Saturday, February 13, 2010

The best time to plant a tree was 20 years ago. The second best time is now.....


I passed my management accountancy exam in 1992. This was the same time when Activity Based Costing, was at its peak (may be first peak). Although I was not aware that such a concept exists then. I heard about this concept for the first time, when I was asked during one of the employment interviews in 1993, “What do you know about Activity Based Costing?” I was zapped, as I had never heard about such a thing in my life. Life took me back to Activity Based Costing again in 2000, when I started reading about anything and everything about it on Internet. Since then I have been hooked on to this concept till date. The only reason why I telling all this is, when I look back, I see Activity Based Costing and my alliance with Costing have crossed 20 years of mark. In those twenty years I have changed a lot, as well as the concept of Activity Based Costing. Let me put the life of the Activity Based Costing for last 20 years as I know it. Correct me, wherever I am wrong, by posting your comments.

In the late eighties, the western organizations were facing competition from the Japanese organizations. These Japanese organizations (especially in electronics and automotive) had a distinct advantage of cost over the western organizations. This was due to the various management techniques used by Japanese (SPC, JIT, Continuous improvements etc.). At the same time some of the western organizations were looking at their decreasing margins, without knowing the real reasons behind it. During that time costing was used for completing the financial reporting by calculating the inventory costs. The same information was used to calculate the product cost. This was obviously leading to ‘not-so-correct’ product costs and decreasing margins.

This is when people built their first ABC models. These models provided them the information that helped them to understand that the basic reason for the wrong business decisions was the cross subsidizing product costs. Organizations then started ABC results for pricing and product mixes as well. With this came the flood of ABC implementations. Both the consultants and organizations went overboard. The models were designed with in detail (sometimes in too detail). The corresponding technological solutions could not cope with such detailed models and this led to the early reactions. ABC is very much time consuming. ABC takes too much time to build. ABC was also used for product costing only, then. So it has to face the criticism of not being customer focused or process oriented. Also the followers of ‘theory of constraint’, also criticized that ABC can only be used for long term decisions and not for short term decisions.

A lot of changes have happened to the modeling concepts, technology, uses of the concepts etc. The irony is that, people criticize the ABC concept even today, based on the state that ABC had 20 years back. It could be so because they do not know about the changes that have happened to the concept of ABC. Let me give a try to provide this information.
The concept of costing was relevant to manufacturing industry, because costing was always synonymous to ‘Product Costing”. With this understanding the modeling was such that each and every ‘penny’ spent was taken to the product. Over the years people have understood the ‘cause-and-effect’ relationship of costs and now the ABC models separates the costs that are caused by Products, Customers, Capacity and costs not related to any of these (business sustaining costs). Due to this segregation of costs organizations are able to calculate and understand the product as well as the Customer profitability also. As the costs are separated for the resources provided and resources used, organizations are able to calculate and understand the resource utilization of the various functions. Earlier capacity was always synonymous to machine capacity. When you know the capacity utilization you can use the concept to plan your resources. Hence, ABC can be used for resource planning.

In the early days of ABC modeling it was a single stage modeling. This means costs from resources taken to activities and from activities to the cost objects (products). The current ABC models can have multiple level assignments. You can assign resources to resources (Cost of HR function assigned to various other functions), activity to activity assignments (secondary activities supporting primary activities), reciprocating assignments (HR providing services to IT and IT providing services to HR in turn). By using various attributes one can view the cost of process at various stages. For example we can see the cost of procurement process at various stages as requisition, purchase order, QC, returns, payment etc. This analysis helps the organization to directly attack the areas that are inefficient. All this is possible because of the technological solution as well as changes that have happened in modeling.

The changes in technology have also happened in parallel. From the early days PC based solutions to the current enterprise wide application, is one way. Various functionalities like multi-stage assignments, reciprocating assignments, attribute etc. For me the most important is the ability to do the multi-dimensional modeling and viewing the results as OLAP views (for those who not so technical OALP are very similar to the ‘pivot tables’ in MS Excel). With the use of this feature we can not only see the product or customer profitability, but this can be seen as product wise – customer wise profitability (or vice-a-versa). This combination can possibly lead to understand profitability of a customer – for a product – sold in a region – thru’ a channel. Alternatively one can see profitability of a customer can be seen for a telecom company as a set of customers falling as private customer – acquired thru’ retail outlet – with a prepay plan – having very low technology aptitude – with a life stage as matured family – age group of 44 to 55.

