Subscription based business models have been around for ages. Newspapers and magazines pioneered the business model and have been relying on it ever since. However, in recent times the subscription model has seen a resurgence in an unexpected manner. A subscription model makes sense for a newspaper due its perishability and the recurring nature of its demand. Products such as shaving products, food and household products are increasingly more often sold on a subscription basis. Dollar shave club, an American men’s razor company that pioneered the selling of razor blades on a subscription basis, was acquired by Unilever in 2016 for a billion dollars.

Dollar shave club was founded in January of 2011 and grew to its billion dollar valuation in merely 5,5 years. This goes to show how disruptive subscription models can be to traditional brick and mortar and click and mortar dominated industries. Part of the success of the subscription business model in industries like the shaving industry is because of the predictable and recurring demand patterns products like razors have. Established companies that sell products with similar demand patterns are quick to adapt to the disrupting trend, offering subscriptions next to their individual sales. Amazon for example sells household products on a subscription basis, and even offer a discount for those who subscribe. Gillette started its own shave club as a response to the disruptive dollar shave club, offering its range of products on a subscription basis. The big brands respond in an attempt to preserve their market share.

One major advantage of the subscription business model is the elimination of the middle man. Which, in theory, allows for undercutting the competition. Companies that have managed to sell directly to consumers have traditionally been able to offer a better value proposition to the consumer. A noteworthy example of this is computer manufacturer Dell that started selling their personal computers directly to consumers around 2007, eliminating the need for retail stores.

By subscribing a customer effectively declares intention to buy a certain product on certain dates. This makes it very easy for the supplying company to forecast demand, resulting in a lower carrying costs of inventory and, in case of a highly perishable product such as food, a lower cost due to deterioration. Predictability of demand may not sound like a noteworthy advantage, but for a company like amazon that generates close to 200 billion dollars in revenue per year inventory management is one of the biggest business challenges.

Not only does the subscription model provide predictability of demand, it also provides an increased payment safety. Many subscription services require direct debit payments. This results in a more stable and more reliable cash flow for the company. Whereas large retail companies have to devote a substantial amount of resources to the collection of revenue, sometimes going as far as establishing debt collection departments. Collecting revenue from customers that haven not paid their bill not only takes time and effort, sometimes businesses are forced to settle for less than the original amount or to completely absolve a bill. With a subscription based business model this problem is less likely to occur and the costs associated with this problem are likely less severe.

Subscription boxes with a return policy is the subscription business model 2.0?

Subscription models that offer a themed box of products with changing contents on a predetermined schedule have been around for a while. This business model has many of the same advantages for consumers as the traditional subscription business model, with the added benefit of being exposed to new products that you otherwise would have missed out on. For the subscription box provider the added benefit is that companies might be willing to provide free of highly discounted promotional products to put into the subscription box. Subscription boxes are a new but growing business. Ipsy, a beauty related subscription box has over 2.5 million paying subscribers that are paying a monthly fee to receive beauty related products.

These subscription box services operate in a very similar way as the previously mentioned subscription based services. But recently subscription box providers in the fashion industry have been offering ways to personalize the content of subscription boxes, and have been offering refund policies. The personalization of subscription boxes is an interesting development that also seems to occur outside of the fashion industry. Vinebox for example offers a questionnaire that allows customers to discover their taste in wine, and adjusts the content of the subscription box to the taste of the customer. Trunk club offers the assistance of a personal stylist to pick out fashionable clothes that fit your style and body. These attempts to personalize the subscription boxes require the cooperation and attention of the customer, the customer needs to invest time into the customization. In return there is an increased chance that the customer enjoys the content of the box.

Trunkclub also offers a return policy, that allows customers to ship back unwanted items from their subscription box. This is odd because a return policy might eliminate one if the main advantage of the subscription based business model, namely the predictability of demand. This trend is most prevalent among the fashion themed subscription boxes, which is not surprising. Trunkclub for example is owned by Nordstrom, a large retail company in the fashion industry. Returned items could potentially be sold through the retail branch of the parent company before they go out of fashion.

