Benchmarking Your Forecasting: When Balance Is Bad.

Why biased decision-making and driver-based forecasting are good - and the balanced scorecard may not be

It's ironic that in a world of abundant information, planning processes are more cumbersome today than when the decade started. According to research from Hackett Benchmarking solutions, the average company, regardless of industry, takes nine months to complete its annual budget.

The balanced scorecard is often at the root of the problem. How can you make the best decisions for your area of responsibility when you give equal weight to goals/missions, objectives, profit/return to shareholders, quality, customer service, innovation and corporate citizenship? To suggest the decisions can somehow consider all these values equally and still arrive at an optimal course of action for a defined area is equivalent to "management communism."

Adding to the scorecard confusion is the fact that managers at most of these companies are forced to make forecasts and budgets with this ineffective tool. It's ironic, again, that budgeting at most companies has yet to be re-engineered when so much depends on sound forecasting. When done well, production lines are fully staffed, the necessary amounts of raw materials are procured and product is available to customers when and where they need it.

Poor forecasting, on the other hand, can ruin a company. If one underestimates demand, production lines are understaffed, forcing labor to work overtime at a wage premium. The scarcity of raw materials on the shop floor forces buyers to accept higher prices when rushing late to market. And these added costs strangle cashflows.

The overestimation of demand has just as many penalties. Massive amounts of data have not improved forecasting. In some cases, the deluge has blinded managers because they can't decipher what is important.

What can a manager do? Rather than try to understand all the information that is available, you can create a predictive model that relies on only a few financial and operative factors to generate a reliable forecast. By limiting the criteria and capitalizing on the speed of measurement offered by technology, you can make better forecasts faster.

Driver-based forecasting does exactly that. It determines profitability by pinpointing those factors that drive revenue, cost and profit. Drivers are internal and external factors, like the effect seasonality can have on the prices of raw materials or the impact of a strong economy on the availability of labor.

To be a driver, a predictor must have materiality and volatility. "Materiality" is defined as any factor that has a significant impact on components like revenue and costs. It is determined by comparing the relative proportion of a financial line item to the organization's total costs. "Volatility" is how much the factor varies from period to period. One must understand the relationship between these two characteristics to determine the variable's impact on the desired state and how often it should be monitored.

A predictive model can be created by evaluating each cause-and-effect relationship that precedes the desired end-state. Is the relationship material? Is it volatile? If the answers are "yes" to both, then some measurement of the activity needs to show up in a timely report to management. This process eliminates from the report any surface or pseudo-drivers that do little to explain the actions of more senior drivers in the hierarchy.

Keep Your Eyes on the Road

Increasingly, best-practices companies are presenting data-customized reports called "dashboards." Just like a car's dashboard doesn't tell the motorist everything about the engine, there is no benefit in overwhelming managers with useless information. Otherwise, one risks the manager taking his or her eyes off the proverbial road - and missing the larger fiscal picture.

A manager's dashboard is a visual interface that repeatedly measures what drives a business. There are some data that all motorists need to know - speed, temperature and fuel level. Likewise, there are specific data that are universally relevant to all managers in an organization, like revenue, margin, total expenses and return on assets. The objective of the reporting tool is to keep managers focused on what's important, so that they make better decisions faster.

Best-practices companies typically achieve better decisions faster by incorporating three levels of driver analysis (read about GM's successful application on p. 46). First, they use operational driver-based analysis to create forecasts that provide direction by correlating data with a manager's area of responsibility. By employing the dashboard method, they maintain or increase forecasting accuracy while delivering reports faster and with fewer staff resources. The approach improves the effectiveness of financial planning by taking a bottom-up view to all aspects of an operation.

Second, they employ strategic driver analysis to measure goal attainment. Goals determine short- and long-term strategies, so a strategic driver is any variable that affects company value or critical success factors essential to meeting goals during respective time periods.

Third, they use value-based driver analysis. A value driver is any variable that affects the value of a company. These drivers are defined at a level of detail consistent with the decision variables directly under the control line of senior management. This top-to-bottom data give decision-makers the information to make daily operating choices that improve performance and potentially impact shareholder value.

Managers need the same ready access to driver analysis that a car's dashboard offers a motorist. Likewise, if trouble is looming, managers need to quickly drill down or uncover a problem area. Best-of-breed technology offers these critical capabilities and more. Technology improves data integrity through common language and interface. It improves report quality through increased control. It quickens response time to ad hoc queries. It improves investigation capabilities through better knowledge management and faster data access. And last - and this is critical - it establishes a platform for additional applications.

Information is power. Its timely delivery can make good managers better, more confident decision-makers. However, the metrics conveying this information must be valid. Thus, time needs to be spent in data determination and gathering when evaluating driver-tree relationships for the predictive model. Validating drivers and pseudo-drivers requires digging through numerous financial reports, special financial studies and management reports and scorecards. Outside expertise is needed to evaluate external variables objectively, as well as to offer qualitative opinions regarding magnitude, importance and reliability of data. Driver-based forecasting is only as good as the data being used.