As I said the costing as well ABC was initially used in manufacturing industry. It has had such an effect on people’s mind that people still think that ABC is useful for manufacturing industry only. In the early nineties came the CAM-I cross.


This diagram helped organizations to understand that ‘Activity’ was the most important part of the ABC model. Process in nothing but set of activities performed in sequence by one or more functions. Analysis of cost drives as the ‘cause of the cost’ helped organizations to use ABC results to improve their processes. Here was the answer to the people who criticized that ABC is useful only for product costing (and hence for the strategic use and not for operational improvements). This modeling concept led to the use of ABC in various service industries because the ‘products’ of those service industries were their ‘services’ and nothing but various processes. Initially the service industry used this concept to calculate the customer segment level profitability. In case of Retail it was used to calculate the category level or product group level profitability. The technological challenge was, it could not handle the millions of assignments. Today the commercial software solutions are available that can handle 100s of millions of assignments, which can be used to calculate profitability at subscription level in telecom, account level in banking or SKU level in retail. The time driven activity based costing equations have helped to model these millions of assignments very easily.

The latest technological evolution is the integration among the various software solutions. Because of this, the data required for the ABC model to update the model can be directly pulled from the ERP or any OLTP application running in the organization. For the ‘non-ERP’ data we can develop a small web-based solution, so that employees can enter the empirical data for the ABC model. The results of the ABC model can be integrated with various Customer Intelligence (CI) solutions like segmentation, retention, campaign management etc. ABC model can feed up-to 20% of the KPI data in a scorecard. This has increased the usability of the ABC information.

The use of ABC has also changed over the years. It started with calculating product costs, to customer, channel, Business unit profitability etc. This has been used to calculate the resource planning and activity based budgeting. After creating the activity based budgets, organizations can start reporting activity based variances. Route optimization in supply chain for various organizations like Retail or CPG, can be supported with more accurate costs of various activities at multiple places. This cost information coupled with optimization techniques can help them to find the optimal route for various vendor-item category-location combinations. ABC information can support any other process improvement program that is running in the organization, by providing accurate costs of various processes and help to prioritize the program. For this we can use the ‘2 x 2 diagrams’ methodology. We can use cost v/s potential to improve (the potential to improve will depend on how badly you are performing the process or activity). From here we can choose the quadrant that is ‘high cost and high potential to improve’. For the activities in this quadrant we add another attribute ‘potential to change’ (that is how easy it is in the organization to change the way in which we perform the activities). The first set of activities would be then ‘high cost – high potential to improve – high potential to change’. This seems to be the low hanging fruit.

It is no longer that the organization would start looking at business as a portfolio of customers. Then managing business is managing these portfolios of customers. To form the customer portfolios, customer profitability would be one of the most important information. Soon the board of directors would start demanding the customer profitability information and will challenge management to act on it. Who knows, shareholders would also start asking for this information. While mergers and acquisitions people will start running ABC project to understand the customer profitability, because in the acquisition, organizations are acquiring a portfolio of customers. The bid would also depend on the portion of the customers that match the profile but also portion of the customers that are profitable and matching profile.

During all this discussion, I wanted to explain that since the introduction of the concept of ABC till today and in future, it has gone through various cycles. Some people have used it, some have criticized it. It has taken its path of crests and troughs. Looking at the maturity of the use of the concept, technology available, consulting resources and competitive market situation, I feel the right time to start using ABC is TODAY. I will modify my title little bit and say “The best time implement ABC was 20 years back. The second best time is now….

10 comments:

  1. It has been my experience in the last 12 years, that most of the driver information is available in the organizations, in some form or other - they may not have been captured eletronically - but they are available in log books, adhoc reports made by the departments from time to time etc. The challenge of getting the information from Ovehead service departments has also been squarely addressed by TDABC.

    Wherever there is a real urge to reduce the costs,getting the right information has neither been cumbersome nor time consuming.

    ReplyDelete
  2. Rajen has written a comprehensive history of ABC. I was fortunate to have gotten involved with it in 1988 with Professors Kaplan and Cooper with KPMG, and I have never stopped my participation (e.g., writing books, articles and blogs).

    One "era" that was disturbing was when ABC models were so over-designed and complex that some failed to be maintained. It gave ABC a bad reputation as if it did not work. The methodology is obviously sound, but the implementtion approach was the culprit.

    Gary

    Gary Cokins, SAS

    ReplyDelete
  3. One more comment ... this about time-driven ABC.

    First, I suggest we stop referring to TDABC as "revolutionary" and ABC as "traditional." TDABC is basically just an alternative technique for selecting a type of activity driver.

