A CUSTOMER WALKS INTO A RETAIL musical instrument store with a guitar and an amplifier, let's say. He proposes to sell the guitar outright, and to trade in the amplifier on more expensive equipment, provided the store's willing to knock off 30% from list. He'll retire the present layaway with cash from the sale and start to pay for the new stuff by applying the trade-in plus a gift certificate as down payment. Most of the remainder will be financed via a commercial lender, with the balance bridged by a credit card if necessary. But hold on: the components aren't in stock. They have to be ordered and transferred from the inventory of another store. The customers asks to be phoned -- after six o'clock -- as soon as they arrive.

At conventional retail enterprises, sequences as convoluted as that take place mostly in the nightmares of clerks. But variations on the script play out in real life so frequently at Daddy's Junky Music Stores Inc., a chain of six outlets headquartered in Salem, N.H., that its sales manager uses the scenario as basic training for counter help. It is also used to make sure the fancy computer system Daddy's installed company-wide last summer can hack the job.

Projecting a mere $3 million in sales back in 1983 -- the year the search for an appropriate data-processing system began -- Daddy's was an unlikely candidate for the sophisticated setup it finally bought. Then again, its founder, president, and CEO Fred Bramante Jr. was an unlikely candidate to run a fast-growing business in the first place. An eighth-grade science teacher -- whose small son inadvertently coined the name Daddy's Junky Music Store in 1972 while commenting on Bramante's first makeshift outlet -- he sensed that buying as cheaply as possible and selling at maximum profit demanded more attention than retailers were accustomed to giving it. Bramante was willing to assume simply from his ability to open the doors that business had been OK the day before. As the business appeared to flourish, however, not knowing for sure began to drive him crazy.

A self-confessed worrier and seat-of-the-pants merchant, Bramante fretted that a portion of the receipts that salespeople traditionally pump through registers as income actually isn't. Between the gross and the net undoubtedly stood undocumented costs from inefficient inventory control, inept financial planning, unrewarding product mixes, stock shrinkage, and other unseen chimeras. For example, accessories such as straps and picks were falling through the cracks because they were so hard to track financially, even though they added up to a full 25% of sales. So was a lot of warranty repair money that, if only Bramante had had proper records, would have been refunded by vendors. He wanted a picture of where the dollar went -- sales trends, growth slopes, inventory turns, sales-incentive payoffs. He needed to see profit, to touch it, to be able to trust it. And -- if all went well -- to spend it.

To be sure, all had been going to well that last year Bramante, now 40, was putting the finishing touches on a guitar-shaped swimming pool next to the mansion he had just built on the coast. So why the fancy computer? Because, he confesses, he was running scared. At under $5 million, Daddy's profile was low enough; but at over $5 million -- which the chain would be shortly -- the operation would alert competitors' jealous attention. What I've done this far has been relatively simple, Bramante figured, so there are probably people out there who can do it as well. But now I want to accomplish things that will make them wonder, How in hell is this guy doing what we can't do?

"I'd love to be invincible," he admits, "but I don't believe anyone can get to that point. So I'll take getting as close as a business can." And tight financial reins would carry him most of the way. One tack was to hire more support staff -- 25 bookkeepers, he estimated, for starters. In bookkeepers, he estimated, for starters. In Bramante's fevered image of some pencil-driven work force, each keeper would be assigned a specialty: Number One, your job is guitar strings. On which strings do I get my best gross margin return on inventory? What quantities of strings, therefore, should I have in each store? The second person would do picks, the third power cords, and so on.

The record keeping would entail phone calls back and forth among the stores, and reviewing sales slips, and toting up inventory levels. "In most small to medium-size retail businesses, the guys who run them have to be good guessers," Bramante says, including himself among the more gifted of the seers. They operate from paper-strewn offices and out of warehouses piled to the ceiling with goods still in boxes. "If you ask for something, the owner says, 'Yeah, I've got one, it must be buried over there somewhere." To them, an operational support staff is out of the question: too expensive. It's more important to get that customer's dollar locked up in the cash register. "Those guys live for today; they don't want to make sacrifices in the area of sales. They're praying," Bramante observes, "and it scares me. What does lack of control ultimately cost? You can go for years without knowing, until one day somebody says, 'Sorry, it's too late now."