Sources:

https://www.fastcompany.com/3068822/beauty-subscriptions-are-a-tough-racket-but-ipsy-is-sitting-pretty

https://www.reforge.com/blog/ipsy-content-influencer-growth

http://www.trunkclub.com

https://www.bloomberg.com/news/articles/2016-07-20/why-unilever-really-bought-dollar-shave-club

The researchers of this paper (David et al, 2017) conducted three tests to find out what the impact is of customized pricing on consumer evaluation, considering consumer’s interpersonal attachment orientation. First of all, they decided to run a survey with 172 U.S. participants. Half of them were randomly assigned to a situation in which they paid the shelf price, others also paid the shelf price or a lower, customized price. After manipulating the  prices for the consumers, they measured their level of satisfaction. They found significant evidence that there is a variation in the consumer evaluation of prices, in case of customized pricing. Thereby, this relation is influenced by someone’s attachment style, namely secure or anxious. It appears that anxiously attached consumers were satisfied paying the shelf price, whereas secure consumers were dissatisfied paying the shelf price. To find an explanation for this phenomenon, a second study was conducted with 270 students as participants. The results supported the earlier findings and showed an explanation for the mentioned phenomenon, namely that customized pricing programs create an enhanced expectation of receiving a discounting price among the secure consumers. This means that these customers are increasingly price sensitive and this could be disadvantageous for retailers, as this could affect their profits. As more and more retailers use customized pricing techniques based on the purchase pattern of their customers nowadays, this has implications; customized pricing is not always effective. A third and last survey extended the findings of the previous tests, as it showed that in the presence of customized pricing programs, securely attached consumers expect to receive discounted prices that are lower than the prices that other customers have to pay. Anxiously customers on the other hand are merely satisfied paying the shelf price despite the attendance of customized pricing systems, unless they are in a disadvantaged situation, meaning another customer receives a lower price.

As earlier research mostly highlighted the benefits of customized pricing as it encourages consumer satisfaction and purchase likelihood (Van den Bos, Peters, Bobocel and Ybema, 2006 & Xia and Monroe, 2010), the results of this study showed that there is also a downside to this sort of customization.  It is indeed the case that consumers positively evaluate customized pricing in which they receive advantage, but the effectiveness appeared to be heavily influenced by interpersonal attachments, as expectations are created among secure attached consumers for a discounted price. On the other hand, anxious attached consumers are decreasingly price sensitive. Altogether,  this paper did not only found the relationship between customized pricing and consumer valuation, which is moderated by interpersonal attachment, but they also found the specific explanation how that is possible, namely the mentioned increasing expectations. Knowing all this now, what are the (practical) implications now for managers for example, so that this information can be used in their advantage? For marketers, it is important information as understanding when and for which customers the customized pricing is effective. Thereby, it is likely that this also applies to other forms of customized offers and services, so not only customized pricing. Managers should segment their customers based on their attachment style. For example, they can offer customized products best to older people, who are securely attached and have a higher level of income, meaning they should try to individualize their marketing approach.

A strong point of this research is the fact that they conducted three tests, with many different participants, that all found the same significant results of the impact of customized pricing and the role of interpersonal attachment. This means there is strong evidence and the validity is high in my eyes. It has huge implications for managers and marketers, as they can adjust their way of customization now on an individual basis to enlarge the effectiveness. One of the weaknesses are the fact that only Americans were involved with the studies, while people from collectivistic cultures such as China might think different about customized prices, so that external generalization is more difficult. Another point of improvement is to have more data on the actual numbers a retailer could benefit from applying their implications. No real existing examples were mentioned, which could be beneficial to really test these implications and to know how much per cent their sales would increase for example.

Article:

David, M.E. Bearden, W.O. Haws, K.L. (2017) Priced just for me: The role of interpersonal attachment style on consumer responses to customized pricing. Journal of Consumer Behaviour

References :

Van den Bos, K., Peters, S. L., Bobocel, D. R., & Ybema, J. F. (2006) On preferences and doing the right thing: Satisfaction with advantageous inequity when cognitive processing is limited. Journal of Experimental Social Psychology, 42, 273-289

Xia, L., & Monroe, K. B., (2010) Is a good deal always fair? Examining the concepts of transaction value and price fairness Journal of Economic Psychology, 31, 884-894

In general web personalization has the capability to achieve two important business goals: increase advertising revenue and increase sales revenue. The realization of these two business goals is directly related to item sampling and item selection. Where Item Sampling takes the form of a user’s clicks on personalized recommendations, Item selection involves the user choosing one of the personalized recommendations as the final choice.

