First steps toward Lean Accounting

Companies seeking to become lean need a new management accounting to support the management of value streams. This new accounting, which we are calling Lean Accounting, must be part of a comprehensive system for the immediate identification and resolution of problems. This article will discuss the reasons why traditional accounting is inadequate and point out some directions for building Lean Accounting.

The two accountings: financial and managerial

Traditional financial accounting, governed by laws, principles and procedural conventions, aims to provide information on the financial position of a given company and on the financial flows it was subject to during a period of time. It is through financial accounting that investors are informed about the return on capital to which they are entitled, creditors learn about the company's solvency, the tax authorities identify their share in the form of taxes to be collected and managers can see the financial results of the business they run.

However, financial accounting does not serve as the only management tool. It meets the control requirements of the tax authorities, investors and creditors - assuming there is no fraud - but it falls short as a source of information on the company's performance across several dimensions. Besides portraying the past in a static way, financial accounting presents overly aggregated numbers, does not reveal the effectiveness of operations, is governed by accounting principles that are at odds with economic reality, etc.

That is why management accounting is necessary. Companies and their managers need other information on the company's performance that traditional financial accounting does not provide. Such information must portray the company's operational reality and can be a precious source of feedback, capable of guiding decisions and the actions that follow from them. But management accounting, as it is structured in most companies, does not seem adequate to us.

Problems with management accounting as it stands today

Implementing and maintaining management control systems - mostly computerized - is expensive and requires a considerable workload. From the lean perspective, the efforts to collect, synthesize, interpret and disseminate information about the organization are type 1 muda (waste): they create no value for the end customer, yet they are indispensable. Since they cannot be eliminated, such efforts must be reduced: the work must be simplified and directed toward supporting the value stream. Not only management accounting but also financial accounting can be classified as type 1 muda .

Another problem: many of the indicators generated are based on mass production premises, since they are tied to production volume and pace: the more and the faster, the better. Such indicators lose their meaning when pull logic is adopted, in which the production pace must follow the customer's buying pace. I deal with this problem specifically in the article entitled "Indicators that deceive".

But the big issue related to management accounting is this: it is designed to support "management by the numbers" in "multi-departmental" organizations. And here we enter the innermost core of the management mindset that has prevailed in organizations for at least 60 years. I do not believe this is the place for details on the origins and characteristics of this management "model," but I will point out some of the problems that stem from it.

"Management by the numbers" is the kind in which the people responsible for running the business learn about its performance through certain indicators and make their decisions on that basis. Indicators, whether financial or not, are elements of feedback whose function is to guide action. This practice is so deeply rooted that it is hard for us to imagine how it would be possible to manage in any other way.

"Management by the numbers" results from - and also feeds - the enormous gap between knowledge, responsibility, feedback and action that exists in companies. These four dimensions rarely go together or refer to the same group of people within organizations. This creates an enormous mismatch and makes improvement initiatives rare: those who hold the knowledge do not do the work, those who do the work do not receive feedback and are not responsible for it, and everyone is "measured" so that those responsible for the business can understand what is going on and only then act.

What makes it possible to create a sustainable dynamic of continuous improvement is the capacity to learn. And we only learn when we know where we went wrong. Imagining that the process of improving can always do without those directly involved in the work is an illusion, and some companies are beginning to realize this; others have known it for a long time...

A "multi-departmental" organization is one in which the whole thinks and acts in parts and, most of the time, in a conflicting way. Thus, companies are divided into departments, sub-departments, business units, cost centers, etc. Not that companies cannot be divided in order to be better managed; after all, there are areas with quite distinct scopes of activity. The problem is that the criteria for this division disregard the concept of the value stream.

As a result, the most important thing is not seen holistically: the steps needed to deliver value to the customer. Note that the value stream usually runs through several traditional departments: purchasing, logistics, manufacturing, maintenance, sales, etc.

"Management by the numbers" is management accounting's reason for being, and the "multi-departmental" organization defines its structure.

Lean Accounting

A lean company must have management accounting geared to supporting the management of the value stream, revealing waste. It must be a means of motivating improvements through meaningful, clear and immediate feedback signals for those involved in the work. And, above all, it must be backed by a structured problem-response system.

