Written by Guenther Tolkmit, Dreams and Details Ambassador
(Forgive the pun: “The Real Value of Numbers” by Mikael Trolle CEO & Partner at Dreams and Details.)
“The challenge will be that many businesses in the future needs to be developed agile. Numbers are still significant, but they should rather reflect the development of our performance; meanwhile, we unleash the potential of our strategy. In that way, we will, through the “real value of numbers”, be able to generate the optimal reachable “budget”.”
(“The Real Value of Numbers” by Mikael Trolle CEO & Partner at Dreams and Details.)
I wholeheartedly agree that this assessment raises, among others, following questions:
- What are the better real numbers?
- Can we practically calculate these values?
- Can they be reliably predicted?
- How often should we determine these real numbers?
This paper is meant to contribute to the crucial discussion of the topic and not as a final destination. More interestingly, the addressed ideas have already been tried out at several companies over the last couple of years, which creates material for evaluation.
What are the “better” numbers?
First of all, management numbers must reflect the truth – the reality of the business – or they will be ignored. This is easier said than done because human beings tend to believe more in numbers than texts. Hence profit and loss statements, as well as balance sheets, are taken for granted though everybody knows that they are only models. And models are by nature only approximations. Just think of accounting practices like wrong debt reservations (revenue recognition), deferred earnings, write-offs, accrued revenues, management adjustments, and finally, allocations which in essence are all arbitrary. For statutory reporting requirements, this is good enough. But for managing the enterprise performance in the days of lot-size 1 “more correct,” details are required.
Secondly, maybe even more importantly, management numbers must be intuitively understandable. Because otherwise they will too, be ignored. As a matter of fact, finance and operations people still don’t understand each other, because their performance measurement systems do not relate to one another. Operations people think in “on-time-delivery” and “just-in-time-sourcing”. Finance people think in costs and revenue or profit. Instinctively people believe that these measures are related to each other, which they mostly are. Till recently nobody could calculate these relationships in euros and cents. As a consequence, nobody was able to determine the relevancy and priority order of typical key performance indicators.
For example, are you sure that improving your customers’ satisfaction actually makes a difference?
So, how about using “cash” for managing your performance? Cash is for sure the least falsifiable number (“cash is a fact; profit is an opinion” (Rappaport 1999, p.15)). Unfortunately, cash has also many meanings. For example, a treasurer looks at cash differently. So why not reverting to “working capital” or even better “net working capital (NWC)”:
Working Capital (NWC)
REVIEWED BY WILL KENTON Updated Sep 19, 2019
What Is Working Capital?
Working capital, also known as net-working capital (NWC), is the difference between a company’s current assets, such as cash, accounts receivable (customers’ unpaid bills) and inventories of raw materials and finished goods, and its current liabilities, such as accounts payable. Net operating working capital is a measure of a company’s liquidity and refers to the difference between existing operating assets and current operating liabilities. In many cases, these calculations are the same and are derived from company cash plus accounts receivable plus inventories, less accounts payable and less accrued expenses.
Working capital is a measure of a company’s liquidity, operational efficiency and its short-term financial health. If a company has substantial positive working capital, then it should have the potential to invest and grow. If a company’s current assets do not exceed its current liabilities, it may have trouble growing or paying back creditors, and finally, go bankrupt.
- A company has negative working capital if the ratio of current assets to liabilities is less than one.
- Positive working capital indicates that a company can fund its current operations and invest in future activities and growth.
- High working capital isn’t always a good thing. It might indicate that the business has too much inventory or is not investing its excess cash.
That said we can postulate that the change in free working capital is an indicator of performance. So, growth in free working capital equals better performance, whereas decline suggests worse performance.
Can we practically calculate these “better” numbers?
Currently, working capital numbers are relegated to the balance sheet. And are not reflected in the profit and loss statement, which is the standard tool for any manager. Furthermore, working capital numbers are typically only available on a high aggregate level. For being useful as a management tool, it needs to be available on the detailed level of work activities performed daily.
A while ago, people conceived the so-called Activity-based costing (ABC) costing method.
Activity-based costing (ABC) is a costing method that identifies activities in an organization and assigns the cost of each activity to all products and services according to the actual consumption by each. …
The aim was to be precise when it comes to efficiency or effectiveness calculations. Following strong initial uptake, ABC lost ground in the 1990s compared to alternative metrics. An independent 2008 report concluded that manually driven ABC was an inefficient use of resources.
Based on this experience, working capital numbers at utmost detail must be calculated automatically. Luckily this is possible now with big data and in-memory columnar databases. Or at least statically; i.e. one can calculate the tied up working capital at any granularity (admittedly with a little trick).
Can the real numbers for value reliably be predicted?
Because we want to plan realistically, in this context, we need to understand the dynamics between operational activities and movement of tied-up capital.; i.e. stretch goals but not goals which are for sure out of reach. Fortunately, this is possible nowadays too. Assuming you have an ERP-system which has run already for several years this system will have a wealth of data which can be analyzed with advanced statistical methods (“machine learning” and “artificial intelligence”). The result is the functional relationship between operational activities (input factors) and financial outcomes (output factors); i.e. you can determine the f in y=f(x) where x=operations and y=finance. And as it is natural with statistics, the f is defined with a certain confidence level (only), which means that you can indeed reliably predict the working capital changes caused by operational changes.
For the first time, finance and operational people have a common language for planning and controlling purposes; e.g. you can tell whether and how your cash position changes if you become more punctual with your deliveries.
There was a similar problem in the communication between doctors and patients. Doctors wanted to convince patients of a specific diet and patients wanted to understand the rationale for these recommendations. This offered ground for the introduction of the Body-Mass-Index / BMI. Meanwhile, BMI is so widely accepted that nobody thinks about it as a common language between doctors and patients anymore. Maybe we can establish such commonly accepted performance indicator for business performance too!
How often do we need to determine these “better” numbers?
You have to calculate hundreds of thousands of specific numbers to assess relevancy and feasibility of performance improvement measures. Even worse, human beings would have to look at these numbers continuously to control their progress.
Obviously, this is not practical at all. Hence you want to strive for automating the performance management controlling as much as possible. Today machines are indeed capable of doing this.
But maybe this is a new chapter in our discourse.
Did you notice that this approach is independent of the actual org hierarchy? This means you could effortlessly perform what-if analyses regarding your company structure as well.
Aggregations makes assumptions about your business which may be true or not. And because this approach doesn’t need any aggregates, it is in its nature unbiased.
Observed practical complications with this approach has more to do with the fact that statistics introduce a certain degree of uncertainty, whereas classical controlling relies on basic arithmetic’s only