The better alternative, he knew in his untechnical heart, would be to harness the right computer. A powerful system not only would save on office space, but would wrest profitability from the whims of chance and put it where it belonged -- into the plans and strategies of his management team.

One problem: Daddy's did not have the right computer. The other: Daddy's didn't have a management team, either.

Taking first things first, in 1983 Bramante employed an ex-magazine publisher as marketing director, then called on a public accountant to head finances. "I may be the only music store in the country to hire a CPA to run the accounting department," the president rationalized, "but I'm not hiring for where I am, I'm hiring for where I'll be."

Like any red-blooded entrepreneur, Bramante hoped to be in nine figures within 10 years. But Daddy's new marketer, David Wright, now 35, discovered that orderly growth was not what Daddy's store managers were interested in talking about. Rather, they insisted on orderly orders: we never get the product we need, they complained. We're sent dozens of something else we can't get rid of, and we don't know where most of the stuff is, anyway.

"The president was saying this is the way it works, and the stores were telling me this is how it goes. I put them side-by-side, and there was no match," Wright remembers. "We had grown so fast that things were never proceduralized. I looked at the projections in his five-year business plan and said, 'Oh my God, how are we going to keep up with this?' There was a sinking sensation that we had no idea what we were doing."

Shortly after that sensation hit bottom, Wright was joined by the other new officer, controller Addison Minott, now 29. The two agreed with Bramante that Daddy's had to be missing out on a substantial slice of the musical instrument pie. But it was also clear that until the stores got what they needed when they needed it, the slice would go begging.

At that time, Daddy's had a makeshift computer that was creakily keeping the ledger and tracking only half of the inventory, mostly big-ticket items. "If a sale was made on Friday, accounting got the paperwork on Monday, and put it into the machine by that Friday. Maybe we got the printouts by the end of the next month," complains Wright, who, like Minott, had arrived with a fair measure of computer literacy. "Everything was between a week to three months behind; it wasn't interfaced day to day."

Nor was finding speedy solutions helped by the fact that Daddy's procedures, scattered among a number of small outlets, are about as complicated as a small retailing operation can get. A straightforward stores-keeping unit, or SKU, approach was used to stock 10,000 or so new items. (In a SKU -- pronounced "skew" -- system, each product that differs even slightly from another gets its own number.) But after that, things took on a life of their own. With one of the largest pools of used instruments in the world (Daddy's claims), around 3,000 more items are pumped through inventory at a given time, but not in quite so straight-forward a stream. At each Daddy's store, the salespeople are authorized to price, buy, and sell used equipment following Bramante-prescribed strictures. They are allowed to write a check to a seller on the spot and can even turn around and resell the item a few minutes later for a quick profit. Since some used instruments are purchased outright and others taken in trade, costs can vary significantly, even for otherwise identical items. Similar but unique, each must be separately listed with its own SKU number. And to confound the tracking system even further, Daddy's is among a dying breed of merchandisers who still loan store-owned goods to help out customers whose own equipment is being repaired.

If salespeople were to be able to do their jobs reliably, the details of such ad hoc transactions had to be made available to the other stores as soon as they happened. The same is true to stock transfers -- intracompany orders to send stock from one store to another. Daddy's transfer routine spewed out 30-page listings only twice a month, and keeping inventory moving among the stores relied on a century-old device called the telephone.

Bramante, too, had reached the point where he could no longer work with laggardly reports. "Our president has a thousand ideas a day," the exhausted Minott was to learn. "He analyzes everything to the nth degree -- salesmen's commissions, product lines, department performance, store performance. We had to make everything go faster. He didn't care if we did the whole thing manually, as long as we could keep up with him and give him the quality of data in the time frame he wants. Which happens to be by yesterday."

