Buying Products & Services

RULE: The overhead cost of purchasing a computer product can be greater than the cost of the product itself. Here are elements that figure into the cost of a computer component:

• Research what to buy ($/hr)
• Find where it can be bought at what price ($/hr)
• Cost to purchase ($/hr)
• Shipping (or cost to pick up (much higher)) ($)
• Cost of component ($)
• Cost to configure the component into operating system software ($/hr)
• Cost to verify correct function and usability ($/hr)
• Cost to diagnose failure and obtain replacement (%$/hr)

If you work with an integrator who sells equipment, his/her prices will be significantly higher than you may see advertised. If you are a business person, you will understand why - the integrator is absorbing a considerable amount of the overhead cost and risk, more efficiently than you could, and is spreading it among multiple clients - your true cost is less, probably a lot less.

If you are a hobbyist, you are not valuing your time, so the cost from the integrator is higher. Of course, if your purchase creates problems you cannot handle yourself, the integrator will send you a "guilt free" invoice for the entire cost of handling the problem. Many integrators will actually encourage you to buy stuff yourself because they love those "guilt free" invoices so much.

If you work with a consultant or integrator, you will find they recommend only a narrow range of products, probably not including the great new products you are reading about in the magazines. They recommend what they know works together and what they know enough about to support, and are spreading the cost of experience over multiple clients. If you insist on the latest wiz-bang product, your true costs are likely to be very much higher.

Reports from shoppers at the big computer warehouse stores tell that a very high proportion of the merchandise on the shelves has already been returned by a previous purchaser. Every one of these returned items represents a lot of time wasted by the previous purchaser. Sometimes the product is actually defective, and may have been purchased and returned more than once. It is clear a business cannot afford this.

Mailorder has similar problems, but returns are more difficult and take more time. Fortunately, the average mailorder buyer is a bit more knowledgeable than the average warehouse buyer and understands the risks they are taking. Mailorder components at sharply lower than normal prices are probably at least one revision out of date, and may not work with newer products. This is usually stated for software, but almost never stated for hardware.

RULE: Buy what your technology partner supports. Original purchase may be higher, but your true cost will be much less. (Index)

Quality - The Industry's Dirty Little Secret

Parts are tested and sorted by performance and quality at the end of the production line. The best go to prime customers, and the marginal ones go to the "lowest cost" market. Cost pressures are so great, every part that actually functions must be sold.

Once the product leaves the manufacturer there is no way to determine its quality status - all are marked exactly the same, and only the manufacturer has the test rigs to determine quality. The only way to assure you are getting "primes" rather than "culls" is to know who handled the product on its way to you.

This little known process goes a long way to explain the high failure rate of "best name brand" components purchased from "lowest price" vendors. (Index)

Benchmarks and Comparative Reviews - Meaningful?

A product tuned for benchmarks may be cranky and may work properly under only a narrow band of conditions. "Buying by benchmarks" puts you at risk of purchasing equipment that does not work well (or at all) in your circumstances.

Comparative Reviews: Here's how comparative reviews are handled by magazines.

• List all the features of the market leader (generally the product with the biggest advertising budget).
• Sharply criticize alternative products for any feature they do not include, whether or not the feature is useful, even if the feature doesn't work well in the market leader.
• Ignore any feature an alternative product has that is not included in market leader, no matter how useful that feature may be.
• Sharply criticize any feature that does not work exactly the same as a feature in the market leader, even if the alternative works better. "Users will be confused."
• Completely ignore practicality, usability and performance. These are not "quantitative measurements".
• If all else fails, add the line "[Leading Product] will include all these features in its next revision, so we recommend a wait and see attitude".
• Publish the least offensive of the many angry letters received from users of alternative products and explain, "we can't compare features not included in the market leading product".

This is not an exaggeration. In the magazine business, the "right" product has to win, but it has to appear fair (kind of like Professional Wrestling on TV). Advertising revenue is at stake!

I remember when DR-DOS came out, including a real (and usable!) full screen system editor. Reviewers panned it for not having Microsoft's EDLIN, unquestionably the worst system editor ever included in a shipping product. Of course, the next version of MS-DOS had a full screen editor too, though one that was not as good.

Are Comparative Reviews Important? Yes. Users of PC software are constantly complaining that products are too complex and too difficult to use. Products that actually are easier to use and less complex fail quickly in the marketplace because they fall short in the feature lists of comparative reviews. People buy the product that got the best reviews, not the one that is most usable.
(Index)