Pricing strategy for cloud computing: A damaged services perspective Jianhui Huang a , Robert J. Kauffman b , Dan Ma b, ⁎ a The Corporate Executive Board, Singapore b Singapore Management University, Singapore abstract article info Available online 28 November 2014 Keywords: Cloud computing Damaged services Pricing strategy Services interruption Spot prices How effective is a hybrid pricing strategy for a cloud computing services vendor that mixes fixed-price reserved services with spot-price on-demand services? This research offers a decision support model to create the appro- priate strategy for IT services based on prior research on information goods, electricity pricing, product versioning, and revenue yield management. The goal is to establish whether interruptible spot-price on- demand cloud computing services — which we view as damaged services — are valuable to the vendor. The results from the analysis of an economic model show that a hybrid strategy outperforms a one-service-only strategy in most cases, especially when clients are sensitive to services interruptions or when task values are highly differ- entiated. A more intriguing finding is that a vendor should permit the possibility of services interruptions even when clients are highly sensitive to their occurrence. The presence of interruptions serves as a quality differentiator between the on-demand services and reserved services, assuring the efficacy of the hybrid strategy. Moreover, a vendor may use capacity limit, in the hybrid strategy, as a tool to further improve its profit. To our knowledge, this research is the first to propose the damaged services perspective as an analogy for damaged goods in the cloud software market. © 2014 Elsevier B.V. All rights reserved. 1. Introduction Cloud services vendors deliver IT resources and software application services via the Internet. The services are scalable and accommodate fluctuations in client demand [2]. The cloud services market has grown rapidly over the past decade. According to Gartner [12], revenue in the global cloud services market was US$111 billion in 2012, a 21.4% increase from US$91.4 billion in 2011, and it is expected to reach US$206.6 billion by 2016. 1 Cloud services suit clients with unpredictable demand for comput- ing power and large batch processing tasks [2]. Flexible provision of ser- vices and usage-based pricing are key enablers [43]. As prices are driven lower [16], clients will rely on cloud computing for all IT-related services. Cloud services are more economical than in-house systems, especially for clients with data-intensive computing. The cost savings can be 95%, an indicator of the high value it provides [44,19]. Various pricing mechanisms have been adopted in the cloud market. 2 Since cloud services are consumed similar to utility services such as electricity or water, most vendors have applied usage-based pricing with services charged by the hour or minute, and client pay- ments are tied to actual usage. Clients, however, have shown concern, since it is difficult to calculate total cost [21,22]. Innovative pricing schemes have been implemented recently. For example, Amazon intro- duced its Elastic Compute Cloud (EC2) in 2006, and charged hourly fees. In 2009, Amazon employed a new pricing method: clients could pur- chase a reserved-services contract by paying an upfront fee. Thereafter, they were allowed to access the IT resources as reserved-services in- stances. By 2009, Amazon was delivering its EC2 services as spot-price on-demand instances. Spot prices often are lower than reserved- services instance prices to encourage use of on-demand services. Ama- zon retains the right to interrupt running tasks and takes back the IT re- sources without notifying clients. The completion of spot-price on- demand services is not guaranteed. Such interruptible spot-price services are an inferior version of interruption-free reserved services; clients will assign a lower value to Decision Support Systems 78 (2015) 80–92 ⁎ Corresponding author at: 80 Stamford Road, School of Information Systems, Singapore Management University, Singapore 178902. Tel.: +65 68280926. E-mail address: madan@smu.edu.sg (D. Ma). 1 Three main types of cloud computing services initially characterized the market: infrastructure-as-a-service (IaaS), platform-as-a-service (PaaS), and software-as-a-service (SaaS). As the market matured, more categories emerged, such as data storage-as-a-service, hardware-as-a-service, desktop-as-a-service, business processes-as-a-service, and others [32]. 2 There are quite a few recent useful survey articles on the pricing of cloud computing services. The research covers: the economics of cloud computing and pricing [18]; the de- sign of a pricing service for grid computing [6]; pricing models and their relationship to the quality of cloud services [1]; fixed and variable cost analysis for monopoly cloud services vendor who offers subscription and pay-per-use pricing [7] with two-part tariffs [8]; and strategic differentiation for cloud vendors with pricing approaches, including price clarity and transparency [20]. http://dx.doi.org/10.1016/j.dss.2014.11.001 0167-9236/© 2014 Elsevier B.V. All rights reserved. Contents lists available at ScienceDirect Decision Support Systems journal homepage: www.elsevier.com/locate/dss