Taking all this into account, Wright and Minott decided what was called for was not finer tuning of the existing computer system, but a radical recomputerization of the entire process. "We felt we weren't even close to keeping up," Wright sighed. They sold Fred hard on it.

Now it was time for Bramante to do some stocktaking of his own. The leading retailer in the industry, a successful chain out of New York City, had just shocked the trade by unexpectedly laying off more than three dozen people. The CEO confided that he had put in a big computer system, but no matter how many people he hired to run it, they couldn't get the damn thing right. So finally he kicked it out and returned to old ways. Thus the layoffs.

This poor guy must have thought the computer was bleeding him dry, Bramante reflected. And the fear seemed real enough: the extra costs could put such a strain on available capital that a business might slip under before the machine had a chance to pay off. You could be forced to handle merchandise that wasn't in the best interests of the company, because you couldn't afford the better lines. Or maybe you'd need 50 products, but could barely pay for 20, so you'd be constantly out of stock. At that level, vendors don't do you any favors -- send the check, or go C.O.D. Meanwhile, the competition could be faring just fine, thank you, and you'd be giving them a chance to take potshots as you struggled defenselessly to gain control. "The question is, can they hurt you badly enough so that you never get beyond that point?" Bramante asked himself. Answer: control would be risky, all right, but without it, he'd never stop running scared.

For the next six months, Wright and Minott pieced together the data-entry forms and flow charts that could handle the information efficiently, at least on paper. When they finished, they had interconnected every area of the business on a day-to-day operating basis. "A lot of the problems small companies run into when they think of computerizing is that they don't have a manual system in place," the now-seasoned Wright observes. "They don't have the data, the files, or the people who understand how those work. First, we established manually what we wanted to do, then we looked for the system into which we could directly funnel all that information at a point in time. We wanted to take the essence of the way we did business and put it on a computer."

And for two years, the duo flew around the country between official duties, like weekend surfers searching for the perfect wave. Except that their grail was even more elusive: financial and operating control of a finicky business at a prepackaged price.

Some vendors offered software that would have forced Daddy's to alter its ways, in effect, some engineer dictating how he thought retail ought to be run. Yet others were willing to loan a consultant to study Daddy's peculiar routines -- at $100 an hour. That largesse was rejected as well, not just because it entailed "major bucks," Minott says, but because "we weren't willing to bring someone else in to duplicate the effort we had just made." Besides, their effort had taken on a special slant. "Most management defines what the system is and computerizes it," Minott says. "But we found you don't really know the system until you're on the floor; everything else is subordinate to that. You have to make the distinction between this is what really happens at the bottom -- this is the way our salespeople think -- and this is what you want to know at the top. Then you determine how you're going to get there."

The system they sought would give store managers the information they needed and would also satisfy their own and Bramante's need for control. "Fred was thinking that way anyway; we were just an extension," says Minott. "The right philosophy was there. It was the data that was so-so. Not only were the numbers to the right of the decimal no good, I had serious doubts about the ones on the left." On the other hand, the last thing they wanted to do was to use already computer-shy employees as guinea pigs in an extensive systems-development program. The storekeepers wouldn't tolerate more printouts they couldn't believe. To Wright and Minott, the mission was obvious: they had to be able to go in one day and announce to the stores and to Bramante, "We have a system ready to go -- and it works."

Easier announced than done, however. At that time, there was little on the market for small retailers. System after system fell short of the twosome's goals. The off-the-shelf program that finally did the trick was uncovered in Large, Fla., at Tyler Business Systems, whose specialized software for consumer-electronics retail outlets, the parties foresaw, could be adapted to instruments as well. Wright and Minott headed South and spent an entire week learning the program. No wonder you had so much trouble finding the right software, the Tyler people told them: we've never seen anybody who controls as tightly as you do -- and without a computer, to boot. Well, the two students lectured their lecturers, when you install a computer system in a company that doesn't have controls, all you get is a faster mess.

