acm-header
Sign In

Communications of the ACM

Virtual extension

Application Service Providers: Market and Adoption Decisions


View as: Print Mobile App ACM Digital Library Full Text (PDF) Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook

IT outsourcing has been a viable business strategy since the late 1950s when companies began to invest significant resources in outside computing services. As IT managers increasingly used time sharing, contract programming, and facilities management, the next few decades saw accompanying increases in efficiency in software applications and services, not to mention decreased operations costs. In 1989, when Kodak outsourced its entire IT department to IBM and EDS, the business world had to scramble to get over its initial shock. Seemingly overnight, "software" and "service" had become strange bedfellows. The rules of the game had radically changed.

Since the mid-1990s, the explosion in Internet connectivity, increased bandwidth, and the ubiquitous nature of computing have made the delivery of software applications from remote data centers both technologically feasible and economically attractive. A relatively new form of IS outsourcing, Application Services Provision (ASP), provides applications to multiple entities from its data center across a wide area network. Similar to the telephone services model, clients pay a monthly fee to access online applications from an ASP. Advances in encryption and firewall technology remove a client's security concerns for using remotely hosted services. Active research on ASP outsourcing (also called NetSourcing) has attracted academic scholars and business practitioners alike,5,10 but few have investigated the specifics of the market and the decision process to adopt the ASP model.

ASPs today are quite different from what they were five years ago, in large part because this dynamic industry has experienced significant merger and acquisition activity. Forty percent of ASPs active in 2001 were no longer in business in 2004.11 Traditional software companies also have jumped on the ASP bandwagon, developing new technologies to deliver on-demand services. The scope of hosted applications for ASPs has also shifted significantly as certain applications were scrapped for lack of profitability, for example, overall enterprise system hosting with no specific industrial solutions. However, a Forrester report2 predicted that as eCommerce continued to grow 30% - 40% in 2005, the demand for licensed software products would decrease while demand for hosted services increases. This prediction seems to be reality, as ASP hosting solutions are increasingly viable and attractive business models.

Thus, to update our knowledge of the rapidly changing ASP market and to help practitioners make better-informed ASP adoption decisions, this article addresses two central questions: What is the structure of the current market? And what factors should clients consider in their ASP adoption decisions?

Back to Top

The ASP Market

The market supports various types of ASPs, but all share the same basic characteristics: 1) online service delivery; 2) usage-based payment; and 3) one-to-many service model.

Between 2003 and 2005, the list of top ASP players changed radically, primarily due to mergers and acquisitions.1 Many new faces appeared in 2005 as well; including Omniture and Crown-Peak (see Table 1 for comparison).

This continuous juggling makes it difficult to determine who the viable ASPs are. Hence, classification of the ASP space is necessary to better understand the ASP market. In this paper, we distinguish between ASPs according to market focus and application characteristics (see Table 2). This classification is based on over four years of observations, surveying, and personal interviews with ASP founders and director-level managers responsible for the ASP business. A brief discussion of these categories follows.

Pure ASPs. Pure ASPs are companies whose major business is online hosting services. They tend to be recently emerged, small or medium-sized companies. Hence, they tend not to be financially strong and will likely be acquired. Pure ASPs either purchase specific software licenses or partner with software vendors to lease their own applications or solutions to clients. Pure ASPs include two subcategories: horizontal ASPs and vertical ASPs.

Horizontal ASPs provide standard applications online to clients across industries with little or no customization. Payroll processing and employee activity management are common examples (such as ADP). Online conference services, such as those provided by California-based WebEx, also require little customization. This strategy can best achieve economy-of-scale and prove to be more profitable and manageable than other approaches.

Vertical ASPs offer applications designed for a specific industry, such as law, healthcare, or hospitality. For example, Partners.com mainly provides the healthcare information system hosting for hospitals; Statability mainly provides online reporting for hotels. Vertical ASPs can leverage deep knowledge of a specific industry and special product features to gain strategic advantage.

