Thursday, January 07, 2010

It took a while for end-user-driven computing to take hold: remembering the "salesmanship" lessons of the 1970s and 1980s

Back in the early 1970s, companies already understood the idea of user-driven computing. At NBC, one of my first tasks was to write an RCA Spectra COBOL program to manipulate a “parameter file” what would eventually set up the options for processing general ledgers on a number of NBC’s owned stations and companies (from a package called Infonational, that ran on Univac 1110 “Ascii COBOL” (36 bit words) in a day that this sort of mainframe computing was thought to have a future, before IBM drove everybody out.

The Univac mainframe system also had a “report writer parameter file” that users coded on sheets and employees keypunched. But even then I.T. was ready for the concept that individually owned subsidiaries of a corporation could continue to operate differently. The economy of scale came from combining processing in a data center, not from making the companies process the same way.

But when I went to CABCO (Combined A&B Medicare Consortium) in Dallas, of six Blue Cross and Blue Shield plans, I ran into a similar issue. I worked on the back-end systems, especially surveillance and utilization review. So I designed a subsystem that would express company-driven parameters, much the way NBC had, as I was familiar with that way to do business. But I was never able to “sell” the idea. Users complained they were confused by all the “options” and did not seem to understand that the point of a combined system is processing efficiency; it is not intended to tell constituent companies how to run their own businesses. This became a deeply political issue (around 1981).

Things have changed a lot then. Now, when companies combine they keep separate systems and replicate to a common mid-tier, or use direct connect mechanisms to provide customer service center end user GUI interfaces.

But was my little “downfall” in 1981 a failure in “salesmanship”? I was on to an approach that had not quite caught on yet, but that would become the business norm by the mid 1980s, as Wall Street started forcing companies into mergers.

When one goes to sell a vision, even to internal customers, one should first determine how widely held that vision already is, in other organizations.

No comments: