Marketing Implementation Articles

Part 3: The Quality Process in Sales

The problem with data base and application design for a sales and marketing information system is that what appears to be easy, becomes very complex very fast. This article is not going to be a step-by-step how-to, but hopefully, more of an explanation of the exercise, and some solid reasons to start the process. We hope to leave you with a different perspective of your computer system, and some useful tools, in the form of questions, that you may use when analyzing a current or future system.

In the above chart we have defined a typical "cube" of all the information a distributor deals with in his sales and marketing data base, with the customer information, product information, and sales territories on the X, Y, and Z axis's. That seems simple enough, doesn't it? Let's assume for a moment, that you have a data base that has the ability to "think" in three dimensions and you can use this magic report writer to define any kind of report you want. We have outlined four "mini-cubes" of information you may want to check our during the year:

Report #1 would be a report for one product (item, product class, vendor or a logical group of products) for all sales territories for all customers.

Report #2 would be a report for a specific range of products for a specific range of customers for one salesman.

Report #3 would be for a few products for all sales territories.

Report #4 would be for one customer for all products.

The point here is, if the data base is designed properly, and if your computer system is flexible enough, this type of information is available. In the past the problem was most computers were too expensive to store all this data on line, and the operating systems too inflexible. In the last five to ten years, however, the price of disk space has dropped dramatically, and modern operating systems have been standardized to allow for better data base management. That has reduced the hardware cost factor in the equation, but the software (data base design and applications) has not taken advantage of this yet.

The major reason for this, in our opinion, is the lack of demand by top management or sales management. In case you haven't found out, the 21st Century is right around the corner, and if you aren't knee deep in the information age, the distribution industry may pass you by.

At the risk of repeating ourselves, the variables for an "average" distributor with 1,000 customer, 7,500 items and 5 sales territories is 37,500,000 possible "cells" in the cube. Now you can add to that all the sub-cells necessary to do some exception reporting on a timely basis. (e.g.: examples #1-4 above) and then add to that the fields you would like to report on, but don't have, like SIC code or county or some aggregate product classes, and you can see why some careful planning early on is imperative.

Computers are generally very dumb. You need to tell them everything you want them to do (programs) and you need to specify every piece of information you want them to save (data base). They will not save data because they think you may need it, nor will they re-program themselves to make things easier for you. You must do the work necessary to get the reporting you desire, or look for a system that does most of it and have it changed to meet your needs.

Below we are going to share with you some of the thought processes we find helpful in analyzing the data bases for customers, products and territories.

Look for Interlinking of Bases

When you approach a customer data base some obvious data elements (fields) like name, address, both sold-to and ship-to, and some sales data are the easy ones to think of. But remember, if you don't define a field for a piece of information your system will not save it. We have looked at many systems, some with well over 500 fields for just the customer data, and many of these fields would store repetitive information in one multi-value field, e.g.: contact name(s) would all be in one field that could hold 99 name vs. contact Name 1 field, contact Name 2 field, contact Name 3, etc. This type of data storage is available in some operating systems and can make data base comparisons by just counting fields tricky. When you are looking at the customer data base keep asking yourself, what kind of information do I now need to run my business as it relates to customers, and what information have I wanted to get but isn't available in my current system? Then look at the data base for the fields necessary to get that information; if they're there then you most probably can have it. The main thing to look for is the interlinking of the customer data base to the products and territory data bases, because that will show you the overall flexibility of the entire sales and marketing data base.

Product's Data Base Can Be the Toughest

The product's data base can be the toughest one to design because of the interfaces to both the customer and the sales territory files. Remember, in the "cube" every report usually wanted some type of product limits? Well, you may want to aggregate many items into a sub-product class (all drills). Then you do this again and again, (e.g.: all drills, to all high speed steel tools, to all perishable tooling). If you do the design work necessary to develop multi-level reports, you're well on your way to exception reporting, and instead of dealing with reams of paper you can look at one or two pages. Then, if you see something either very good or very bad you can "drill down" through the next level of information, on an exception basis, to find out what the reason is for your concern. With a well designed product data base and set of reports, the answers to many of your product questions will be both fast and accurate. This will help you to make better decisions on your lines, your vendors and over-all sales performance.