In this paper by Ying Ho and Bodoff (2013) a scientific gap is identified which limits the understanding of how web personalization can be used to increase advertising and/or sales revenues. In order to fill this gap the researchers try to develop a theoretical model of these behaviours, item sampling and item selection, and their attitudinal antecedents. Additionally, this model is based on the integration of two theories:  the Elaboration Likelihood Model and the Consumer Search Theory.

Elaboration Likelihood Model

The Elaboration Likelihood Model, ELM for short, models the effects of a user’s elaboration of individual persuasive items on his or her overall attitude. The model thus shows how a user’s elaboration, defined as the extent to which a person carefully thinks about an argument (Petty and Cacioppo 1986a), of individual recommendations influence a user’s attitude towards the personalization agent, in turn influencing the users decisions to select a personalized recommendation (Petty and Cacioppo 1986a, 1986b; Tam and Ho 2005).

In lay man terms, with the Elaboration Likelihood Model one is able to define the degree to which a user will cognitively process a given recommendation. This is quite limiting as it does not allow for the investigation of the number of recommendations the user investigates, something important if the business goal is to maximize user clicks. Hence the ELM model is combined with CST, Consumer Search Theory.

Consumer Search Theory

Where ELM is limited to individual recommendations the CST model models the umber of a items a user inspects on his way to the completion of a search task. Thus, in CST the item source is seen as a distribution where each act of inspecting an item is viewed as sampling an item from the original distribution. Where ELM thus focusses on an individual recommendation CST is able to capture the entire array of inspected elements prior to the completion of a search task. This breadth of sampling of CST and depth of processing from ELM allows for the exploration of influence attitude formation, which in turn affects item selection.

Studies

In order to develop a theoretical model of item sampling and item selection, the researchers employ both a lab and a field study. In the lab study a personalized online bookstore is developed and used in combination with a thought-listening technique in order to capture participants’ depth of processing. Despite being a widely used method in order to capture ELM and depth of processing thought-listening technique is an obtrusive measure and hence a field study was used to increase external validity.

During the field study the researchers collaborated with a large music provider in the Asia Pacific region in order to develop a personalized music website. During this 6 month period participants visiting this music website either received good or bad recommendations and where hence tracked on their behaviour.

Results

As visible in Table 1. one can see that for hypothesis 1-5 there are consistent results for both the lab and field study. Whereas hypothesis 6 was significant in the lab study such a significance could not be observed in the field study. This difference in observation however, could be explained due to the large difference in available alternatives and hence a possible higher required attitude confidence for  significance to be visible.

Apart from acknowledging several hypotheses the researchers also discovered two significant predictors of item section they had not hypothesized about. First, the NFC (Need for Cognition), the extent to which individuals are willing towards cognitive activities, exerts a negative effect on the person’s actual selection of a personalized item in the lab study. Second, attitude confidence, the degree of certainty a person has in a choice, has a positive effect on item selection in both the lab and field studies.

Strengths and Weaknesses

First of all, in the lab study the researchers made use of a though listing method, this method although widely used might have altered participants behaviour and thus jeopardize the result of hypothesis 1. In order to compensate for this result, the researchers used a second, field experiment, to increase external validity. Apart from using a thought listing method the researchers also made sole use of experience goods in their study. Arguably they did make use of these goods because clicks are meaningful in such a setting, however one may argue this limits generalizability as the developed model might be inconsistent for other types of goods.

Sources:

Ho, S. Y., & Bodoff, D. (2014). The effects of Web personalization on user attitude and behavior: An integration of the elaboration likelihood model and consumer search theory. MIS quarterly, 38(2).

Lavie, T., Sela, M., Oppenheim, I., Inbar, O., and Meyer, J. 2010. “User Attitudes Towards News Content Personalization,” International Journal of Human-Computer Studies (68:8), pp. 483-495.

Petty, R. E., and Cacioppo, J. T. 1986a. “The Elaboration Likelihood Model of Persuasion,” Advances in Experimental Social Psychology (19), pp. 124-205.

Petty, R. E., and Cacioppo, J. T. 1986b. Communication and Persuasion: Central and Peripheral Routes to Attitude Change, New York: Springer.

Tam, K. Y., and Ho, S. Y. 2005. “Web Personalization as a Persuasion Strategy: An Elaboration Likelihood Model Perspective,” Information Systems Research (16:3), pp. 271-293.

