All posts by mvanspaendonck


To Keep Or Not To Keep: Effects of Online Customer Reviews on Product Returns

By Madeleine van Spaendonck (365543ms)

In the US, the current average return rate for products bought online is approximately around 30% of purchases (The Economist, 2013). Most returns take place due to customers’ negative post-purchase product evaluation rather than product defects. One factor that is found to have an impact on this is the role of Online Consumer Reviews.

This is what Minnema et al. (2016) investigated in their study “To Keep or Not to Keep: Effects of Online Customer Reviews on Product Returns”. It uses a multi-year (2011-2013) dataset from a European online retailer that offers both electronics and furniture products. The paper examines the impact of three OCR characteristics (valence, volume and variance) on return decisions (figure 1). The researchers evaluate the net effect of OCRs, looking at its influence on both purchase and return decisions.

Screen Shot 2017-03-10 at 23.41.04.png


The hypotheses examined are based on the ‘expectation disconfirmation mechanism’. Post-purchase satisfaction results from the combination of customer expectations formed at the purchase-moment, product performance, and the difference between them. Negative expectation disconfirmation therefore decreases satisfaction, leading to a higher return probability. Therefore, higher expectation levels should lead to higher purchase and return probabilities, while higher expectation uncertainty should lower these.

Main results

Figure 2 presents a summary of the results of the study.

Screen Shot 2017-03-10 at 23.42.40.png

A particularly counterintuitive insight is that overly positive review valence (whereby the current OCR valence is higher than the long-term product average) leads to not only more sales but also a higher return probability. A potential reason for this is that OCRs induce the customer to form product expectations at the moment of purchase, leading to higher purchase probability. However, high expectations due to overly positive reviews may not be met. This leads to negative expectation confirmation, which then leads to higher return probability. Review volume and variance mostly affect purchase decisions, having little to no effect on product returns.

Strengths, Weaknesses and Suggested Improvements

While the majority of scholarly work in this field focuses on OCRs effects on product sales, this paper also addresses the lack of understanding of its effects on product returns. Taking into account both aspects is vital, because the prediction of OCR effects on retailer performance may be overly optimistic or pessimistic if only its effects on sales are considered. The study also shows that OCR effects advance beyond the moment of purchase and have the power to affect the decision to return a product. However, the model did not incorporate other information sources available at the purchase-moment that affect return-likelihood, such as product descriptions and pictures provided by the retailer. A comparative analysis could be used to evaluate whether reviews or retailer-provided information have the strongest impact on returns.

Managerial Implications

The study highlights the importance of considering product returns when evaluating OCR effects, as overly positive reviews may have negative consequences for retailers’ financial performance. Overly positive reviews, leading to more product returns, result in large reverse logistics costs. To reduce negative expectation disconfirmation, retailers should provide information and tools (besides OCRs) that allow consumers to set the right expectations and see if the product really meets their needs.


Minnema, A., Bijmolt, T.H.A., Gensler, S., Wiesel, T. (2016). “To Keep or Not to Keep: Effects of Online Customer Reviews on Product Returns.” Journal of Retailing, 92(3), pp. 253–267.

The Economist. (2013). Return to Santa. December 21, (latest accessed March 8, 2017), commerce-firms-have-hard-core-costly-impossible-please-customers- return-santa

Source for cover photo:

Ministry Ideaz, (2016), How do I return a product I no longer want? [ONLINE]. Available at: [Accessed 8 March 2017].

The Effect of Perceived Impact on Crowd-Funding Contributions

By Madeleine van Spaendonck (365543ms)

When deciding to fund a project on a crowd-funding platform, does it matter to you how close it is to its target? This is what researchers Kuppuswamy and Bayus (2017) investigated in their study “Does my contribution to your crowdfunding project matter?”. Prior scholarly work in this field has focused mostly on the significance of early contributions, and their ability to signal quality and lessen project uncertainty (Colombo et al., 2015). They found that people financially support projects when they believe their contribution will have an impact. Using a panel-data approach, the study examined 10,000 randomly-selected funded and unfunded Kickstarter projects (posted between 2012-2014), with the following variables.

Screen Shot 2017-03-05 at 21.08.58.png

The results reveal the following phenomena, forming the basis of the proposed “impact theory”:

  1. Additional backer support for a project will be higher as its cumulative funding approaches its target goal.
  2. Additional backer support for a project will drop sharply after the target is reached. After this point, people are likely to prefer other projects that do not have sufficient funding and where their financial help is thus perceived to have more impact. Results 1 and 2 combined form the ‘goal gradient’ effect.
  3. Moderating factors: this effect is strongest when backer support is likely to have the highest impact; this is when the project is close to its funding deadline, has a small funding goal, or has limited early support [figures 1, 2 and 3].

Screen Shot 2017-03-05 at 10.29.49.pngScreen Shot 2017-03-05 at 10.33.39.pngScreen Shot 2017-03-05 at 10.34.34.png

The figures illustrate U-shaped patterns for funding contributions over time. From a customer-centric perspective, backers are thus motivated by the target goal of the project and its proximity. For both pro-social reasons and the opportunity to receive the promised rewards, backers want the project to succeed (Gerber & Hui, 2013).