    Next, one should consider the special conditions where TDABC may apply. It is not for everyone and every problem. There can be debates that TDABC is a lower administrative effort and cost to construct and maintain, but I would argue the opposite. Therefore, to incur the extra effort for applying TDABC I suggest you confirm the presence of these types of conditions: highly repetitive processes, predominantly direct expenses with relative less complexity and magnitude of indirect expenses, and management's suspicion and high concern of the existence of unused or excess capacity (for TDABC to "deduce.") If these conditions are present, then TDABC may apply. If it were me implementing, I would first use the ABC rapid prototyping and iterative modeling approach for a couple of weeks to "see" the architecture of the ABC cost assignment network to gain understanding and buy-in from the eventual users. (There is an IMA Statement of Management Accounting describing this titled "Implementing ABC.")

    At that junction, you could iterate one or more times to further refine the ABC model and then switch it on as a permanent, repeatable ABC production system. Or if TDABC makes sense at that junction, then convert the ABC model just constructed to TDABC.

    Gary

    Gary Cokins, SAS

    ReplyDelete
  4. Gary,

    Wow - its been a long time since I have talked/heard from you. I think we met at an ACB Technolies user group meeting in New Orleans.

    You are absolutely correct that implementation is the key factor in ABC failures. I believe that most ABC implementations are too complex and often result in 'Shadow' accounting departments. Your rapid prototype idea is a good first step for many organizations prior to investing in a complete roll-out.

    I've out of the ABC world for a few years - any ideas on how to get back in. I'd love to talk with you; feel free to call me at 510-301-1397.

    Best Regards,

    Mike Wagner

    Also- thank you Rjen for the post - I have no idea how you found me! Excellent overview of the ABC discipline. It brings back memories when I had the opportunity to listen to Kaplan in the early 80's before ABC was even a discipline.

    ReplyDelete
  5. Hi Rajen,
    spot on. I have had the opportunity to apply the ABC model for services more in a crude form then really in a scintific model. I can identify & agree with the points you are making. We did achieve instant results, from practically a situation where we were spending more then the revenue, ABC helped us to narrow our focus to areas that needed to be controlled. We did by identifying 3 out of 12 heads of expenditure (I agree with vasudevan- most of the data is readily available within the departments - someone is required to collate it to make sense, which I was pushed into against my will initially). Behold! we did get the desired results. since then I am a firm believer of the Activity based costing.
    I rather enjoyed the 20 yrs background to the subject. A sapling is indeed planted in my mind. I hope to track this column and learn more from illuminaries such as Gary, Mike and yourself. Thanks
    Vijay

    ReplyDelete
  6. Hi All,

    Gary, as usual, is absolutely correct about complexity killing ABC models. I've personally been involved in large enterprise projects where the client wanted a high number of low level activities and wanted daily surveying of staff to ensure the model was "accurate", this is simply not sustainable and we were able to convince them otherwise. Also our experience is the same as Vasudevan's where we want to capture as much of the existing data to populate driver formulas to make the update process as easy as possible. If the update process overly burdens the organization then it has a higher chance of failure, like capturing detailed times for TDABC, unless that data is available electronically.

    Mike - there is a CAM-I quarterly meeting coming up in your area: June 6 – 9, 2010 – Sheraton Fisherman’s Wharf, San Francisco, CA, if you would like to mix with some ABC folk again. www.cam-i.org

    Cheers,

    Lea Patterson, Pilbara Group Inc

    ReplyDelete
  7. Some vertical M&A focuse on product integration. This allows end-to-end product delivery.
    For horizontal M&A, companies focuse on customer or channel integration to enlarge market share. Beverages industry is a good example. May be ABC is a good tool to tell the story.
    However, I wonder what is the best time for a company to implement ABC? Before, during or after M&A?

    ReplyDelete
  8. I am currently reading Tom Davenport's books on Business Analytics and found ABC as a Business Intelligence application for internal processes in them.

    One of my key take-aways is that ANY application of analytics and especially the insights from analytics need to end up in the corporate culture and be used and understood on a daily basis.

    In 'Competing on Analytics' he describes stages of analytical capability and it is interesting to see how ABC failures correspond to his early/lower stages of analytic competency where there is lack of executive commitment, quality data, analytically skilled people, insights permeating into decisions, and even strategic leadership.

    I like an old quote of Gary Cokins:
    "If your senior leadership cannot articulate the basic principles of an improvement initiative, then employees will never achieve or sustain the initiative."

    ReplyDelete