Why We’re Moving to Pen-Based Computing

Or what’s wrong with existing systems

by Sharon R. Masters

Editor’s Note: The following is a first-person account of why one major corporation is rapidly moving to pen-based computing. Because this project is still in the evaluation stage, the company is left unnamed.

We are a large consumer products manufacturer with a widespread direct sales force. For a number of reasons, both the sales and support staff have come to the realization that our seven-year-old computer system (used for tracking sales force call data) is no longer adequately serving our business needs, nor does it provide the technological gateways to be useful in the future.

This current system is an enterprise-wide application that tracks product distribution, retail territory coverage, and selling/merchandising activities. It’s actually comprised of three separate but integrated components:

• Handheld computers used by the sales reps to transmit information to the data collection PC’s located in the data center, which in turn transmit to the mainframe. Sales reps go from store to store, gathering data and keying into a unit the size of a small desk calculator. Once the information has been input, it cannot be changed, and once transmitted, it is not accessible from the handheld unit. The units have only that one data entry function, so the sales rep has no other computing power in his/her hands.
• PC’s in field offices used to maintain the information relating to the geographical territories. Databases are updated by clerks from information provided by the sales reps. Additions, deletions, and changes are transmitted weekly to update the mainframe database. Programs on the PC (“inquiries”) can run against certain PC’s databases or submit batch remote job entry (RJE) against the mainframe database to be printed at the PC.
• The mainframe is host to the central database, which contains the data from all field office PCs.

Reports are generated in the weekly and period-end processing runs (13 four-week periods in a year). After weekly processing, data relating to the handheld computer input is downloaded to the field office PCs.

The problems with this system are:

• Practically every piece of the system (hardware and software) comes from a different vendor. At last count, there were 15 stops on the responsibility trail.
• Having two sources for the same information is not reliable (PC databases and mainframe databases are supposed to reflect the same information) and out-of-sync problems occur for any of a number of reasons.
• In some cases, source code is not complete or current; in others, there is no source code nor is the original vendor (if still in business) willing or able to provide it.
• The corporate architecture is changing and none of the existing pieces fits the new guidelines;
• Business needs have changed beyond the scope of the original products.
• The hardware is old and failure rates are increasing, some parts have been replaced and others cannot be replaced without significant expense.

Consequently, support staff and field-office personnel spend far too much time putting out fires (some of us have even taken to wearing red suspenders and hard hats). We are too often catching up with lost data, fixing out-of-sync databases, applying band-aids to programs, and finding work-arounds to program bugs and quirks. All of this led support staff and sales personnel to the conclusion that there had to be a better way and we began our search.

When surveying the systems offered by one leading computer vendor, we were impressed by the desktop and laptop systems — but what caught the sales staff’s eyes were pen-based systems. After examination, it was evident that the hardware would fill our needs, but there were questions about the software. However, the vendor knew of and had worked closely with a software development firm that has an application in production which performs much of the functionality of our existing system.

We are currently compiling the presentation document needed to convince sales management that there is a better way, and that pen-based computing could be a significant player in this solution. In addition to fulfilling the data collection and reporting needs, pen-based computers would allow the sales reps to have additional computing capability and power, with resulting increased productivity and account management.

Pen-based computing, in combination with developers who know how to take full advantage of the pen-based architecture, gives us hope that future systems will not leave us limping along with outdated technology in another seven years. It will be possible for us to assume the tasks of writing code to update programs or of relying on the vendor if need be. We are pushing for this project to have 1992 dollars. Maybe then we can increase the computing power of all involved personnel and provide the tools to manage the business.