Predict, Implement, Motivate

Once the predictive tool is created and the right staff mix is in place to interpret the information, a corporate-wide process must be institutionalized to permit timely delivery, ready access and appropriate responses. Such a system must provide managers with the tools they need to interpret business-related data, provide it when it's needed and, importantly, motivate managers to act accordingly.

Without processes supporting good people and best-of-breed technology, driver-based forecasting will fail to deliver faster budget cycles and better cash flow. The reality is that a measurement does not have to be perfectly balanced, but it must be biased. It should dictate the company's strategic objectives, be tailored to a manager's area of responsibility and address what differentiates your product and services. How it is biased will depend on the priorities of a particular industry or company. For instance, 3M may be biased toward innovation, Motorola toward quality and Ritz-Carlton toward customer service.

By focusing on what is important and moving to simpler tools of information, companies can improve the level of understanding managers have for the strategic goals of the organization. Their ability to understand the vision of senior managers, coupled with timely and readily available information, will result in better and faster decision-making. The impact of improved decision-making will be revealed when quantitative measures are overlaid on goals and critical success factors, applying a yardstick to what was once considered esoteric.

According to the latest Hackett research, 42 percent of companies fail to share strategic plans with managers. A driver-based forecasting system cannot bridge this gap alone. Yet, the right compensation can complete the link.

Less than 60 percent of companies tie incentives to the satisfaction of strategic plans, while 97 percent tie compensation to financial plan results, as reported in Hackett research. Systems like this risk motivating managers to focus on short-term finances at the expense of long-term strategy.

Best-practices companies, on the other hand, are defining performance measures that are tied to strategy, then building tactical and financial plans according to these priorities. They are finding that this structure aligns management motivation with strategies and reduces planning cycle times.

For top-performing companies, the marriage of a driver-based forecasting system and appropriately aligned incentives is improving the budget planning cycle dramatically. The budget of the average company, according to Hackett research, contains 230 line items. Best-practice companies, on the other hand, are keeping budgets to as few as 40 line items or better. And some have actually discontinued the annual budget in favor of a rolling forecast.

Budgets with numerous line items practically ensure numerous variances. This results in highly skilled professionals allocating too much of their time on the lower value-added activity of analyzing the differences. The average company, according to Hackett research, invests more than 25,000 person days annually in strategic, tactical and financial planning, reporting and forecasting for every billion dollars of revenue, compared to 6,000 at top-performing companies. One cannot ignore how much valuable time and energy are being lost to low value-added activity, rather than invested in thinking about ways to grow a business.

Regardless of their goals, best-practices companies create systems to deliver timely information and the tools needed to interpret data as it relates to business strategy by incorporating the following seven critical success factors.

Priorities: The dashboard should be limited to data from inside and outside the firm that measure how well a CEO's vision is being achieved. By including only those metrics that intrinsically drive business, managers expend energy and effort on that which is strategically important.

Technology: Investment in e-mail or Internet software enables the timely delivery of and ongoing access to concise, focused information that improves the speed and quality of day-to-day decision-making.

Incentives: Motivate managers to act as needed by tying individual incentives to specific financial results. This helps give top-down direction to managers, so they understand where within the business they need to focus staff energy and attention.

Accountability: Customize reports to what is relevant to users, matching content with accountability and responsibility. Otherwise, you risk confusing them with information on areas they have no control over.

Recognition: Validating those who make better decisions faster helps define desired behavior for others.

Communication: Don't just show numbers. Explain to managers how data relate to the achievement of goals, reinforcing corporate objectives and highlighting gaps in performance.

Credibility: Substantiate reports by providing the same ready access to information deeper down the chain. If a red light flashes on the dashboard, a manager should be able to pop the hood and quickly uncover the source.

For driver-based forecasting to succeed, a lot of up-front tactical thinking must be done. The manager's tool set of people, process and technology is just a commodity without a vision behind it. How do top-performing companies make this set a value-adding force that differentiates them from competitors? They start by recognizing that information has power. They acknowledge it has the power to destroy a company, like when knowledge workers are blinded by too much detail to understand what the best decisions are. And, better yet, they understand it has the power to focus a manager's attention to that which is critically important to the value of the organization.

The true value differentiator for any business is its people and the quality of the decisions they make. Driver-based forecasting asks managers to consider only that which is important to the areas of performance they can affect. Technology delivers the tool to them, improving knowledge worker efficiency and effectiveness, and enhancing a company's competitive strength and bottom line.

Louis Eyermann is director of CEO|solutions at AnswerThink Consulting Group, based in Miami. You can reach him at [phone]. All statistics are from the ongoing benchmark sponsored by FEI and Hackett Benchmarking|solutions. They are property of Hackett Benchmarking|solutions, 1999.