It is not easy to understand and manage changes in attributes and the impact of those changes. Strategy must recognize and cope with change over time, so needs a method for quantifying both scale and speed of progress. Now we must not only work out how key resources are changing, but also the quality of those resources. How is this done? The table below looks at how the customer base and sales for the manufacturing firm from briefing 24 would evolve if it started with a large number (240) of small customers (each buying 35 units per month), and set out to win larger customers, each buying 105 units per month. Its customer base increases, but its total sales rise still faster, and over two years the average customer size increases from 35 units/month to more than 58. It does not actually end up with 355 customers each buying at this higher average rate, of course. Instead it has the original 240 buying at the low rate and an additional 120 customers buying at the higher rate.Table: Adding larger customers to the manufacturing firm’s customer base. Table 1: Examples of attributes for tangible resources

Start of month	1	2	3	…	22	23	24	end
Customers	240	245	250		345	350	355	360
New customers per month	5	5	5		5	5	5	5
New sales per new customer units per month	105	105	105		105	105	105	105
Increase in total sales rate	525	525	525		525	525	525	525
Total sales units per month	8,400	8,925	9,450		19,425	19,950	20,475	21,000
Average sales per customer	35.00	36.43	37.80		56.30	57.00	57.68	58.33

This change in customer quality can be displayed as a time chart (below). As in earlier chapters, the causal arrows reflect specific, explicit relationships:

New sales each month = new customers per month * new sales per new customer

Average sales per customer = total sales per month divided by customers Depicting changing customer quality as larger customers are won. The attribute Total sales per month is shown as a resource, in parallel with the Customers resource. This is because, like customers, it accumulates and would deplete if customers were lost. In other words, sales per month is something that “flows with” customers, so this relationship is referred to as a “co-flow” structure. There are two potentially confusing features of this structure to clarify.

• The attribute of Total sales: units per month is increased by the arrival of new sales from newly won customers. The units of this increase are therefore “units per month, per month,” so the phrase New sales each month: units per month is used to avoid confusion.
• Why is the attribute resource given as total sales? If we want to track the average customer quality, why does the lower resource not contain average sales per customer? The reason is that the math works out much more easily if the lower resource contains the total sales, rather than the average per customer.

Extending the bathtub analogy introduced in earlier briefings may help clarify why this works. Imagine that the resource itself is the amount of water in a bath. The quality of interest is the temperature of that water. If you want a warmer bath, you can add hot water. How much warmer it gets depends on how much heat the added water brings with it. The initial temperature is the initial quantity of heat divided by the amount of water that shares it, and the final temperature is the new quantity of heat divided amongst the new, larger quantity of water. Unlike our manufacturing firm, however, the water does not stay divided into two separate amounts, one of which is hot and the other cool, but gets mixed up so that the whole bathtub reaches a similar temperature. Two options to improve the customer-base Adding larger customers is just one way to increase the average size of the customer base. An alternative is to remove smaller customers. Consider what might be done if the company’s initial 240 customers who are buying an average of 35 units per month actually consisted of 120 buying 50/month, and another 120 buying 20/month. The company could allow the smaller customers to be lost, leaving itself with a customer base increasingly dominated by larger customers. If five smaller customers were to go each month, the company would end up after 24 months with only the 120 customers buying 50 units/month. Although the company improves its average customer quality by this mechanism, it does of course suffer a fall in total sales. Consequently, it could end up making lower profits in spite of the better quality of its customers, depending on how much it can reduce costs as customer numbers fall. Until next time… [hr_shadow] If you would like to receive the series from the beginning in your email inbox, please register on our website and subscribe to Briefings in “My Account” Doing it right: attributes may “break the rules,” on defining resources Surely the additional resource in this figure is breaking the rules that were so carefully laid down in earlier briefings about what is or is not a resource? We insisted then that variables such as sales, revenue, costs and profits are not resources. Since their units are “XXX per time period” they are instead flows of orders or of money. That rule remains important, but the attribute resource in this figure is being used only to describe the quality of the customers, not as a resource in its own right. We can (and should!) still calculate Total sales from Customers multiplied by average sales per customer and the resulting chart of Total sales would be exactly the same as the lower chart shown here. Sorry if this seems pedantic, but it’s right – and important. Read more about the book on our website