In part 1 of this blog, we looked at how H-P and Tektronix were structured and why that worked well initially for them. Eventually, though, things changed, and not for the better.
Toward a new way
In a hierarchy, all essential company functions — design, manufacture, sales, and support — are part of the company. Those at the top must be involved in all these functions to some extent. It worked out in the case of H-P; Bill Hewlett tended toward engineering leadership, while David Packard had a better head for business. At Tektronix, founder Howard Vollum was somewhat more like Hewlett, and Jack Murdock was somewhat more like Packard. What additional technical advancements might these founders have achieved had they not been burdened with nontechnical, organizational issues?
At Tek, some functions that might have been farmed out to contract manufacturers were brought into the company because nobody else could do them well enough. Tek had operations in ceramics (hybrid circuits), CRTs, a big sheet-metal building, and (later) integrated circuits. In time, these operations (not CRT manufacturing, the component at the hub of the analog oscilloscope) were able to be outsourced. Even the high-speed IC operation was sold off in a joint venture with Maxim.
At H-P, the split forming Agilent was similar in effect. The large corporation had become unwieldy in a hierarchical organization, and a reduction into differentiated functions was required to manage the complexity and regain focus (i.e., to establish clear goals). The lesson to be learned is that there is not really any need for large organizations — only large capitalization for the few processes requiring it. The hierarchy persists because no viable alternative has displaced it.
Contract manufacturing is sustained by the low labor rates of emerging countries. If low-labor-rate contract manufacturers can manufacture to the requirements of designers, then what is a company like Tek or H-P really contributing? Is Tek's real contribution to the world simply oscilloscope products? The apparent answer has been yes. The correct answer is no. This is shown by an event in Tek history when, by the US government's action, companies were able to copy Tek scopes and produce lower-priced models.
Tek was not bringing the manufacturing of scopes to the world as its essential contribution — its distinctive competency . What was it instead? Innovation. What was unique at H-P and Tek was in the design function and its enhancement through employees at all levels. Proprietary component manufacturing has also diminished greatly or become distinct functions. Contract manufacturing for all but specialized, high-performance electronics is becoming a commodity, especially as the technical competency of emerging countries quickly increases.
For a time, both H-P and Tektronix sold products through Neely, a contract sales company. Sales were not a distinguishing function of either company. Nowadays, with the Internet and word-of-mouse advertising, the sales process is also changing. So is customer support, which often is relegated by machine mindedness to impersonal, irrelevant, and frustrating Internet boilerplate responses to customer inquiries.
Product maintenance is now provided by companies specializing in the calibration of instruments. The characteristic deficiency of the support function indicates an implied hostility toward customers. That attitude also presages a change in the way designers relate to users.
In part 3, we will look at where this new paradigm is taking us. Are we still building quality products? And what are we paying for such products (in the short term and the long term)?
- Is There a New Way Ahead for Electronics Enterprise? Part 1
- An Instrument on a Chip? A Look Back
- Z Meter on a Chip? Impedance Meter Integration and Readout
- Getting From Scopes to Semiconductor Innovations
- Can We Put Other Instruments on a Chip? Part 2
- ‘Scope on a Chip? Why Not a DAS Onboard?
- Put an Oscilloscope on a Chip: Why Not?