Independent Software Vendors (ISV) are software companies that develop and host their own software applications. They are typically not "first mover" in the ASP market but they tend to be well-established, top-tier software vendors, strong in software development and support, and financially healthy. High familiarity with their own software enables ISVs to readily provide application customization and integration with bolt-on solutions. ISVs seem to enter the ASP market for tighter association with small or medium-sized customers, as the director of a leading software vendor said, "We simply feel we deserve a piece of the pie." ISVs may grow a new ASP business within the existing organizational structure or partner with many small ASPs to provide applications.

We distinguish between two types of ISVs. In general, horizontal ISVs provide a broad suite of products and services or specialized applications across industries. When a client tends to outsource a large package of applications (such as an IT department), it may negotiate all necessary services from a single horizontal ISV rather than multiple small ASPs. Microsoft and IBM are obvious examples.

Vertical ISVs provide industry solutions to customers within one industry. A typical example would be ERP vendors that provide industry-specific solutions for advanced planning and optimization of enterprise operations. Another example is Blackboard, which provides course management systems to education industry.

Back to Top

Factors that Influence ASP Adoption

Many companies have been confronted with ASP adoption decisions, and these decisions will grow increasingly important over the next few years. Focusing on the key factors that influence these decisions, we surveyed over 200 decision makers across all industries in medium and large companies. We also conducted personal interviews with ASP managers, founders of pure ASPs, and directors of ASP centers at large enterprise software vendors. Based on personal interviews, survey responses, and current market observation, we developed the following ASP adoption decision recommendations:

  • Recommendation 1:
      Examine closely the uncertainty in your internal environment; the more frequently your internal operations incur unexpected change, the greater the tendency to favor ASP adoption.

When environmental uncertainty is high, the traditional IS outsourcing literature does not generally recommend outsourcing. However, because we distinguish between internal uncertainty (experienced by the organization) and external uncertainty (experienced by the market economy), we find that when uncertainty (unpredictable change) governs internal operations (such as, fluctuating customer demands and material sourcing), managers believe that ASP use results in greater cost benefits. About 35% of survey participants who expected high internal uncertainty preferred the ASP model. These clients believe that the use-based payment and short-term contracts allow them to benefit from adjusting the scope and/or usage of rented applications according to a change in demand and/or supply. A nationally ranked university we studied decided to outsource the course management system to Blackboard to contend with demand uncertainty in student accounts management and faculty requests for features and functionality.

  • Recommendation 2:
      Carefully consider the trade-off between enjoying the benefits of adopting an ASP versus increasing vendor dependency.

Companies who resist ASP adoption are often concerned with vendor dependency. To offset this fear, managers advise piloting a partial outsourcing arrangement so that both sides can test the viability of the approach. Many ASPs will offer a 3- or 6-month trial before requiring an extended contract (such as 12 years) with clients. ASPs discourage application customization during this trial period to avoid sinking unrecoverable labor-related costs into the new customer. Statability, for example, allows only renewing customers to customize an applicationand then only with a higher priced, extended contract.

  • Recommendation 3:
      Examine closely the customization requirements and scope of each application.

Standardized software applications with little customization characterize the vast majority of ASP hosting today. These standard applications allow a relatively easy transition from in-house systems to outside vendor.7 On the other hand, we found that many companies with special customization requirements strongly desire ASP solutions. These companies are primarily interested in reducing maintenance and in-house development costs.

The vast majority of ASP hosting activity has evolved around applications with narrow, shallow scope (such as, email) as opposed to broad, deep applications (such as, ERP). Most managers surveyed (58%) preferred a step approach of adopting point solutions (e.g., HR, CRM) instead of broad, sweeping solutions (such as, ERP across the board). Only a small fraction (3.6%) would consider the latter approach because a wide application scope requires a more comprehensive evaluation of ASPs, from technology and business capabilities to their financial strength.

  • Recommendation 4:
      Carefully evaluate ASP's financial stability and consider your exit strategy.

