Distribution —a positive sales factor
Page 81
Page 82
If you've noticed an error in this article please click here to report it so we can fix it.
HE growing recognition of physical stribution as a tool of marketing and sales plicy was the underlying theme of a recent lk by Mr. Wendell M. Stewart at a British Lstitute of Management forum in London. ntitled "Cost analysis and customer serce", it was the third forum in a series woted to physical distribution. As a vice-president of Chicago-based anagement consultants A. T. Kearney and o., Mr. Stewart recognized at the outset at many companies did not have immentely available the costs required for desion making in this sector of their business 3ivities. Alternatively what costs were vailable were dispersed over several deartments and left unrelated.
A first step to obtaining the information .isential to such decision was to determine that was meant by physical distribution. [ere Mr. Stewart listed the following activies as being interrelated within overall istribution: Distribution planning and ost analysis; inventory management; inound freight on raw materials, sub assemlies, supplies, etc.; receiving and storage of ibound materials; order processing and elated data processing and communications ctivities; packaging; warehousing at facories; local delivery to customers; customer ervice.
Mr. Stewart said that customer service perated in a similar manner to quality ontrol to manufacturing. How well the istribution system was operating was indiated by the end objective, namely satisfied ustomers.
Not enough attention, however, was ;iven to initial planning and analysis. What ittle was done was too short term and tsually limited to accepting the volume of products provided by the manufacturing lepartment, an operational function and not planning in the real sense. Also analysis of listribution costs was a function too often left solely to an accountant.
If the number of customers dealt with, say 3,000 upwards, was too large to be considered as a whole for the purpose of cost analysis, it should be broken down into "freight centres". In the UK such areas could possibly be equivalent to the counties.
Such a geographical pattern then became the basis for distribution planning. The object was to achieve a relatively simple way of analysing distribution costs and not overburdening oneself by becoming too complex and too detailed.
Here Mr. Stewart interposed that in relation to planning, management should not be continually immersed in "fire fighting" current problems when in fact they should be concerned with long-term projects. He said that control by exception had become an accepted function of production management and administration. Likewise distribution cost analysis was concerned with the allocation of costs to those exceptions.
In essence distribution was an attempt to reconcile relatively stable production with a volatile marketing policy.
Penalty' costs
Here Mr. Stewart told of a company with a sales bonus policy which resulted in onethird of its quarterly output being distributed in the last three days of a quarter. This sudden and marked increase in sales seemingly justified the policy. But no account was being taken of the additional cost of errors and overtime that this upsurge in demand of distribution services entailed.
This was just another case of no-one being responsible for collating all the "penalty" costs. When it finally was done it was realized that the whole exercise was unprofitable.
Inbound freight was a function which some would contend was not a distribution problem and should be the concern of the purchasing manager. But whether or not inbound freight costs were at present lost in the supplier's inclusive invoiced charges, it was essential to have such information if proper consideration was to be given to the running—including back loading—of an own-account fleet of vehicles.
Although it was not necessary for a distribution manager actually to control order processing it was an essential element of distribution costs. He should therefore be aware of such costs so as to facilitate comparison between one distribution process and another. Here he emphasized that it was necessary to remove the traditional boundaries in inter-company organization so as to avoid decisions being made on a political rather than a functional basis.
As a management consultant he was not concerned with who was in control. But it was essential that he should know all the distribution costs.
More organizations were setting up a customer service as a form of quality assurance programme. But few had a continuing customer-service programme. For example they might be able to show that their average delivery time was seven days without being able to show for how many customers this might be too late or even too early. Few distributors had such a standard set of customers that they all needed the same standard of service.
Moreover, with so many companies developing a wide range of products it fol lowed that there would have to be varying levels of distribution service if they were to be provided on an economic basis. Here again, Mr. Stewart insisted, control of customer service must be by exception—he added: "Don't try to control every single service".
Mr. Stewart then gave examples of the cost of physical distribution in six broad industry categories as a percentage of sale. They were: Food and food products 29; machinery 10; chemicals, petroleum and rubber 23; paper and allied products 17; primary and fabricated metals 26; and wood products 16. The all-industry average was 21 per cent.
A breakdown of the cost of physical distribution as revealed in a survey for a client (with an overall percentage of 21.8, i.e. very near to the all-industry average) was as follows shown as a percentage of sales: Administration 2.4; transportation 6.4 (inbound 2.1, outbound 4.3); receiving and dispatch 1.7; packaging 2.6; warehousing 3.7 (in-plant 2.1, distribution 1.6); inventory carrying costs 3.8 (interest on investment 2.2 taxes, insurance obsolescence 1.6) and order processing 1.2, so giving the total cost of physical distribution as 21.8 per cent of sales.
Much of the present traditional forecasting and available statistics were too broad, Mr. Stewart considered. What, for example, could one usefully do with an overall figure such as "UK sales for 1970 will be £Xm". Admittedly it was initially time consuming to break down such overall figures, but once that exercise had been accomplished it was relatively easy to maintain them.
Such a breakdown was also necessary if transport costs were to be determined for each product category. This involved knowing what the average size of the order was so that one could be in a position to measure costs relative to both customer and market total. Otherwise if profitability went down there was no means of knowing what caused the drop.
In distribution costing there was ‘too much acceptance of the status quo. Thus a trend for customers to live further out of town could go unrecorded if regular resurveys were not made. Likewise there could be changes in order sizes.
Therefore distribution cost analysis should be on the basis of a variable budget and not fixed costs. This, Mr. Stewart claimed, was the only effective way of distribution costing.
A major problem was to obtain actual cost. In too many cases the data that was available was frequently factory oriented rather than market oriented, i.e. total outbound transport cost from a plant rather than transport cost to each market from each factory source. There was a need for a complete re-orientation of costs to market by market and product by product. That provided the only solid long-term approach and one should begin to accumulate costs from such an orientation.
Because the level of service provided had direct bearing on distribution costs, Mr. Stewart considered in depth the means of measuring present customer-service levels. As to what measures were appropriate, factors to be considered would be the number of orders shipped complete, say, within 24 hours and the number of line items, number of cases etc.
It was necessary to find out what priority customers attached to each line order. Thus a service level of 95 per cent of items dispatched in 24 hours could, in fact, reveal a much lower level of service for particular items.
The level of satisfactory service could vary with each customer. It was therefore essential to find out what the customer thought was important and not what the distribution manager, with the best intentions, considered to be a prior requirement.
Once it had been established what level of customer service was appropriate to meet a particular case it was then necessary to devise means of measuring and checking that level of service.
An initial difficulty was that one often did not know the date customers received deliveries, assuming they were not made on one's own lorries. But some companies had found it worthwhile to enclose a postcard with the goods for customers to mail back a record of the time of delivery and conditions of the goods. Mr. Stewart added that it had not been unusual to get a response as high as 40/50 per cent in the return of these postcards.
In checking the level of the distribution service, it had been found worthwhile to time stamp the initial order and then successively along the distribution chain-this could involve seven or more stages. The biggest blockage was often found to be in credit processes. But the alleged delay in the delivery was often found to be the excessive time taken in the customer's order reaching the effective point within the company. This was a factor which should not be omitted when reconsidering the siting of warehouses.
But only by obtaining the cost for the alternative levels of service could one assess whether a change in service levels was worth the extra cost. In other words it must be shown that it had a positive sales value.
The nominated-day concept of delivery was universally becoming more popular. The meaningful dates from the customer's point of view were those from the origination of the order until delivery.