For the exclusive use of N. Roughgarden, 2015.
S
w
909E05
NUWAY SOFTWARE
R. Chandrasekhar wrote this case under the supervision of Professor Derrick Neufeld solely to provide material for class discussion.
The authors do not intend to illustrate either effective or ineffective handling of a managerial situation. The authors may have
disguised certain names and other identifying information to protect confidentiality.
Ivey Management Services prohibits any form of reproduction, storage or transmittal without its written permission. Reproduction of
this material is not covered under authorization by any reproduction rights organization. To order copies or request permission to
reproduce materials, contact Ivey Publishing, Ivey Management Services, c/o Richard Ivey School of Business, The University of
Western Ontario, London, Ontario, Canada, N6A 3K7; phone (519) 661-3208; fax (519) 661-3882; e-mail cases@ivey.uwo.ca.
Copyright © 2009, Ivey Management Services
Version: (A) 2009-04-15
In November 2008, Thomas Henshell, founder and president of Nuway Software (Nuway), sat at his desk
and wondered how to price Nulogic, the company’s new mobile software product. Until now, Nuway had
built a positive reputation among its customer base by focusing exclusively on customized mobile software
projects, each sold using a cost-plus pricing model. Such a pricing approach had ensured profitability so
long as Nuway consistently delivered projects on time and within budget. But Nulogic was different. In
essence, this internally developed application acted as a kind of “Swiss Army knife” enabling Nuway’s
developers to quickly insert common software routines into their mobile applications (as opposed to
developing each routine from scratch), thus saving time and money. The company reasoned that outside
mobile software developers would also find Nulogic valuable. Pursuing the Nulogic commercialization
opportunity would involve selling a generic packaged software “product” (rather than selling a customized
project) to a market of uncertain size (rather than to a known individual customer).
Did the cost-plus pricing model still make sense for Nulogic? Henshell pondered over the problem:
A review of the cost-plus pricing model seems necessary for three reasons. First, with
Nulogic being delinked from project sales and turned into a product, we will be looking,
for the first time, at economies of scale. We have not had economies of scale in the
projects business previously because each project is specific to the needs of a customer,
and the resources deployed to service a given project cannot be scaled up to service
multiple projects simultaneously.
Second, the mobile products and services market is changing. Insourcing will be on the
rise whereby customers will internally develop their own programs for mission-critical
requirements. This will result in a ready market for a basic platform that can be used as a
building block to develop specific applications and help reduce programming time for
customers. Nulogic is just that kind of platform. It will position Nuway to take advantage
of the new market trend.
This document is authorized for use only by Nicholas Roughgarden in Pricing Strategy – BMIM1 – Charlestown at Hult International Business School, 2015.
For the exclusive use of N. Roughgarden, 2015.
Page 2
9B09E005
Third, the demand for custom mobile applications software is so huge that Nuway has not
employed sales or marketing persons so far. But marketing becomes necessary when
Nulogic goes out as a standalone product. Providing a list price will then be imperative.
Against this backdrop, I’m not sure whether cost-plus pricing will do for Nulogic. We will
have to find a price that is fair to both Nuway and its customers. It should also send the
right signals of value. What is that price?
SOFTWARE INDUSTRY
The global packaged software industry was valued at US$262 billion in 2007. It was forecast to grow at
7.7 per cent for the next five years, contrasting sharply with the double-digit growth rates of the late 1990s.
The slowdown had primarily affected the demand for discretionary products and services; demand for
mission-critical software was not only strong but was likely to remain so for the foreseeable future. The
United States was the single largest consumer of software (see Exhibit 1). The industry had three market
segments (see Exhibit 2).
Because it was packaged both as a product and a service, software differed from most conventional
customer offerings. Software’s value was not amenable to easy quantification, but varied from customer to
customer, as did customers’ compelling competitive needs of the moment. Once software was developed,
the cost of additional units (or copies) of a software application was almost zero. If any incremental cost
was incurred, it was generally linked to the type of distribution channel. Retail prices of an application in
the software industry ranged from free to more than $500,000 (see Exhibit 3). The price range was
indicative of raw software costs and did not include the costs of add-ons such as maintenance and service.
