BE Radio Jan/Feb 1995 Managing Technology Buying new technology By William Fawcett As in engineering, management decisions are best made with quantitative analysis. This is especially true in purchasing, where an overwhelming array of selections can cause the fiscal gridlock that some have called paralysis by analysis. If we remain subjective, we are often vague and uncertain, while quantitatively we can proceed with confidence. Some preparation is mandatory: Get a good staff/management consensus on just what functions in the workplace are in need of improvement. Ask for compelling reasons why these areas are deficient. Some good examples: Is money being wasted on low productivity? Is the station being put at a competitive disadvantage? Are there untapped sources of revenue out there? Is staff morale being sapped? Is the on-air product suffering? And so on. The outcome of this study should provide you with a reasonable list of needs, which should be prioritized and fashioned into a reasonable master plan, perhaps extending over several years. Next, technical solutions should be sought, in general terms, for the first problem that you will tackle. Consult subject-matter experts in that area -- either on staff or outside -- to determine who the key manufacturers and developers are for these technologies. Then you can get down to the real work of analyzing the details of these players' offerings. An empirical approach Before starting your analysis, first obtain some basic information on the equipment. A few phone calls to some friendly vendors will provide specifications (cut sheets) and pricing information. Thanks to the fax machine, what once took days or weeks can now be done in hours. Take that information and determine which items are minimally qualified to do the work. Once you have narrowed the field, you must make a purchasing decision. This is where things often stall, especially in the realm of the subjective. To get through these doldrums, let the numbers do the talking. If you don't have a spreadsheet program, get one. It is nearly impossible to make any of these kinds of decisions without one. An initial exercise will determine the selection criteria for your candidates. Here are some sample categories: ease of operation, audio quality, compatibility, durability, serviceability, factory support, distance to factory, use of off-the-shelf components, efficiency (transmitters), physical size. Add your own as appropriate. Be sure to include criteria that may be unique to your specific application. Set up a spreadsheet listing the selected criteria as columns heads. The candidates are then placed in rows. Rate each candidate on a scale of 1 to 5 in each category. Then make a column that will average all scores, perhaps giving added weight to more critical criteria. This simple process allows to you quantify what you probably already know subjectively. Just going through the exercise will help you decide which criteria are most significant, however. Keep in mind that the best choice may not be the product with the best score at this point. Cost analysis A second exercise considers the cost of the equipment. It is best to compare on a cost-per-year basis. Such annualizing allows direct comparison between candidates with differing life expectancies or operating costs. Once again, it's back to the spreadsheet -- either a new one or an extension of the first one. On this spreadsheet, columns can include the initial cost of the unit, freight charges, estimated repairs during the life of the unit, estimated parts expense, installation cost, power consumption (transmitters) and expected length of useful life. Base the estimated parts and repairs costs on your own experience or the recommendations of other experienced users. Add up the cost columns and divide by the life and you get the annualized cost. Again, each application will have unique criteria. The life of the unit need not be exact -- you are just trying to make some relative comparisons. In general, you can use 7 to 10 years for most professional equipment, 3 to 5 years for consumer products and 20 years for towers. Make your own guesses, but be consistent among the candidates so that the comparisons are valid. This exercise may reveal hidden costs of operation that can make a more expensive item the best buy. No attempt is made to address cash-flow games like buy-now pay-later schemes -- such reasoning is short sighted and can ultimately lead to business failure. Tables 1 and 2 present a few examples based on actual, recent purchasing decisions. In Table 1, while the tube transmitter has the lowest entry- price, it shows a higher annualized price because of its lower efficiency and need for periodic tube replacements. At one time, solid-state transmitters were seen as less reliable, but now this technology has matured and that is no longer the case. In fact, I cannot recall having to replace an output device in a single transmitter of current design (whereas I have replaced many in ca. 1985 and earlier designs). -------------------------------------------------------------------- Table 1. A sample analysis for purchase of a 1KW FM transmitter. All three units are assumed to have equal (10-year) lifespans. All 3 units have -----ten year-------- 10 year (or equal) Tube Power Life Price cost consum repairs annual Solid State Name Brand $23000 $0 $8760 $1200 $3296 Solid State Off Brand $17000 $0 $8760 $2400 $2816 Tube type- economy $12000 $3350 $11388 $3600 $3033 -------------------------------------------------------------------- Table 2 considers the commonly asked question of whether to buy professional or consumer CD players for a manually operated on-air control room. While the argument can be made that the expensive unit has better specifications, I can't hear the difference -- can you? Perhaps the pro unit has a jog-wheel or some other neat features, but these may not be needed in your on-air applications. If the consumer unit meets your minimum criteria (Exercise 1 above), then you should select it, based on the results of Exercise 2. If you need the additional features and can afford the pro-model, you should go with it. But remember, if you buy the pro model and technology advances as fast as it has in recent years, you might see future consumer units outpace the performance and features of your expensive pro model during its longer lifespan that you are locked into. (Today's consumer CD players certainly outperform the pro units of a few years