Their idea was to turn off the old system and go right into the new without an over-lapping backup, but they had lingering concerns about the potential mess that could be made by members of the staff who didn't trust computers. So in preparation, eight simple Apple IIe's had been brought into the stores to breed familiarity. "Apples ask dumb, cute questions that people feel comfortable with," explains Wright. "'Are you sure want to do this? . . . You didn't put the disk in the disk drive, so I can't read the information. . . .' The Apples let people know a computer could help them. It was a perfect transition."

Software and hardware came in under the budgeted $200,000 -- a virtually imposible feat even a year or two back. The modems in each store, for example, transmit data over telephone lines at an ultraspeedy 9,600-baud rate; in the recent past there wasn't any such thing as a 9,600-baud modem -- never mind a price tag of only a few thousand dollars each.

Ultimately, what clinched the package was its ability to satisfy the pivotal criterion they had sought from the beginning -- realtime data, posted at the point of sale. That meant an interconnected flow in which keyboard entries generated at the counter made their way simultaneously into every other aspect of the business -- inventory, cost of goods, gross margin, accounts receivable, sales taxes, payables, customer profile, and so on. The president could then shape any of that information into immediate guidance on complex decisions that used to take months to resolve -- if, indeed, ever.

In the highly integrated program, sales data from the day before is gathered the next morning for entry into the general ledger. At the close of each month, all receipts already will have been recorded and all transfers brought up-to-date. Less than 3 days are required to review and approve the month's figures, and in 10 days or less, profit and loss statements for each branch can be issued.

The end of the month has become far less hectic than the stay-till-you-get-it-done days of yore. Because they're carefully fed every day, financial statements come together faster, and the company finds that closings are sort of business retreats: blissful periods of reflection and planning. "We can look at things 15 ways from Sunday," says Wright, "knowing that the summaries are accurate. What's happening day to day is being managed day to day, and if it looks good day to day, it will look good at the end of the month."

But getting a P&L out in record time isn't the crux of the matter. In an industry whose average inventory turn notoriously lingers at about two per year, it was more important for Daddy's to look at its performance in buying and selling product lines. "If you can make return on inventory investment work, you can make a retail business work," Wright preaches. "Where are you going to make your most money out of the sales you project for the month? It's not the sales dollar, but the gross margin over time that runs our business. A lot of retailers haven't caught on to the significance of that multiple." The significance is that it's customer demand that is calling Daddy's tunes, and the lyrics are clear: buy high-margin products that are easily sellable at the counter, but be sure you can restock them quickly so you don't tie up cash on the shelf.

Daddy's new computer system clusters products based on the way customers perceive differences in them -- all guitars that sell between $300 and $500 might be one customer-defined grouping, for example -- and analyzes gross margins across the board within each demand category, as well as by specific brands. "We might find that we get a better return on investment over time on one brand versus another for the same type of item. So which one are we going to restock? Getting the customer to come again back fast is a priority, yes," Wright concedes, "but so is wringing the most from that sales dollar."

Accordingly, as soon as the system went on line, Bramante was given a daily report that included such breakdowns as sales projections for each store, and gross margin return on inventory comparisons by manufacturer, department, and item type. Costs that are less than 20% of sales are flagged. Recently, a keyboard was sold for $1,895 that was carried at a cost of only $11.40. "It used to be that we would shout, hooray, we made a lot of money. Now," says Bramante, "we accept it for the error it probably is." Because Daddy's floor staff is trained to cut its own deals with customers, Bramante closely watches for merchandise whose costs came in at more than 85% of sales. A side benefit of daily reports is that the perpetrator of too-deep discounting can no doubt remember the exact circumstance and perhaps can explain the reason for it, inasmuch as it took place only hours before.