The Internet was facilitating rapid expansion of a software development model called Open Source
Software (OSS). Using various electronic tools, volunteer OSS developers from around the world
collaborated by revising and reposting software source code and thus incrementally improved upon one
another’s work.1 The source code was referred to as “open” because it was freely available for others to
download, unlike proprietary commercial software code that was generally “closed” to anyone except the
software company that owned and sold the program. Some examples of highly successful OSS software
applications included the Linux operating system, Apache web server, Perl programming language,
MySQL database and Firefox web browser. Revenues from OSS desktop operating systems alone were
expected to grow at double-digit rates over the next few years and to garner a total market share of about
nine per cent by 2012.2 Although the initial software acquisition cost was usually zero, some corporate
users avoided OSS applications due to their inherent lack of warranty and vendor support. At the same
time, some firms sensed that the lack of support created a market opportunity, and so entered to provide
warranty and support services for popular OSS applications. For example, in 1994, Red Hat Inc. began to
offer warranty and technical support components for the Linux operating system.3 Where a strong OSS
alternative existed, commercial software vendors were under increased pressure to extend their pre- and
post-sales services, beyond provision of the software application itself.
1
Kim Johnson, “Open-Source Software Development,” http://chinese-school.netfirms.com/computer-article-opensource.html, accessed January 6, 2008.
2
Software Pricing Trends: How Vendors Can Capitalize on the Shift to New Revenue Models, PricewaterhouseCoopers,
2007, http://www.pwc.com/techforecast/pdfs/SoftwarePricing_X.pdf, accessed January 7, 2009.
3
“What Does Open Source Mean?” http://www.howstuffworks.com/question435.htm, accessed January 15, 2009.
This document is authorized for use only by Nicholas Roughgarden in Pricing Strategy – BMIM1 – Charlestown at Hult International Business School, 2015.
For the exclusive use of N. Roughgarden, 2015.
Page 3
9B09E005
In terms of product pricing, three models were common in the commercial software industry: license
pricing, subscription pricing and value-based pricing.
License Pricing
In license pricing, the vendor priced the software either per customer (for customized software) or per unit
(for software sold over the counter as a generic one-size-fits-all application). Payment was received onetime and upfront, and the customer acquired lifetime rights to use the product. Because the license never
expired, it was sometimes referred to as a “perpetual” license. License pricing was also called the
“blockbuster” model, because companies that licensed software in this way anticipated that massive sales
volume on product release would generate large profits that could sustain them until the next blockbuster
was ready for market. Large software companies had many ongoing blockbusters. Because it was
imperative for a vendor to recoup costs and secure the targeted margin at the point of sale, license pricing
was typically based on some form of a cost-plus formula. The income from license pricing was nonrecurring, and accounted for as revenue at the time of delivery.
License pricing enabled two other revenues streams: maintenance fees and service fees. Although
customers were typically not obliged to accept these additional services, those who adopted the software
for mission-critical applications often did so willingly. Maintenance fees comprised upgrades and
enhancements to the software, whereas service fees consisted of consultancy, training, support and
integration assistance. Both fees were aimed at enhancing the productivity of the software already sold to
the customer. Unlike one-time license fees, maintenance and service fees were treated as recurring revenue
by the vendor (i.e. contracts were periodically renewable, and revenues from installed user bases were
reasonably easy to predict over time).
When a software vendor received a large or growing proportion of revenue from its licensing fees,
compared with its maintenance and service fees, a strong positive demand was indicated for the company’s
core products. Conversely, a higher proportion of revenue from maintenance and service brought into
question the demand for the company’s core products and its long-term sustainability.
