and low fueling times. Fuel cell vehicles (FCEVs) have a shorter history, especially a high dynamic/hype beginning in 1997 with the ZEV-program in California until 2006, when commercialization did not take place and led to disappointments (Bakker and Budde, 2012). The main reasons for this disappointment were slow technological developments (e.g., cold start) and high costs for the fuel cell as well as the dependence on a significant amount of complementary infrastructure (the “chicken-egg-dilemma”), logistics, and fuel stations. Although some progress has taken place in the last 5–10 years, there are still only 216 fueling stations and approximately 500 FCEV on the roads in demonstration programs worldwide (LBST und TÜV Süd, 2013; Bünger, 2013). Major OEMs, gas companies, politicians, and others signed several agreements, forming, for example, H2 Mobility in Germany or the United Kingdom to concentrate market introduction activities. The goal for Germany is to build up 400 H2 fueling stations (50 today) and introduce several thousand FCEV until 2023 (Daimler, 2013). Because of the high market entry barriers and correlated uncertainties it is very hard to estimate market potentials. A study for the European market investigated by a coalition of science and industry estimated that a coexistence of BEV and FCEV can be achieved until 2050 starting between 2020 and 2030 (McKinsey, 2010a).

9.3 The problem: How to engage consumers in the energy system transition?

As a part-solution to achieve climate neutrality and associated goals, active consumer engagement is called for in current energy and environmental strategies (Blättel-Mink, 2014; Pitt et al., 2017; Schot et al., 2016; Ympäristöministeriö, 2017). Active consumers and prosumers can contribute to electrification by producing, self-consuming, and storing electricity, as well as by buying and using EVs. As with sustainable innovations in general, the prosumers have to be many and come in many forms to have an actual impact on the system.

A growing solar energy prosumer base depends on affordable solar PV panels and the introduction of policy instruments such as FITs creating economic incentives for the diffusion of solar PVs, EVs, and other RETs. In the absence of incentives progress is typically slow. Box 9.1 provides an example of challenges for the prosumer base in Finland, where so far only modest support exists for solar energy diffusion among small-scale producers (Haukkala, 2019; Kotilainen, 2020).

Box 9.1

Scenarios for the prosumer base development in Finland

In order to shape the transition and accelerate it, the prosumer base must be substantial (Foxon and Pearson, 2008; Hillman et al., 2011). Finland has a small prosumer base (Ahola, 2019) and lacks powerful economic incentives, such as FIT, to residential solar PV production, while no official targets to increase solar production exist (Haukkala, 2019). Homeowners do get tax relief for the installation costs of solar PV systems, estimated to decrease the cost by a maximum of 10% (FinSolar, 2019). Incentives in Finland to install solar PV in apartment buildings are very modest, but legislative changes were underway in 2020 to enable tax and distribution fee exemption for energy sharing within a building. Moreover, the residential solar PV payback times are still long compared to most markets in Europe.

In a study conducted in 2017, Finnish energy experts were interviewed in order to build scenarios for the prosumer base development in Finland. It was concluded that prosumer base growth would both benefit and challenge the Finnish energy system. One of the clear benefits would be increased amount of RES and distributed generation (DG), as well as new business opportunities for different stakeholders. The perceived challenges were related to the threat of potentially uncontrollable behind-the-meter systems that could destabilize the whole electricity system; commercial problems to the incumbent energy sector firms, and a potential off-grid movement led by increasing amount of consumers (Kotilainen, 2020). The experts did not, however, anticipate a rapid growth of the prosumer base in Finland until the 2030s. The table below presents brief summaries of prosumer base development scenarios in Finland based on these interviews. The experts deemed the “Prosumerism stalls” the most likely scenario within this time frame and in the current regulatory structure.

The research furthermore identified factors potentially affecting the development of the prosumer base in Finland, shifting it either toward prosumerism as an integral part of the energy system or toward an off-grid movement challenging the electricity system and markets. The factors identified were further evaluated against their potential impact and uncertainty. The most critical uncertainty factors concerning the prosumer base growth related to the developments in politics and regulation, new energy technologies (especially storage), and consumer influence.

What is then required to support small-scale prosumers' RES adoption? For consumers to evolve into prosumers, appropriate enablers are required (IEA-RETD, 2014). Most technology enablers, such as solar PV, are widely available, but the prosumer base is far from its full potential (Kampman et al., 2016; Sajn, 2016). In order to produce, sell and store energy, and to participate in flexibility schemes and innovation activities, a mix of different types of enablers is needed. The International Energy Agency's Renewable Energy Technology Deployment (RETD) program identified some of these: technology, economic, behavioral, and national conditions (IEA-RETD, 2014) (see Fig. 9.1). We next review the constraints related to the economic, behavioral, technological, and policy dimensions and introduce some policy instruments to tackle these.

9.3.2 Behavioral constraints

Environmentally sustainable behavior typically requires the individual to make an additional effort. For example, recycling, using public transport, and energy conservation at home necessitate adapting one's daily routines. Research on consumer motivations to invest in energy production is hence a popular research theme. In their study on microgeneration adoption, Balcombe et al. (2013) considered motivational aspects such as finance, environment, security of supply, uncertainty and trust, inconvenience, and impact on residence. From a more theoretical point of view, intrinsic and extrinsic motivations (Ryan and Deci, 2000) have been found to have an influence on technology adoption. Moreover, sociological research on consumers' decision-making and pro-environmental behavior has focused on the value-belief-norm (VBN) theory (Stern et al., 1999), theory of reasoned action (TRA) (Fishbein and Ajzen, 1975), or theory of planned behavior (TPB) (Ajzen, 1985). What the theories have in common in predicting behavior is the integration of beliefs, attitudes, perceptions, and subjective norms. The technology acceptance model (TAM) originally developed by Davis (1989), Davis et al. (1992) uses a similar approach whereby the perceived usefulness and ease of use are pivotal and can to some extent predict intentions to use a new system. While originally tested mostly in the context of adoption of IT products, the TAM—and its many versions—has also been applied more recently to energy and sustainability research (e.g., Alam et al., 2014; Broman Toft et al., 2014; Chin and Lin, 2016; Naspetti et al., 2017).

People are furthermore motivated by different drivers. The theory of the diffusions of innovation by Rogers (1962), explains how innovations become adopted in temporal sequence by several adopter groups (see also Fig. 9.2). Innovators and early adopters adopt the innovation first and then the subsequent adopter groups follow: early majority, late majority, and laggards. The early adopters are more relaxed about using immature products and pay less attention to cost than do the later adopter groups. While early adopters keenly promote innovations, most innovations fail to cross “the chasm” (Moore, 1991) between the early and late markets. The products offered to the later adopter groups need to demonstrate concrete value and ease of use. Pro-environmental innovations have demonstrably special characteristics in terms of their diffusion since consumers are sometimes prepared to pay more for environmentally beneficial innovations. To get to this point, there may be a need for policy support to kickstart the diffusion, especially among the early majority adopters.

Figure 9.2. Diffusion of innovations by adopter groups (Rogers, 1962). Chasm between early adopters and early majority (Moore, 1991).

Based on Rogers, E.M., 1962. Diffusion of Innovations, first ed. Free Press of Glencoe, New York, NY; Moore, G.A., 1991. Crossing the Chasm. Harper Business Essentials, New York.