Strengths, Weaknesses and Future Research Directions

The paper offers new insights into crowd-funding behaviours by empirically studying its dynamics over time. For example, the ‘impact theory’ can explain the “Kickstarter effect”, which is the observation that more than 90% of projects that achieve at least 30% of their goal will eventually reach their target. People want to make an impactful contribution, which means that projects that are near – but not past – their target are most likely to receive support. Other crowd-funding phenomena, such as herding, cannot account for this by themselves. A weakness of the study is that the outcome measure is only focused on whether or not a contribution was made. To determine whether people voluntarily contribute more when they believe it will make an impact, a future research direction would be to measure contribution amounts.

Managerial Implications

The study highlights several practical implications for entrepreneurs. Setting the appropriate goal has a high impact on potential funding. A too-high goal makes it challenging to get close enough to the target for the goal gradient effect to arise. However, a too-low goal may prematurely halt contributions, because support declines after the target is reached. Furthermore, communicating the target goal and the goal process in the form of updates/reminders can increase contributions, as this also triggers the goal gradient effect.


Colombo, M., Franzoni, C., & Rossi-Lamastra, C., (2015). Internal social capital and the attraction of early contributions in crowdfunding projects. Entrep. Theory Pract. 39(1), 75–100.

Gerber, E., Hui, J., (2013). Crowdfunding: motivations and deterrents for participation. ACM Trans. Comput. Hum. Interact. 20 (6), 1–32.

Kuppuswamy, V., & Bayus L, B. (2017). Does my contribution to your crowdfunding project matter?. Journal of Business Venturing, 32(1), 72-89.

Cover photo:

Gil C. via for VentureBeat, (2017), Kickstarter Headline [ONLINE]. Available at: [Accessed 3 March 2017].

Sounds Like Music To My Ears

By: Madeleine van Spaendonck (365543ms)

The Problem Situation

Do you ever pay attention to the music you hear in your favorite store? Many shops and hospitality businesses in the Netherlands still make use of outdated mix-CDs and standard playlists. Considering it has become increasingly important for retail businesses to provide a dynamic brand experience, how can background music be used to optimize the customer journey?

Atmosphere and its business model

Amsterdam-based Rockstart-startup ‘’ addresses this situation with its new B2B music service, ‘Atmosphere’. Its key resource is its pool of musicians, DJs and producers, called ‘curators’. New clients undergo an extensive intake-procedure that allows Atmosphere to create a ‘music identity’ that reflects the company’s brand identity, target audience and desired customer experience. Consisting of a collection of moods, sounds and emotions, this allows the platform to match brands with the most suitable curators for them. Atmosphere allows curators to use the music on its platform to continuously assemble new playlists on a monthly basis. A streaming app is then used to play the music on-location. (Atmosphere, 2017)

Atmosphere’s value proposition is a better customer experience for brands and a new earning model for artists and music experts. It also incorporates feedback to create better playlists every month and learn from each brand profile to improve its services. Businesses pay Atmosphere on a monthly basis for using the platform, and the curators on the platform decide the price of their service.

Co-Creation Efficiency Criteria (Carson et al., 1999)

Atmosphere is a two-sided platform that connects retail/hospitality businesses with ‘curators’. The business model allows for joint profitability, as it enables businesses and curators to interact to create value together and maximize their payoffs. A study conducted by the Stockholm School of Economics found that background music that matches brand identity can increase store sales by at least 30% (Johansson & Moradi, 2015), which presents a measurable potential financial output for businesses. The curators suggested by the platform are picked by a particular company on the basis of the quality of their playlists and close fit with the brand, which incentivizes them in terms of effort to deliver the most suitable music sets and thus get rewarded in return. To facilitate this, Atmosphere invests in algorithms and models to create accurate brand profiles.

Furthermore, internal institutional arrangements are present in the form of ‘rules of the game’. Multiple curators are suggested to the client, who can make the final choice based on music samples. To stay ‘matched’ with a company, the chosen curator must continuously produce high quality work; otherwise, the company will switch to another curator. In terms of institutional environment factors, the legal environment of the business model poses the most significant threat, as songs are often copyrighted. Atmosphere has addressed this issue by acquiring the license rights for all the music that is available on its platform. This allows it to be used for commercial purposes. However, this is a continuous process; if Atmosphere wishes to attract and keep customers on its platform, it needs to constantly update its music offering.


Atmosphere. 2017. How We Work. [ONLINE] Available at: [Accessed 14 February 2017].

Carson, S. J., Devinney, T. M., Dowling, G. R., & John, G. (1999). Understanding institutional designs within marketing value systems. Journal of Marketing, 115-130.

Johansson, G., & Moradi, J. (2015). What Does Your Brand Sound Like?. [ONLINE] Available at: [Accessed 13 February 2017]. 2017. Company Web Page. [ONLINE] Available at: [Accessed 14 February 2017].

RetailTech. 2017. Artists Select Music For Retailers . [ONLINE] Available at: [Accessed 14 February 2017].

Silicon Canals. 2017.’s Atmosphere will find the right tunes for every company. [ONLINE] Available at: [Accessed 13 February 2017].