License pricing had several variants. Companies whose products worked well together often adopted
complementary pricing. It enabled them to price a particular software product low, often at a loss, and the
complementary product was then sold at a price high enough to both cover the loss and generate overall
profit. This approach was similar to the one used by ink jet printer manufacturers: printers were sold at a
very low price but profits were made on the sale of ink cartridges for which demand was perpetual. Price
bundling was used when the unit costs of two or more software products being sold as a package were
sufficiently low that selling them collectively at a discount was more profitable than selling each unit
separately.4
Subscription Pricing
Around the time of the economic slowdown in 2001, subscription pricing of software started gathering
momentum as an alternative to the license model. Companies sought increased accountability from
vendors and wanted to track the performance of their software installations more closely. They wanted the
flexibility to switch vendors if performance was not satisfactory. Most importantly, companies were
4
Software Pricing Trends: How Vendors Can Capitalize on the Shift to New Revenue Models, PricewaterhouseCoopers,
2007, http://www.pwc.com/techforecast/pdfs/SoftwarePricing_X.pdf, accessed December 21, 2008.
This document is authorized for use only by Nicholas Roughgarden in Pricing Strategy – BMIM1 – Charlestown at Hult International Business School, 2015.
For the exclusive use of N. Roughgarden, 2015.
Page 4
9B09E005
looking for ways to continue to invest in information technology (IT) projects without the obligation of
major upfront capital payments. These changing customer needs demanded a corresponding change in how
vendors charged for their software. Application service providers (ASPs) were among the early proponents
of the subscription model, promoting innovative approaches, such as “pay-per-use” (i.e. costs to the
customer rose or fell depending on usage) and “try-before-you-buy” (i.e. customers could test a limited
version of the software with a trial account either at no cost or at a reduced rate).
Several advantages were inherent in subscription pricing. For a software vendor, subscription pricing
meant income consistency (e.g. a vendor that sold a three-year subscription for $90,000 could record with
certainty $7,500 of income for each quarter). A lower initial price allowed the vendor to aim its pitch lower
in the target organization’s hierarchy, thus reducing the sales burden and speeding the adoption decision.
Subscription pricing could enlarge the available customer pool by making software affordable. In most
cases, this approach made maintenance fees mandatory, which led to an additional steady revenue stream.
For a customer, the subscription model allowed payments to be written off as expenses in the period they
were incurred, unlike the license fee model that required costs to be capitalized over a longer period.
Customers now paid only for the software services that were consumed and could scale their use up or
down according to internal demand. Subscription pricing also gave the customer both a carrot and a stick
in negotiating with vendors to ensure that performance remained commensurate with fees charged.
The subscription model also had disadvantages. The vendor could lose customers who now could choose
not to renew a subscription when the subscription term ended. The vendor could lose not only the
subscription revenue but other benefits as well. For example, if the contract had included an option such as
hosting a customer’s data, on termination of a contract, the vendor would have to write off hosting-related
investments, which were often considerable. Because the revenues from a subscription model were ad hoc
rather than one-time, the vendor would have fewer financial reserves to invest in research and development
(R&D) and to fund the development and launch of next-generation products.
The downside for the customer was two-fold. The customer did not own anything at the end of the
subscription period because this model did not lead to asset creation. If the vendor was hosting its
applications and the contract was not renewed, the customer faced the threat of losing access, and
potentially ownership, to its own operating data.
The lines distinguishing the licensing and subscription models could often become fuzzy. For example,
some analysts considered Microsoft’s Software Assurance Program, which offered regular product
upgrades at an annual premium, to be a subscription offering, whereas others treated it as a maintenance
offering. On a different plane, salesforce.com sold its customer relationship management (CRM)
application at a subscription fee of $6,000 per annum, whereas Microsoft sold its CRM package,
Dynamics, for a license fee of $100,000.5
Cost-based pricing, which was at the core of both licensing and subscription models, had several
limitations. Practitioners of the activity-based costing (ABC) methodology readily acknowledged that
product costs were rarely precise; and imprecise costs meant imprecise pricing decisions. Cost-based
pricing was tactical and easily matched by competitors. This approach accelerated the process of
commoditization and undermined long-term profitability. Cost-based pricing placed a vendor’s own
interests ahead of the customer’s interests (see Exhibit 4).