Vendors, too, have been feeling the wrath of electronic data. Precomputer, Daddy's would purchase low-priced items by the thousands and, like most manpower-thin retailers, not bother to track the results. As long as they thought they were selling them, they kept buying them. But instinct no longer rules the warehouse. "This is a down-and-dirty, street-fighting, horse-trading system," admits Wright. "The worst news you can give a rep is, 'In 15 minutes, I can tell you what we've done to this point on anything of yours we carry.' Suppose we want to know if our return on investment on guitar picks is working out for the first 10 days of this month. If not, we won't replenish -- or we'll replenish with someone else's product, or we'll make the margin better." A leading brand was one of the first casualties of automated reins when Daddy's analyses disclosed that a competitor could supply the same sort of instruments at lower cost. Shown unassailable figures, the manufacturer was so chagrined by the low margins and tepid turnover that he bought back the unsold stock.

After the first six months -- from July to December 1986 -- Daddy's claimed that margins had improved by 2% to 3% (comparisons aren't wholly reliable yet, since the old figures are so sloppy). And internecine wrangling over shelf-stocking has subsided. "Thanks to the inventory tracking capability," reports Wright, "we don't go through the I-don't-have-this-yes-you-do routine anymore. I pull out the transaction by item number. 'It says on December 10th you received five, you transferred one, you returned one to the warehouse to be put back into inventory, and you sold two on these tickets. Therefore you have one left -- and it better be there somewhere."

Restocking is done through a retailing version of Just-in-Time. Sales analyses spot deviant curves within a couple of weeks, and when an item that was selling at, say, an average of 3 per week suddenly goes to 12, purchasing attends to the anomaly instantly. Like most companies in the industry, Daddy's used to order haphazardly every few weeks whenever a particular rep was scheduled to come in. Now, inventory is reviewed daily instead of monthly, and ordering occurs in response to critical minimum/maximum levels, rather than to the calendar. And the JIT nature of restocking is solidly rooted in what manufacturers' turnaround times actually are as revealed by the computer, rather than on what vendors promise. As a result of such efficiencies, Daddy's claims, inventory costs have been reduced by close to 15%.

Indeed, buying efficiencies may be approaching a time when budgets based on predicted availability of cash aren't meaningful. With the computer, inventory and sales feedback is so fast that you simply order what you can sell. "In the old days," says Minott, "the budget people would say there's only room for so many of such-and-such. Now we go from the retail end and say, 'My return on investment is 250%, and I'm going to buy more whether you like it or not, because we're making so much money on them that I don't care if you have to borrow to pay for them." Wright agrees: "If you're fierce about it and everything is integrated, the day-to-day stuff virtually runs itself. One arm can't be out of whack because we'll know it in a second."

When David Wright joined Daddy's Junky Music Stores in 1983, its founder was planning to add one or two stores a year. Complained his harried sales manager: "I'm not saying we shouldn't do that, Mr. Bramante, I'm saying we can't." Today, an insouciant Wright replies: "Two, 5, 10 -- no difference. You have the cash? You have trained salespeople? Where do you want them?"

For Wright, the biggest difference the computer system has made is in the precision of planning sessions. "We used to debate whether we actually made money selling magazines. I'd vote yes, Addison no, Fred maybe. We never had the data when we needed it. Those questions are no longer arguable. Five minutes at the keyboard, and the system tells us. There aren't any more vague discussions when everyone wonders, but doesn't know."

"The computer has given us the ability to manage," Minott concurs. "We would spend 75% of our time paper-handling and processing transactions, and 25% analyzing. Taking hours to decide the best way to do something is common when a management team doesn't have all the information. If it were a bigger company, it would only be a multiple of us. I can see how management sits around and plays Group Think for days on end."

Adds Wright: "Pretty soon Addison will go back to being controller, and I'll go back to marketing with the data I need. Life will be boring."

And Minott: "I often contemplate the mess our business could be in. But in January we can go skiing, rather than having to clean up December."

And the proud Daddy himself: "This takes the guessing out of it. The best guesser in the world can't do what we can do."