Pure ASPs have proven to be successful after surviving years of aggressive merger and acquisition activity, but it is prudent to proceed cautiously with such relationships. Even stable ISVs have disappeared due to consolidation. For example, PeopleSoft acquired JD Edwards, but then succumbed to a hostile takeover by Oracle which then acquired Siebel and Retek. Unpredicted bankruptcy, merge or acquisition introduces hidden costs to the client. The clients may be forced to use unfamiliar vendors with possible higher fees and unpredictable services quality.3 Clients are also concerned with support and integration issues after an acquisition. Companies should expect to deal with continuous vendor uncertainty in this space for the foreseeable future.

In order to minimize the risk, as an important component of entrance strategy, a client needs to carefully evaluate the financial viability of the ASP before using ASP services. Large data center, well-maintained facilities and knowledgeable professionals might be the endorsement for financial robustness.

Clients also need to develop exist plans. One exit strategy involves extending the company's knowledge base in areas such as system optimization and maintenance, through knowledge transfer from the ASP to the company. This strategy increases the chance of a smooth transition back to an in-house solution. In fact, 75% of survey participants indicated that knowledge transfer is a practical strategy to compensate for internal deficiency, and is often a factor in the adoption decision. Another option is to keep an eye on alternative ASPs with similar application services. No matter which option will be adopted, the company needs to develop a plan for data management when the relationship with an ASP terminates that includes: how to return the data back to the client, how to destroy the copies of data stored at the ASP site, and how to transfer the data from one ASP's data center to another ASP's data center.

  • Recommendation 5:
      Carefully develop a relationship with your potential ASP manager (preferably executive level) to build trust.

Companies should trust their ASP along multiple dimensions: reliable services, confidence and integrity of management and staff, and responsive support and maintenance. We found that initial trust tends to improve the chances of a successful outsourcing relationship. This initial trust before the formal contract could be built based on the intensive reference check on an ASP's existing customers and comprehensive application demonstrations from ASP's on-site visits. These activities enhance a client's understanding on an ASPs' service capability and the fit between application functions and the client's requirements. We also found that social and personal relationships between client and ASP managers significantly increases client's trust, and facilitates a more personal handling of ASP issues. For example, the manager on the client side might expect (and receive) the personal involvement of the ASP's CIO to either avoid or resolve a problem.

Table 3 summarizes four success cases with ASP services: two vertical ASPs, one horizontal ASP, and one vertical ISV. We profile certain characteristics of these relationships (Economic, IT deficiency, and Trust) based on public information and personal discussion.

  • Recommendation 6:
      Expect the ASP concept to become increasingly viable and desirable over time as new technologies and business models alleviate current concerns.

Managers should consider two factors before adopting an ASP. First, enterprise software applications are quickly being commoditized. ERP systems (for example, back office) no longer provide competitive advantages through significant customization, because distinctions between ERP vendors have dwindled.6 Recent evaporation and consolidation of these software companies clearly indicate that ERP systems have matured; other applications such as customer relationship management, supply chain management, and portals are soon to follow. The customer bases are saturated. Many customers are more concerned with rising maintenance costs.6

Second, advances in technology promise to better facilitate the software-as-a-service concept. Service-oriented computing architecture (see October 2003 issue of this journal) and web services make business processes accessible within and across enterprises. IBM worked with Visa and its affiliated banks to create a web services system for call centers, allowing the agents to transmit credit card disputes directly to both Visa and their bank. As application development technology, web services could also be used by the ASPs to provide system-independent applications to customers.