5
Software Pricing Trends: How Vendors Can Capitalize on the Shift to New Revenue Models, PricewaterhouseCoopers,
2007, http://www.pwc.com/techforecast/pdfs/SoftwarePricing_X.pdf, accessed December 21, 2008 pp 14-25.
This document is authorized for use only by Nicholas Roughgarden in Pricing Strategy – BMIM1 – Charlestown at Hult International Business School, 2015.
For the exclusive use of N. Roughgarden, 2015.
Page 5
9B09E005
Value-Based Pricing
Value-based pricing was a more recent development. It was based on a simple customer premise: You
provide value; we pay for it. The complexity existed in determining the value and reaching a consensus on
how to monetize the value. In value-based pricing, the price charged, the method of payment and the
timing of payment were all up for grabs. In contrast to cost-based pricing, value-based pricing charged a
price relative to the customer’s perceived value of the benefits received. The value was unique to a
company and was not easily duplicated.
One form of value-based pricing was tiered pricing. Software vendors such as IBM, Nokia and Adobe had
experimented with and, in some cases, had embraced this approach. For example, Adobe assigned different
prices to its Acrobat products for different customers on the basis of how the products were being used.
Adobe customers who created basic portable document format (PDF) documents paid less than Adobe
customers who wanted to use more sophisticated features, such as commenting and digital signatures. By
integrating the “willingness of customer to pay” into the pricing model, tiered pricing marked a major step
in the evolution of value-based pricing.
Some vendors, such as Plexus Systems, let customers set their own price using three key factors:
contribution to business value (linked to the ability of the software to generate positive cash flow for the
enterprise), market value (based on the prices of comparative commercial options available in the market)
and enterprise development cost (based on what it would cost the enterprise to develop the software inhouse vis-à-vis the cost of outsourcing it).6
Vendors could approach the question of valuing their software product from three main perspectives. First,
they could list all of the benefits they believed their offering delivered to all of their customers. Although
pricing on the basis of benefits was a departure from conventional cost-based pricing, it was still a onesided, “one-size-fits-all” approach that did not genuinely consider benefits from a given customer’s point
of view. Second, vendors could acknowledge that the customer had alternative solutions available and
highlight key points of difference in the vendor’s favor. Because this approach did not take into account
the customer’s view of those differences, it again risked value presumption due to a failure to understand
customers’ specific needs. Third, vendors could form their value proposition on the basis of the specific
elements that mattered to a particular customer segment and communicate these benefits in a way that
conveyed an understanding of the customer’s business priorities.7 This perspective required extensive
research into customer value by market segment, which was a time-consuming practice. It was this third
approach, however, that enabled pricing to become a strategic tool to build a sustainable competitive
advantage (see Exhibit 5).
COMPANY BACKGROUND
Established in 2004, Nuway was born out of a perception that traditional software development approaches
were fundamentally flawed. The founder was committed to three software development principles: getting
it right the first time, meeting fixed budget and time constraints, and ensuring zero maintenance. A typical
Nuway project lasted six months and was billed incrementally to the customer on a cost-plus basis, on
completion of individual milestones, for approximately $100,000 (see Exhibit 6). Nulogic was built into
6
Software Pricing Trends: How Vendors Can Capitalize on the Shift to New Revenue Models, PricewaterhouseCoopers,
2007, http://www.pwc.com/techforecast/pdfs/SoftwarePricing_X.pdf, accessed December 21, 2008.
7
James C. Anderson et al., “Customer Value Propositions in Business Markets,” Harvard Business Review, March 2006, pp.
91-99.
This document is authorized for use only by Nicholas Roughgarden in Pricing Strategy – BMIM1 – Charlestown at Hult International Business School, 2015.
For the exclusive use of N. Roughgarden, 2015.
Page 6
9B09E005
the total price of each project, at a cost of $6,000 (derived arbitrarily from 40 hours × $150 per hour). In
Nuway’s first full year of operations, sales revenues totaled $400,000; this figure had doubled in each
subsequent year through 2008. Nuway’s gross margin was 25 per cent.
