Digest Topics

Authoring & editing
Budgeting, reporting, & metrics
Case studies
Content management
Enterprise knowledge management
Government applications
Graphics & posters
Learning & Education
Legal applications
Ontologies & topic maps
Personal knowledge management
Point of view articles
Publishing
Reviews
Studies & surveys
Research & searching
Slides & presentations
Taxonomies & metadata
Technologies & tools
Trends & strategies
Usability & user behavior
see also A - Z index

Two perspectives on ROI

Articles in focus:

The business case justification for simulation software. (iRise Corporation white paper, September 2004). This link takes you to a registration page, but the paper itself is free.

"Evaluating the costs and benefits of end-user development" in Proceedings of the First Workshop on End-user Software Development (2005) After you download this PDF file, use the Adobe Acrobat FIND command to search for "costs." You'll find this paper on page 43 of the document.

See also: "Get ready for end user development" (Montague Institute Review, January, 2006)

About the authors. The author of the business case is an anonymous marketing person at iRise Corporation. The author of the workshop paper is Alistair Sutcliffe, a Professor of Systems Engineering and Director of the Centre for Human Computer Interface Design at Manchester University (UK).

Articles summarized. The iRise white paper is designed to help CIOs and CFOs justify the $250,000+ cost of the vendor's software package. The product allows business analysts to create working models of applications that look and behave as they would on the Web. The goal is to minimize the delays and cost overruns due to lack of user involvement in the software development process. The remainder of the white paper consists of six "scenarios" that illustrate projected savings and revenue enhancement: 1) the cost of rework, 2) managing outsourced development, 3) cutting time to market, 4) cutting prototyping costs, 5) enhancing competitive advantage, and 6) increasing user adoption rates.

The workshop paper is targeted to an audience of computer scientists who specialize in "human-computer interaction" (HCI). HCI is a recognized interdisciplinary field concerned with how people behave when they are using computer hardware or software. The first part of the paper describes a metrics approach that compares perceived software costs and benefits prior to acquisition with actual costs and benefits after implementation — complete with equations and graphs.

In the second half of the paper, the author uses his metrics framework to analyze the perceived and actual costs of a real project: a university content management system (CMS). It's clear that he's using his own experience at the university's School of Informatics, one of the 10% - 15% of groups on campus that he classifies as "advanced." Average users can create simple Web sites using Microsoft Front page, while advanced users can create dynamic sites using features such as databases, multimedia, and animation. According to the author, the university's CMS had more costs than benefits for advanced users. Its scripting language was difficult to learn. It was hard to integrate existing JAVAscript and database applications. There were technical problems. Not surprisingly, there was widespread resistance to using the new system, and it was not clear that the university would continue to support it.

My comments. Both documents deal with some of the same issues — rework costs, user adoption rates, and the flexibility to innovate — but the first author sees them from the top down while the second author sees them from the bottom up.

To the author of the white paper, I would address the following questions:

• Who is the "business analyst" for whom this software is intended? Is there more than one kind of analyst (e.g. average, advanced, multi-functional)?

• How much training will each kind of analyst require to become expert at using the software?

• What kind of prototyping are the analysts already doing? How will they react to a new prototyping tool? Could the results be just as good if they were given more training and support for their existing tools?

• How hard is it to turn the prototype into an application? Will there be data quality and integration problems? Can the company's metadata architecture effectively support new applications?

• Who will turn the prototype into a production-quality application? Will business units be free to acquire implementation services from any source they choose, or will they be required to use the in-house IT function? How will their choice of implementer affect project schedules and costs?

I would ask the author of the workshop paper:

• Would you agree that it's desirable to have a consistent "look and feel" to university Web pages (a primary selling point of a CMS system)? If so, what do you think is the best way to achieve it?

• Would you agree that it's important to provide a secure university network that resists viruses and malicious attacks and protects sensitive data? If so, what is the best way to achieve it without stifling local creativity and initiative?

• What is the best way to reduce costs for the 85% - 90% of university Web site authors and developers who aren't advanced users? Would you advise them to invest in learning the tools used by the School of Informatics (i.e. JAVAscript and the Microsoft .net programming environment)? Can the fruits of your development work be shared with them?

• How would you define "usability" and Web site "quality," since a university Web site may have multiple audiences: e.g. prospective students, undergraduates, graduate students, faculty, administrators, and the community at large? How important are advanced features such as multimedia and animation to these audiences?

What we need is a meeting of the minds, since both the top-down and the bottom-up perspectives have merit. But what does this mean in practical terms?

I got some clues from one of our members in the financial services industry, whose company is formalizing a program to support end-user development. Three years ago, the firm established the role of "technical end user developer" within the IT function. Today this person is a manager supervising four employees.

At first, the IT manager resisted the move because the programmers looked down their noses at Microsoft Excel, Visual Basic, and Access — the most common end-user development tools. Today, they've substituted the challenge of working with cutting edge technologies with the instant gratification that comes from helping end users get quick business results. It's a good thing too, since many of their applications are mission critical and need to be migrated to IT for stability and reliability.

The technical staff advise and guide end users, write code, and help with IT migration. The company also offers a training program for business analysts in Visual Basic, SQL (Structured Query Language), and "connectors" such as APIs (Application Program Interfaces) and plugins.

Bottom line
From the enterprise perspective, commercial prototyping and content management software may indeed reduce rework costs, minimize time delays, promote a consistent "look and feel" for Web publishing, and save time for the average user. From the end user perspective, especially the advanced user who's already made an investment in learning how to use desktop prototyping tools, enterprise software can mean an additional learning curve, time invested in rework and integration, and loss of flexibility — not to mention frustration and alienation.

To get the best of both worlds, organizations need to develop a support system that includes end user training, technical support, standards, and policies. The benefits are three-fold: leveraging the development work already done, reducing the cost of future developments, and safeguarding mission critical applications so they are always available when needed.

For original articles by Jean Graef, see the Montague Institute Review.