Article discussed: Schemmann, B., Herrmann, A. M., Chappin, M. M., & Heimeriks, G. J. (2016). Crowdsourcing ideas: Involving ordinary users in the ideation phase of new product development. Research Policy, 45(6), 1145-1154.

Introduction

Crowdsourcing ideas from customers is a contemporary phenomenon that has enjoyed widespread attention from both scholars and business practitioners. These events, albeit temporary or long-term, generate countless of ideas for the focal company, but only a handful are actually implemented. This is particularly true when ideas originate from ordinary users as opposed to experts, or lead users. It is thus pivotal to scrutinize this ordinary user group in order to improve efficiency of the yet expensive and time-consuming filtering process.

With this in mind, Schemmann et al. (2015) conducted a cross-sectional research in an effort to derive ideator and idea-related characteristics that could explain whether a crowdsourced idea from ordinary users is implemented. Based on prior studies and their analysis, they conclude that the following factors all positively relate to actual implementation:

• Attention paid to crowdsourced ideas of others (H2)
• Idea popularity (H3)
• Idea potential innovativeness (H4)

Synthesis of theory: value from juxtaposing user bases

The most interesting theoretical contribution of this paper stems from its focus on the ordinary user, contrasting most existing research, which emphasized the role of lead users. That is, whether ideator and idea-related characteristics can determine actual adoption by the organization. Motivation is found to result in more successful ideas due to individuals being stimulated to develop a qualitative strong idea (Bayus, 2013). However, no such finding is found to be significant within the current research. In addition to the ideator-based characteristic, idea-related characteristics also influence the eventual adoption of the idea. Whereas lead users are valued for their knowledge and expertise, ordinary users leverage an even more powerful mechanism called Wisdom of the Crowds (Surowiecki, 2004). The sheer number of ordinary users results in an extremely accurate prediction of whether an idea is likely to be successful or not. Furthermore, given that innovative ideas are positively related to implementation (Witell et al., 2011), we can infer that the increased diversity within ordinary users is likely to produce more novel ideas than lead users.

Such a juxtaposition of user base enables firms to tailor crowdsourcing initiatives to their respective objectives and move away from crowdsourcing failure rates that have been observed to be as high as 40% (Castellion & Markham, 2013). That is, firms can tweak two mechanisms that naturally emerge from this study’s insights. First, crowdsourcing participants can be limited to a certain user base. Secondly, crowdsourcing platforms can be designed to includes functions such as peer evaluations and idea visibility. The latter brings us to the study by Bockstedt et al. (2016) who distinguishes between blind and unblind innovation contests. Specifically, unblind contests are likely to enhance the positive relation of hypothesis 2 as the platform facilitates learning from others.

Methods used: weaknesses and improvements

The present study was conducted on a sample of 1456 ideas for which the company’s review process was completed. The data stems from an open idea call of a large beverage producer, which was publicly available. Although such a method bears generalizability issues, it can still be justified from a practical perspective. However, in addition to the limitations stated by the authors, another weakness relates to the codification of variables. That is, all variables are examined as dichotomous variables, with exception of ideator motivation. All hypotheses measured with dichotomous values are found to be significant, whereas the continuous variable used to test hypothesis 1 is rejected. Not only does such an outcome arouse suspicion, there is much room for improvement. As illustration, the independent variable attention paid to other ideas could also be coded as a continuous or count variable, measured by relatively time spent on other’s idea page or the number of other’s idea pages visited, respectively. Although arguments for both methods are evident, the authors could at least include such an empirical analysis as a robustness check. As a result, findings become more credible, increasing the likelihood of actual utilization when managing crowdsourcing platforms.

Conclusion

Thus, the present study presents insights into ideator and idea-related characteristics that determine idea implementation in long-term crowdsourcing initiatives. By analysis of ordinary users, the authors complement existing views with another relevant dimension that could help business practitioners manage difficult socio-technical dynamics inherent to crowdsourcing. The higher success rate of crowdsourcing is in turn, likely to spur innovation across a multitude of industries. Although such aspirations seem distant, achieving great feats is realized by taking one step at a time.

