Implementation & ABM

Sometimes, ABM can seem too difficult to implement. That's because of the multiplicity of cost-drivers. Some companies contrive to identify over 200 of them. While using just one driver is wrong--three is better, five is even better--200 can be far too high, as it increases the cost of the cost system more than the savings it can generate. Such problems often arise from using the services of consultants. Typically, a cost-bloated company in bad shape turns to an ABC consultancy, which says that it needs to do a full study using five people, over the next 10 months. This generates 200 drivers. Imagine, for a moment, that you spend 40 hours a week on a particular operation. The consultant lists the 200 activities that the company is spending money on, and asks you to allocate those 40 hours between those 200 drivers. Naturally, this never makes sense, but consultants do it to stretch their three-month assignments into three years. While using 200 drivers may provide more accurate results in theory, the data it generates in practice turns out to be wrong. Although the system may be grandiose, nobody can understand or use it. And the data collection carries a high cost.

So, an optimal number of drivers must always be used. My own rule of thumb is a maximum of 20 or 25 activities, but there may be some activities that share the same driver. For example, the number of transactions, or the number of vendors, may be common drivers to two or three activities. My data needs are really 10 different activity centres, with distinctly-different drivers.

It's in prioritising and identifying the key drivers that the company comes in. When I conduct an interview for implementing ABM, I always ask not for all the activities of a business, but for the five most important ones. It is possible that these cover between 80 and 90 per cent of the total business. I believe that it is better to be approximately right rather than precisely wrong. With 200 drivers, you may achieve fourth-place-of-decimal accuracy, but the integer side is questionable. To create another, connected, rule of thumb: any ABC engagement should take a maximum of three months. That's what the CEO must assert, and add that he'd like the exercise restricted to 20 activities, and 10 different, distinct, drivers. That's it. Within that perimeter, the system will not only work, but can also be used and leveraged for ABM.

Quality & ABM

Quality is great. But take the example of the Houston-based service company, Wallace, which got the Malcolm Baldridge Award for quality in 1991. In 1992, it went bankrupt. Is high quality leading to bankruptcy? Of course not. Motorola got the Malcolm Baldridge Award, and is doing extremely well. The quality movement has raised its marketshare and customer base. And quality has become the culture in the company. But you must determine the amount of quality you really need. For instance, do you really need a failure rate of one in 1,000,000? Isn't one in 10,000 good enough? Second, there's a lead time for quality.

An investment of, say, $10 million this year in quality programmes will generate benefits only three or four years later. However, a company that has cash-flow problems in that period will not exist long enough to reap those benefits. That is exactly what happened to Wallace. Contrary to what the book says, quality is not free. There is a cost involved. You have to be discerning.

Importantly, ABM allows the cost of quality to be measured, the benefits to be evaluated, and the failure rate to be examined. Consider a company that has identified external failures as the cause for losing business, and is investing, say, Rs 1 crore in improving the external failure rate. However, the question is, why is there a high external failure rate? Is it due to a particular component or sub-process? How can it be fixed? A deep analysis of its activities will enable the company to get down easily to the root causes of external failure. Next, the cost of external failure, in terms of, say, how many customers defect, can be calculated. Then, after improvements have reduced the rate of failures, the benefits, in terms of the quantum of reduction in the cost of external failure, can be measured. And, finally, so can the cost of fixing the problem since doing that also involves activities.

A second aspect of quality that ABM can be involved in is in responding to the paying power of the customer. Earlier, companies used to follow cost-based pricing--accumulating the cost, adding a margin, and fixing the final price. But now we are seeing price-based costing, or target costing. It starts from what the customer is willing to pay for a product--say, Rs 100. Then you put aside Rs 10 for profits, and Rs 15 as the cost to sell post-manufacturing. So, Rs 75 is your target cost at the manufacturing level. But if your manufacturing cost is Rs 87 at the moment, ABM will enable you to drop it back to Rs 75. All these factors make ABM not just an accounting language, but a business language for the entire corporation to understand--and use.