Grid computing is another popular concept. It turns networks of computers into a single virtual machine to take advantage of the idle time available within this group of processes. Besides its application in scientific collaboration and finance (analyzing stock portfolios), a potential commercial application is inventory management, where large amounts of point-of-sale data must be collected and aggregated into a data warehouse, possibly distributed across multiple sites and accessed by multiple suppliers. However, a number of technical challenges have yet to be resolved: how to integrate heterogeneous processing and applications computing resources, and how to standardize communication protocols. In addition, grid computing charges customers according to the computer processor that runs the application. Software vendors seem less than keen to move beyond the traditional pricing model.9

Forrester describes a rather comprehensive infrastructure, referred to as Organic IT, for automatically managing and sharing virtualized software, processors, storage, and networks across all business services.4 This next generation infrastructure will allow providers to deliver application services with unprecedented speed and reliability.

IBM has perhaps been most successful at leveraging web services and grid technology to market and deploying the "on-demand" concept. In a personal conversation with Alfredo Gutierrez, an IBM manager, he says this is "an enterprise whose business processesintegrated end-to-end across the company and with key partners, suppliers and customerscan respond with speed to any customer demand, market opportunity or external threat". With these technologies, the basic ASP model will continue to improve, evolve, become more viable, and gain acceptance. In addition, the ASP model explores new billing options: fixed rental, time/materials usage based payment, or contract pricing (companies pay extra if they exceeds the contacted usage amount).8 Various billing options also make the ASP model more flexible and attractive.

Back to Top

Conclusion

The ASP business model and ASP players can help clients employ the most advanced technologies at relatively low costs. In the predictable future, as technology developments gallop ahead, the software industry should become increasingly service oriented, but clients need to examine their internal situations, the dynamic market, and potential ASPs comprehensively before deciding to outsource with an ASP. As services and players grow more diverse, ASPs will play more important roles in client organizations by leveraging their internal capabilities and achieving competitive advantages.

Back to Top

References

1. ASPnews, top ASP list found at www.aspnews.com.

2. Bartels, A. US IT spending benchmarks for 2005. Forrester Report, May 24, 2005

3. Barthélemy, J. The hidden costs of IT outsourcing. Sloan Management Review 42, 3, (2001), 6069.

4. Gillett, F. E. Organic IT 2004: Cut IT costs, speed up business. Forrester Report, May 18, 2004.

5. Lee, J., Huynh, M. Q., Kwok, R. C., and Pi, S. IT outsourcing evolution - past, present and future. Comm. ACM 46, 5, (2003), 8489.

6. Rettig, C. The trouble with enterprise software. Sloan Management Review 49, 1, (2007), 2127.

7. Rose, J.W., and Westerman, G. Preparing for utility computing: The role of IT architecture and relationship management. IBM System Journal 43, 1, (2004), 519.

8. Susarla, A., Barua, A. and Whinston, A. Making the most out of an ASP relationship. IT Professional 3, 6, (2001), 6367

9. Wailgum, T. Grid held hostage. CIO report, August 1, 2005 (found at www.cio.com)

10. Walsh, K. R. Analyzing the application ASP concept: Technologies, economies, and strategies. Comm. ACM 46, 8, (2003), 103107.

11. Young, A. Management update: Application outsourcing key trends and business drivers. Gartner Report, 20 August, 2003.

Back to Top

Authors

Yurong Yao (yyao@suffolk.edu) is an associate professor of Information Systems and Operations Management at the Sawyer Business School, Suffolk University, Boston, MA.

Edward Watson (ewatson@lsu.edu) is E. J. Ourso Professor of Business Analysis, Director of Flores MBA Programs and Interim Associate Dean of Graduate Programs at the E. J. Ourso College of Business at Louisiana State University, Baton Rouge, LA.

Beverly K. Kahn (bkahn@suffolk.edu) is an associate professor of Information Systems and Operations Management at the Sawyer Business School, Suffolk University, Boston, MA.

Back to Top

Footnotes

DOI: http://doi.acm.org/10.1145/1785414.1785447

Back to Top

Tables

T1Table 1. Top ASPs in 2003 and 2005

T2Table 2. ASP Categories

T3Table 3. Customer Success Cases with ASP Application Hosting

Back to top


©2010 ACM  0001-0782/10/0700  $10.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2010 ACM, Inc.


 

No entries found