The limitations of typical balanced scorecards
A short while ago, I had a conversation with the Chief Administrative Officer (CAO) of a company in the natural resources space who corroborated my thinking. They had adopted a balanced scorecard approach only to find that a) they could never meet all their targets or b) they would totally outperform on some and underperform on others. Monday morning quarterbacks were able to explain the reasons for deviations, but they weren’t able to translate those explanations into more accurate predictions going forward, let alone properly anticipate impacts from business changes. The CAO said: “every year, we predicted that one of our divisions will do better; however, even though the revenue would come in, the division would still deliver lower margins.”
The company could predict the behavior of some metrics reasonably well like revenue and total volume by division, product and region. They could even predict what their inventories would be. But they couldn’t translate these predictions into detailed activities, costs and profit. In other words, their balanced scorecard was infeasible.
But why? To answer this, let’s consider how the balanced scorecard is developed and the technologies that power it.
The (lacking) Three-step Process of Creating Scorecards
Step 1. Gather historical data
First, business intelligence analysts gather historical information on business performance like volumes, average selling prices, customer metrics/trends (like market size and share), cost of goods sold, sales and marketing costs, etc. The more aspirational BI analysts add sustainability and human rights into the mix by looking into their impact on their environment and tracking their suppliers’ behavior.
Step 2. Create a forward-looking picture
Second, these metrics are translated into a forward-looking picture of the business. At this point, business analysts that are usually in financial planning & analysis (FP&A) crank up their Excel spreadsheets to complement their BI tool to produce a more integrated, forward-looking picture of their business. For example, they may translate market growth and market share into forward-looking volumes by assuming a market share growth (usually positive) for their key services and product lines. Next, they make assumptions about the average selling price and convert costs into average costs per unit of service or product. They can then forecast the performance of the business. Sustainability and other aspirational targets are typically tracked, but nobody really knows the impact it has on the business holistically.
Step 3. Compile and present the data
Third, everything is compiled together and presented to the management team for review. With management feedback, assumptions are revised and the spreadsheets are updated. “We should be able to increase efficiency here;” “Our market share and prices will grow faster in this business;” “The price of labor will stay constant instead of growing;” “The price of oil will decline;” and so on. With sufficient horse trading, everyone eventually agrees on the new set of targets.
The result is something everyone commits to, but it either undershoots the real potential of the business or it’s not physically and mathematically possible to achieve. That’s less than ideal, right? Here’s why this tends to happen.
Why is Your Balanced Scorecard Lacking?
BI tools don’t holistically represent your business
The BI tool and Excel sheets do not represent the business in the way it actually runs and are therefore easily subjected to manipulation of assumptions with uncertain outcomes. Inter-relationships between demand, resources and activities are usually assumed at the volume and financial levels. They’re not, however, assumed in terms of materials, routings, BOMs, marketing spend, new products, etc, despite the fact that those are the real drivers of cost and profitability
Business contraints are being ignored … And your CEO has no idea
Business constraints are mostly ignored or, at best, are marginally considered. Organizations have all sorts of constraints like regulations, managerial policies, bank covenants, price elasticities, budgets and physical factors (like throughput and yields from call centers, factories or an operating room in a hospital). Without explicitly representing constraints, the calculations that inform a balanced scorecard may not add up. The biggest problem is usually not this, but the fact that management often doesn’t know this.
Your scorecard isn’t optimized
Balanced scorecards are not optimized. Most spreadsheets and BI tools that inform balanced scorecard decisions are based on hypotheses — i.e. “I have an idea and I use the tools to calculate the impact.” Yikes. As more people add their ideas, it becomes a blend and assumptions begin to overlap and impact each other. On the contrary, optimization searches the optimal across all potential ideas.
The problem in business is not optimizing each function; the problem is how to optimize the whole of the parts.
For example, how does management in a large CPG company decide if it’s better to invest in toothbrushes in Indonesia versus ice cream in Brazil? Conversely, how does a hospital CEO decide whether to invest in ambulatory centers and outreach versus additional doctors and robotics for the inpatient department? One might think these are too low level for a CEO, but they’re the drivers of the balanced scorecard. Therefore, they define the targets for the CEO.
But Don’t Ditch Your Balanced Scorecard
The balanced scorecard is a great concept, and those of you who have adopted it need not despair. What you need is simple: the ability to represent the underlying drivers of your business and the ability to link them together in order find the best path, all while recognizing the operational realities of business. With a little modification, your scorecard will not only have a greater impact on your business, it will drive wider adoption within it. Plus, you’ll be prepared when management adds the next layer of complication, like mandating reduction of carbon emissions to 2010 levels!