Founder Thomas Henshell was a self-taught programmer. As a young student he was more fascinated with
building software than with building Lego. Henshell worked for several years in retail computer sales. In
1999, he was hired by grocerygateway.com, an up-and-coming online grocery store, to be its lead software
developer. In 2004, Henshell chanced upon an exciting business opportunity to develop a mobile software
application to support “on-the-go productivity.” He chose to leave grocerygateway.com and start his own
company, Nuway Software.
Said Henshell:
Developing mobile software is the most challenging of all forms of software development.
It separates the men from the boys. The main challenges are that, compared with desktop
computers, the processors are slow and the devices run with little memory. To have
excellent performance you must write good code. Without it, the mobile device is useless.
Writing good code is the biggest challenge for any programmer and it is particularly so in
mobile software
Similar to the adoption of the Internet during the mid-1990s, mobile technologies in the new millennium
were enabling a ubiquitous communication channel globally, and gaining increasing importance for
marketing products, delivering services and collecting payments at the customer’s doorstep. Numerous
hardware manufacturers and software vendors were seizing the new opportunity by forming mutually
beneficial partnerships. Mobile was the new frontier. As The Economist noted in a review of corporate IT:
The mainframe, the original computing platform, was dethroned by minicomputers, which
in turn gave way to personal computers, which are now being pushed aside by hand-held
devices and Smartphones. Computing power will become more and more disembodied
and will be consumed where and when it is needed.8
A mobile software vendor could thus position itself as a solutions provider at the cutting edge of
technology, working to improve the productivity of its clients’ operations and transforming the way they
ran their businesses (see Exhibit 7). The software was a source of competitive advantage to Nuway’s
customers because it served their mission-critical needs.
Nuway had partnerships with two U.S. hardware manufacturers, Motorola and Intermec. Both were early
adopters and between them had quickly attracted and maintained 80 per cent of the mobile hardware
market worldwide. Motorola and Intermec were manufacturing rugged hand-held devices for which
Nuway was building software suited to specific customer requirements. The hardware makers were
marketing their devices to companies operating broadly in the transportation and logistics industry. Nuway
would work closely with the clients of Motorola and Intermec to develop custom software for each client,
individually, on a project basis.
Manufacturers of hand-held devices often freed themselves to focus exclusively on the large-scale
hardware market, by engaging in non-exclusive partnerships with software development firms. Custom
software development houses, on the other hand, required close interaction with a single customer for as
8
The Economist, October 25, 2008 “A Special Report on Corporate IT” pp1-15 www.economist.com /special reports.
This document is authorized for use only by Nicholas Roughgarden in Pricing Strategy – BMIM1 – Charlestown at Hult International Business School, 2015.
For the exclusive use of N. Roughgarden, 2015.
Page 7
9B09E005
long as six months in order to form a sufficiently deep and clear understanding of the customer’s unique
internal business processes, and so deliver an information system that truly helped the customer. Custom
software development processes simply could not be rushed or pushed beyond a certain point.
For each project, Nuway invoiced its customers directly for a pre-negotiated, fixed sum. Selling the
resulting software to multiple customers was not an option because each software package was specific to
the needs of each particular customer situation.
Globally, mobile software was a fragmented business comprising thousands of enterprises and no major
players. Nuway had approximately two dozen competitors located in Ontario alone. In 2008, Nuway had
nine customers and aimed to increase that number to 15 in 2009. In addition to the transportation and
logistics industry in which it already operated, the company had identified five industry sectors as potential
future customer pools: retailing and hospitality, manufacturing, field mobility, energy and utilities,
government and healthcare. The company had a staff of six — all programmers — who, in contrast to
programmers’ average tenure of 18 months at most companies, talked of “staying at Nuway until
retirement.” Their loyalty was due largely to the corporate culture (see Exhibit 8) and incentives such as
profit-sharing.
NULOGIC
Early in 2008, Nuway developers assembled a “toolkit” of approximately 20 common mobile software
functions that were frequently reused from one project to the…