Too Much Work, Not Enough Time

The final axis in the cube is the sales territories. This can be the most important one to develop, because of the direct effect it has on the bottom line. The more information, on good form, you and your sales people have in a timely manner helps them use their time more effectively, which will help your bottom line. If you could sell all your products to all your customers, you could probably increase your sales ten times without taking on one more customer or line. One of the major reasons, I believe, this doesn't happen, is because most mature sales territories have too much work and not enough time.

This happens because the sales person doesn't have enough information in an easily used form to show him what the most profitable use of his next hour, day, month or year could be. They need to be able to analyze, in a logical way, the products, their customers use, the dollar potential of each customer, their time available to establish that account vs. the time it could take to establish the next account. Then each quarter or so, do this type of ranking all over again to make sure the sales received are worth the time spent. You can flip this around to focus on products too, so that you make sure that each product line is worth the time spent. On the territory axis again there are links required back to the customer file for things like mailing lists and specific customer information, and also the product file for the similar reasons.

Well, now you have all three data bases built to give you the ability to trap all the information you need to use, and the reports or screens necessary to retrieve it in a fast, usable manner.

Let's think about what you had to do to get to this spot. First you sat down and listed all the individual elements of information you wanted to have to report the sales effort of your company. Then you listed the interrelationships to each other all those elements might have. The tricky part then became finding a computer system and operating system that could handle the complexity of the data base design necessary to implement your marketing data base. And if you have done a good job you now have the ability to track and report on maybe 500 fields for 1,000 customers and 500 fields for 5 sales people plus 500 fields for 7,500 products, or if each customer has only one salesman assigned, and only buys 10% or 750 items from you, multiply 500 x 1,000 x 500 x 1 x 500 x 750 = 9.375 x 10 to the 13, or 93 trillion possible locations in your cube. If you are buying fixed field disk space this is not possible; if you're using a variable field operating system the cost, if possible, is very large. So what do you do?

Get Data Without Breaking the Bank

This is where a good working knowledge of computers combined with a good working knowledge of your business allows you to have most of your cake and eat it too. By choosing the proper operating system and eliminating double storage (only put data in one file and let the others access it as needed) plus limiting the collection of some minor data, but allowing over-rides on collection limits as needed, a solid sales and marketing data base can be developed using existing technology, without breaking the bank of the average distributor or manufacturer.

There are many software packages on the market today that do some data base marketing, but we have not seen many that are sophisticated enough to handle the average company's needs in the future. The ideal situation would be to have it on your main computer system, but most of them have not been designed to deal with this type of inter-related data manipulation. The next best thing would be to have a self-contained, fully integrated system that would use your main system's transaction data to update its data base on an on-going basis, and still be flexible enough to add custom fields or create ad hoc reports for your company's special needs.

There are a few more terms you may want to remember if you area talking data base marketing. They are "relational data bases" (meaning you usually store the data in only one place, and the other files that need it refer back to that place), and "SQL" or structured query language (meaning a retrieval system that will use a standard system application to generate reports, on an ad hoc demand). Both of these concepts, if available, make any system more flexible and user friendly. Another consideration is single or multi user; if you want the salesmen to have on line access to some of this data you need to have a multi user system. Without going into a long explanation, our advice would be to stay away from a DOS-based LAN type system. DOS was never originally intended to be a multi user operating system, and when you use a DOS/LAN network you start to pay more for hardware and system software plus the necessary additional level of system knowledge required to run it. You could have spent less on the same P/C solution with a slightly more expensive operating system.

If you would like a starting point, Industrial Distribution Association has a publication named "Developing and Using a Marketing Data Base" for $69 for members or $138 for non-members or you may want to contact the association you are affiliated with to see if they have a similar publication. But don't try to use these as total specification manuals, because your company probably will vary from the norm, which will create immediate gaps in your sales reporting.

There you have it. A once-over on the next change the distribution industry is starting to go through, and the last department inside your company to be affected by your computer. Just imagine, a sales force that actually plans its work using more than just a gut feel, a sales force that actually gets rewarded for spending its time effectively, and management that actually has the information available to make the best decisions possible, and a way to implement and follow them up. If you need some help starting down this path and can't quite find it, give us a call, maybe we can help you out. GOOD LUCK!!