Bibliography

Bayus, B.L., 2013. Crowdsourcing new product ideas over time: an analysis of the Dell IdeaStorm Community. Manage. Sci. 59 (1), 226–244

Surowiecki, J., 2004. The Wisdom of Crowds: Why the Many are Smarter than the Few and How Collective Wisdom Shapes Business, Economies, Societies, and Nations. Doubleday, New York, NY.

Witell, L., Löfgren, M., Gustafsson, A., 2011. Identifying ideas of attractive quality in the innovation process. TQM J. 23 (1), 87–99.

Castellion, G., Markham, S.K., 2013. Perspective: new product failure rates:influence of argumentum ad populum and self-interest. J. Prod. Innov. Manage.30 (5), 976–979.

Bockstedt, J., Druehl, C., & Mishra, A. (2016). Heterogeneous submission behavior and its implications for success in innovation contests with public submissions. Production and Operations Management, 25(7), 1157-1176.

There have been many research developments in the field of crowdfunding, in the past decade. While the focus for IS practitioners has been on optimising crowdfunding efforts in order to reach a specific measure of success (usually financial success eg: target funding amount), less focus has been placed on identifying the psychological and motivational queues that lead individuals to contribute monetarily to campaigns.

What makes people want to contribute in the first place, to a crowdfunding campaign?

Zvilichovsky et al (2017) have devoted their research efforts on investigating the motivational factors that lead individuals to contribute to crowdfunding campaigns. Their research analyses how consumers increase their participations when they believe their contribution is pivotal to product creation. In crowdfunding platforms, consumers consciously and willingly fund a product that does yet no exist in the market. The essence of crowdfunding platforms is to involve individuals in early stages of the product development process. This context sparks consumer motivations that transcend traditional transactional interactions.

Rather than being motivated by a sense of generosity towards the laborious efforts of the entrepreneur, consumers are driven by the notion of seeing the product become a reality and become in possession of it. The authors conceptualise making-the-product-happen motivation as an attempt by consumers to contribute to the creation and commercialisation of a product by means of monetary funding and making-it-happen motivation as an attempt by consumers to contribute to the realisation of the campaign’s financial goal. When consumers feel that the existence of the product depends on the success of the crowdfunding campaigns and that their financial backing is pivotal to the creation of the product, making-the-product-happen motivation peaks.

The authors conducted a total of four studies controlling for future product availability, funding mechanism and funding stage target to provide empirically validated answers to the question above.

• The first study, tests whether all-or-nothing funding mechanisms vs keep-it-all funding mechanisms see differences in the willingness to back propensity of consumers. The participants were provided information regarding an air cleaning system and asked using a 7-item likert scale how likely they would be to pay $50 to back the campaign. The result of the first study determined that all-or-nothing funding mechanisms drove the participants’ making-it-happen motivation and willingness to back the campaign.

• The second study tested whether consumers were more willing to back a campaign that was closed to reaching its target vs one that had already reached it target. The results found that consumers’ perceived impact on making-the-product happen, led to a higher willingness to participate in a project whose financial target is close to its goal.

• In the third and fourth study, the researchers tested whether future availability of the product would affect their willingness to contribute. They found that when future availability of the product depends exclusively on the success of the campaign (thus that the backer’s financial donations are pivotal in making-the-product happen) consumers are more willing to back the campaign.

To test for generalisability the authors of the study conducted a field experiment on 193,312 campaigns by analysing the average contribution per backer compared to the length of the campaign.

Apart from validating the results found in the lab study, they discovered that most of successful campaigns achieve their goal only by a slight margin above the target (computed to be $195 or three $65 donations) indicating the importance of communicating the importance of the backer’s role in product realisation process.

The authors found very interesting insights that serve as insights to campaign designers to optimise their probability of success. If entrepreneurs are able to effectively and sincerely communicate to consumers the importance of their role as backers in making-the-product happen then the results of the study indicate a positive correlation with financial success (for the entrepreneur) and future product availability for the consumer.

While the study did provide insights as to which motivational factors lead consumers to back campaigns, it is not short with its limitations. The data collected from Kickstarter does not report the amount contributed by backers, which had to be assumed by the authors. Furthermore, the results applied only to consumer products and not to public goods, who see crowdfunding as a viable source of funding.

The above limitation set a precedent in the direction that future research should take in further developing the crowdfunding literature. The researchers hint that studying the interactions that occur in the post-crowdfunding stage would provide a more complete portrait of crowdfunding literature.

