Do you know the 'vital signs' of your business?

Most of us rely on monthly and year-end financial statements to keep track of the health of our businesses. But I suspect that instinctively you keep your eye on some other vital signs as well. What you may not have done is examine in an organized way the key data you should be monitoring regularly.

I have never been in a job, either during my 30 years with Fortune 500 companies or in my own distributor operation, in which I haven't isolated and monitored the vital signs unique to that particular business. As a result, I have always felt in control and have had an early sense of even subtle shifts in the business.

About 12 years ago, when I became the chief operating officer of Kearney-National Inc., I saw just how important this can be. Earlier, Kearney had acquired an inner-tube manufacturer with sales of about $30 million, which had been owned and operated by a very sharp entrepreneur. When the owner left, the new management restructured the company along "modern" lines. One of the first changes was replacing the entrepreneur's tracking system with a modern standard-cost system. The trade was a disaster.

After digging into the mess, I realized that the prior owner had one of the cleverest monitoring systems I had ever seen. For example, he received a daily report on the percent of virgin versus off-grade material being used in a critical mixing process. This figure kept track of how aggressively the operators were taking advantage of cheaper materials to achieve good final product. It also let him know when the operators had lost control over this large and pivotal process. The tip-off was that they used a higher proportion of the more expensive virgin material as a quick way to regain control. Since the owner knew that the mixing operation was the heart of the company, not only did the percentage tell him the economics of that operation, it also gave him a feel for how things were going overall.

There were more than a dozen other figures that this man tracked on a daily or weekly basis. His system had a wonderfully simple elegance to it, and I can assure you, if his operation missed a beat, alarm bells went off immediately.

The new standard-cost system would have been great as a complement to the prior owner's system, but it wasn't even a close substitute. The standard-cost data came out monthly, lacking sensitivity and timeliness. (The company never did recover from the transition, which included other, similar blunders. It was finally broken up and sold piecemeal.)

* What kind of information should you track?

Although some information, such as sales volume, is common among most businesses, other information is very specific to your own. So your first task is to identify what best measures the ebb and flow of your business. In my $12-million automotive-parts distributorship, the elements are sales, gross margins, out-of-stock levels, personnel counts, and overtime hours.

Your company may be entirely different. A service company, for example, might be interested in hours billed, time spent on development, and so forth. A commercial real-estate business might be sensitive to a tally of square feet for which commitments have been received. In a company where shipments lag behind orders, booking data may be vital.

* How often should you get the data?

Some of the information you track may be daily, some weekly, and some monthly. For example, rather than wait for the end of the month, I track sales and gross margins from our invoice register every three days. Why three days? Single days are too sensitive: the numbers jump around due to random events, and I'd be reacting to aberrations rather than trends. Three-day totals smooth out the spikes and reveal bona fide trends. In your business, daily data may be just right. In another, it may take a week for the statistics to settle down.

The point is, there's no simple formula for determining the right time interval. You have to combine experimentation with common sense. Do not, however, settle for monthly data when more timely feedback would give you a better check on the tempo of your business.

* How do you organize the information?

How the information is presented makes a world of difference in how effectively it triggers action. It's often difficult to make any sense of a raw number, such as total sales or hours booked. But the same information presented as a percentage or ratio may give the data instant meaning. We convert total sales to sales per day. Not very profound, but it makes the information meaningful and provides immediate comparison with our sales plan, which is also stated in sales per day.

For you, percent of available hours billed may be more meaningful than total hours billed. Or overtime hours as a ratio to sales may be more significant than just total overtime hours. Again, there are no correct answers or absolute techniques. Whatever works best passes the test.

* How do you gather the data?

There is a great tendency today to believe that all information must come from a computer. If the information you need is readily available from the computer and easily transcribed into your vital-signs log, then, of course, use it. Do not, however, pass up the opportunity to gather the data manually if that is the fastest way both to initiate the system and get the results on a day-to-day basis. In fact, the data you need may not be in the normal computer flow.

In my case, sales and gross margin data come from routine computer runs. It would be difficult to gather them any other way. But we collect personnel counts and overtime information manually from time cards. We get order-fill statistics in still another way: statistical sampling, which is an approach seldom used in companies (see "On Your Own," below).

Your primary goal is to develop an information system that will let you know immediately when disturbing trends start to develop, and to present data in such a way that you can act on the information in time to head off problems.

Sometimes you'll find that monitoring basic vital signs is not sufficient, and you'll want to create a special onetime information gathering and reporting system. For example, when I was general manager of the communications wire division at General Cable, demand exploded for a particular type of interior wire after the 1972 California earthquake. Suddenly we were under the gun to produce well beyond our normal levels. Our reputation was on the line. Production started to increase, but not at the rate we needed, so we had to create a tremendous sense of urgency -- and create it fast.

The solution we came up with was to have each shift superintendent phone the vice-president of manufacturing, at home if necessary, and report the production figures after each shift. While this may have violated a whole array of management principles -- for instance, bypassing layers of management -- it got attention and, more important, results. We dropped the system as soon as the numbers began to come up strong. You don't want to overwork this approach, but when the need exists it's a very powerful addition to your regular tracking systems.

Charles J. Bodenstab is chief executive officer of Battery & Tire Warehouse Inc., in St. Paul. Previously, he held executive positions with United States Steel, General Cable, Kearney-National, and Gould.


What to do when the data you need isn't routinely accessible

Sometimes you'll find that the information you feel is important to monitor isn't available, either on your computer runs or by manual tallying. Then you'll have to get creative.

In my business, one of the signs of trouble is when we don't have products in inventory to fill orders. But getting information about order-fill levels is extremely difficult, if not impossible, on a routine basis. You'd have to log into the computer every inquiry, both from the order desk and the sales staff, whether the product was in stock or not. That is no small task, and one with very little return.

So we initiated a statistical sampling program to be done every six to eight weeks. For one week, our order clerks manually log whether each line item requested is in or out of stock. All it takes is a pencil stroke for each entry. When the order desk gets too hectic, we stop the sampling until things calm down. This method gives us a good sample of our customers' requests and our ability to fill orders.

As far as we can tell, the information we get by sampling is fully representative of the total picture. There's no reason to believe that the mix of orders during the hectic hours of the day is different from the slow period or that the orders given to the field sales force are different from those called into the desk.

You, too, will no doubt find areas in which getting the information you need requires looking beyond your normal data-gathering processes. My experience says it's well worth the effort.