ago.) -------------------------------------------------------------------- Table 2: Purchasing CD players: consumer or professional? (*-IHF/PRO interface has 6-year life, so it contributes $25 to annualized cost.) Street Hi-Z Add Remote Life Price Interface Start Repairs Annual Top of the line T_______ model 6 $1700 $0 $0 $400 $350 El-cheapo T_______ model 1 $140 $150 * $20/unit $0 $185 *Hi-Z interface has a 6 year life ($25/yr.) -------------------------------------------------------------------- Some assembly required This brings up a broader, related issue. In making annualized cost comparisons, be sure to factor in the price of using "cutting- edge" technology. Include the amount of extra time that may be required to de-bug the system. Include shipping the unit back to the manufacturer three times. Include loss of use of the equipment while you are still trying to get it to work. And don't forget that the equipment may become incompatible with the rest of your equipment if it turns out to be an intermediate step, which will effectively shorten its useful life. In considering whether to include cutting- edge technology in the candidate pool, consider the following rules. (I have always said, "Given an unlimited budget, I would spend it," but most of us are forced to live in the real world.) Rule #1: Stay away from the extreme cutting-edge. Recent problems with the Pentium processor (involving high-precision math) have proven this true. Do you really want to beta-test products for the manufacturer at your expense? Cutting edge technology may be unreliable, unproven and uneconomical. Beware of uncompetitive, "sole-source" technology. Cutting edge technology is always expensive: A scientific pocket calculator that cost $500 in 1975 now costs $50. New 486 computers sell for less than the IBM-XT (8088) that sits on my desk doing transmitter logging duty. Rule #2: If possible, let the technology mature. Will today's cutting edge be tomorrow's standard, or is it an intermediate step that will rapidly become obsolete? For example: diskette-based digital "cart" machines. While certainly a step up from analog tape-loops, they still require physically moving, stacking, sorting (and dropping). I suspect they will be surpassed by hard-disk based systems, which are falling rapidly in price. Rule #3: Look for products that utilize consumer (mass-produced) technology. Early CD decks cost ten times what a decent consumer CD player now sells for. Some early versions of hard-disk audio storage devices were built on proprietary main-frames. Today, most utilize inexpensive PC mainframes, and take advantage of the falling prices of high-capacity drives due to consumer demand. The same is true of broadcast automation systems and many digital audio workstations. Significant economy of scale applies when development and tooling costs are spread over such a large quantity of product. The move from proprietary components to off-the-shelf components is often a sign that the technology has matured. Certain devices, such as transmitters, will never be mass-produced, but many transmitters use off-the-shelf power supplies and RF modules from third-party vendors. The use of common, non-esoteric technology ensures that the equipment, parts and knowledge will be available when repairs are needed. Remember that well-executed standard technology will always beat poorly-executed new technology. Evaluate products for forward and backward compatibility. By staying one step behind the cutting edge, you gain great cost benefit, and you also can better assess a product's likelihood of becoming the new standard or just an intermediate step that will be soon discarded. By now you're probably thinking that I have no sense of adventure or curiosity concerning new technology. Consider this -- adventure doesn't pay the bills. Just ask anyone who has been burned by a new system that failed to meet its manufacturer's claims. In any case, as long as you realistically factor in the hidden potential costs of adopting very new technology, you can then proceed to make a fully informed decision. Deriving a single-figure rating After the first step of criteria selection and scoring, and the second step of determining annualized cost, you may determine that several units meet your minimum expectations. In this case, you can select the unit with the lowest annualized cost from those units making the first cut. Of course, some cases may involve some highly critical, specific criteria, so you may want to reanalyze the remaining candidates by running them through the first exercise again, this time with higher minimum expectations and/or different weightings. Then select the unit with the lowest annualized cost from the qualified candidates once again. If the front runner isn't obvious by this time, a third exercise may prove helpful. Once more exclude candidates that don't meet the minimum score in the first exercise, and then compute the ratio of criteria score to annualized cost (i.e., score divided by cost). Higher numbers are better. Now you can place the candidates in rank order. Select the highest scoring unit that is within your budget constraints. This should assure you that you're getting the most for your money. Maybe you're thinking that this system does not take into account other factors (like the salesman is my sister-in-law's cousin's ex-husband). Actually, any factor that enters into the equation can be literally entered into the equation. If you find yourself thinking, "Yeah, but...," go back and modify your selected criteria or weighting factors. This is where a spreadsheet helps -- you can fiddle with the system and instantly see the results. Not all purchases will require all three of these steps. Sometimes, after the first one or two steps, the answer will be clear. Often, your original gut feelings are confirmed, but it's important to keep an open mind, and not tweak the empirical analysis just to make it turn out the way you want it to. It is easy to become overwhelmed by a vast array of equipment choices and slick salesmen. By quantifying the subjective, you make use of an important tool. Making decisions "by the numbers" gives credibility to your selections, and gives you confidence that you are delivering good bottom-line performance. (It also gives you something to fall back upon -- and rebuild from -- if unforeseen problems should subsequently turn your decision into a disaster.) It's still a crap shoot, but at least the dice are loaded in your favor. copyright 1995 Intertec Publishing Corporation