Zvilichovsky, D., Danziger, S. and Steinhart, Y. (2018). Making-the-Product-Happen: A Driver of Crowdfunding Participation. Journal of Interactive Marketing, 41, pp.81-93.

Nowadays firms have increasingly adopted personalization strategy to provide customers with the option to choose on important feature parameters at the price similar to the standardized products. For example, Dell is well known for its success in mass customization. (Randall, et al., 2005) Subway is also a kind of customization and received huge success. (Choi & LEE, 2015) However, the academic defines the personalization in a different way. It has been suggested as a revolutionary approach to market segmentation. The company now treat the individual customer as individual segments by satisfying their very specific need. The strategy can boost sales and in such bring competitive advantage to the company.

However, it is not necessarily always true for every company. There are many attributes can influence the attractiveness of personalization option, for example, price premium, effort and monitoring quality. If we take a closer look in the effect of effort that the customer needs to take in the personalization process, we can find that the more complex the personalized process is, the less the customer will likely to personalize the item. (Choi & LEE, 2015) Especially the company failed to convey information to the different type of customer. Just like the example of Dell, the inexperienced user cannot tell the meaning of “Memory”. (Randall, et al., 2005) Some of those customers just decided not to make the choice because the complexity exceeds the optimal level.  The product utility that consumer could get through personalization is not strong enough to cover the inconvenience.

This research argues that the customers prefer standard products over the personalized product when the range of personalization is perceived as excessive. The customer is more likely to select standard products over personalized alternatives when faced with inordinately complex decision-making. In order to test the pattern of consumer responses regarding the personalized product and the standardized product, the author decided to use the attitude of customers towards product and the purchase intention as the key factor.

H3: Customers will demonstrate a more positive attitude for standard products over personalized products with a higher level of personalizing options beyond a customer-perceived optimal point.

H4: the customer will demonstrate a higher purchase intention for standard products over personalized products with a higher level of personalizing options beyond a customer-perceived optimal point.

The hypothesis is tested by using stimulation experiment. There are 195 undergraduate students in Korea are recruited. They have been randomly assigned to one of the seven sub-group according to the extent of personalization options. The basic setting is the standardized product, and the other six have the different level of personalization in an ordinal ranking. The incremental increase in the personalization level allows the author to compare between groups and examine the relationship between personalization level and customer response.

The result shows that the customer product attitude shows an inverted U shape as the number of personalizing options increased. This is also the case for the other factor, purchase intention, as the number of personalization option increase, the purchase intention increase until the optimal level and then decrease. (Figure 1 and Figure 2)

The author then conducted an ANOVA analysis on the product attitude and purchase intention respectively for the standardized product and the personalized product (with most extensive personalization options). The results support the Hypothesis 3 while the result of Hypothesis 4 was not statistically significant. To test the mechanism of complexity as a mediator, they also test the hypothesis based on Baron and Kenny (1986). The perceived complexity mediated the effect of product type on product attitude. The mediation tests show that perceived complexity is the influence factor for the preference over standard products and the personalized products when comparing the standard product with the overwhelming personalization options.

The logic of this research is very easy to follow, they successfully demonstrated the importance of setting an appropriate level of personalization to companies that wanted to implement the personalization business model. However, the question arises when evaluating the regression model. Has the author controlled enough factors to isolate the effect of independent variables on the dependent variable? For example, if we look at the product nature, the result doesn’t necessarily apply to the experienced user with the clear expected product. They will likely to appreciate the extensive personalization options to fine tune the product to maximize the fit with their expectation. In addition, the author used to watch as the testing product. It is a relatively straightforward functional product. The personalization option does not include any configuration regarding the function itself but only the design. Will the research show a different result if we test it on Dell computer? The answer is unknown. If we look at a further research in the background of web companies. There is a phenomenon called “filter bubble”. It suggests that with those filter bubbles people are restricted to the filter options and shapes our view to the worldview.  If you are interested, please have a look at the following video.

Bibliography

Choi, J.-E. & LEE, D.-H., 2015. Customers do not always prefer personalized products: The role of personalized options range in personalization. International Academy of Marketing Studies Journal, 19(2).

Randall, T., Terwiesch, C. & Ulrich, K. T., 2005. Principles for User Design of Customized Products. California Management Review, 47(4), pp. 68-85.