Traditional management accounting provides "relevant" information to companies' top management, but there is no transparency for most people. Feedback is not very democratized, and problems are identified late, with the risk that they will not be identified at all. There is a very large distance between the real causes of problems and the "symptoms" that appear in the form of indicators in management reports: the people close to the work are able to identify the causes; managers only see the "symptoms"; if the "diagnosis" is mistaken...

So what should be sought is to make the right signals available to the right people at the right time and to create the means for the response to be fast, precise and triggered in a structured and systematic way by those who are close to the work, coordinating and operating the value stream or the areas that support it.

Value stream coordinators must be in touch with feedback elements that reveal its holistic performance. Operators must be in touch with indicators that reflect the result of their individual actions and of the team they belong to and interact with in their work microcosm. People who work in support processes need to know whether those processes are in fact supporting the value stream or are an obstacle to its good performance.

An alternative approach

Brian Maskell, during one of the courses he taught at Lean Summit 2002, presented a very interesting set of indicators for measuring the performance of the business as a whole, of value streams and of individual cells. Some companies already use some of these indicators, but few do so in an integrated and effective way, that is, democratizing the information and creating the means for actions to be triggered. Below is a summary table with the measures suggested by Maskell as a "starter set":

Strategic Measures	Value Stream Measures	Cell/Process Measures
Sales Growth	Sales per person	Day-by-the-Hour Production
EBITDA	On-Time Delivery	WIP-to-SWIP
Inventory Days	Dock-to-Dock Time	First Time Through
On-Time Delivery	First Time Through	Operational Equip. Effectiveness
Customer-satisfaction	Average Cost per Unit	-
Sales per employee	Accounts Receivable Days Outstanding	-

Source: Lean Summit 2002.

I will not comment on the indicators: I include them here in order to bring an objective alternative approach into the discussion. Even though this is only a "starter set" of measures, as Maskell made very clear during the course, some very important dimensions will be brought to light if these indicators are used: customer satisfaction, delivery performance, inventory levels, the ability to produce with consistent quality, cash generation, etc.

One step further

The nature of the indicators and the transparency given to them are very important factors, but what is fundamental lies beyond the numbers: what really matters is the way in which the actions suggested by the feedback signals are put into practice, by whom and with what actual results. What we want is to create the means for improvements to happen naturally and to be part of the very nature of the work.

Interesting examples of this integration between feedback, action and immediate results come from Toyota.

One of them is the role that the heijunka box (or production leveling board) plays on the shop floor. It is a visual tool developed by Toyota whose function is to sequence kanban cards according to the constraints of the production process, communicating the production pace and mix . An operator familiar with the board can understand what is happening in their work environment with a simple glance at it: it is possible to know whether production is behind schedule or not, what must be produced and when.

Heijunka and the context in which it is used show us how to use simple signals so that those who are involved in the work, and are responsible for it in Toyota's case, are also immediately informed about the results of their actions. And what is more important: the response to problems is also immediate and aimed at preventing the problem from happening again....the knowledge is built into the system!

Another example: when they detect a problem, operators have the autonomy to stop the production process. Could there be a clearer and more immediate signal that things are not going well? Depending on the severity of the problem, other signals (visual and audible) are activated. These signals trigger instant, coordinated actions, which have a predetermined time to solve the problem.

Knowledge, responsibility, action and feedback go hand in hand at Toyota. The two examples cited, which are part of the so-called "Toyota Production System," are emblematic embodiments of this union.

A concept under construction

Lean Accounting must therefore be understood as a tool for organizational learning in the pursuit of ever better future states. We are proposing something that goes beyond accounting itself. It is about creating structured problem-response systems, in which actions, guided by carefully chosen signals, are triggered by those who are directly involved in the work and should be responsible for it. Lean Accounting needs to be a means of bringing knowledge, responsibility, action and feedback closer together in companies.

As the title of this article says, I have tried here to sketch the first steps toward what we are calling Lean Accounting. It is a concept under construction, whose practical applications are still very incipient. We believe this is an extremely important topic and we are eager to exchange experiences and knowledge. Your collaboration in developing and applying Lean Accounting will be fundamental. That is why we ask people - and companies - who are interested in the concept and its application to get in touch with us. We are open to criticism, suggestions and the joint development of knowledge.