Strategic Planning: "Hey, AI, what's your opinion?"

Thursday, 16. May 2019

How Artificial Intelligence and Advanced Statistics Are Changing Strategic Planning
Strategic Insights: Part II

A guest article by Ronald Herse & Florian Dreher

In part I of the publication series Strategic Insights, the potential benefits of extended statistics and machine learning for strategy and planning work were outlined. As a more in-depth introduction to the topic, Part II discusses ways of increasing the quality and reliability of forecasts in planning.

Every year, we provide forecasts on market volume, sales and earnings for our applications, products, customers and regions as part of complex planning rounds. In our volatile businesses, we often have the feeling that a glass sphere could serve us as well as our attempts at forecasting. The process itself is perceived by many managers as costly with too little benefit for their own area of responsibility. Too many different assumptions of many involved actors and the high complexity of our planning structures are a challenge. The meaningfulness and usefulness of the forecasting work should not, however, be called into question despite all the difficulties:

  • 1. The planning of market volumes, sales and earnings over the coming years forces us to think through our business consistently.
  • 2. It is an excellent tool for the advance coordination of all activities of a division and is an important source for personnel decisions.
  • 3. Advance planning is still one of the central tools for effective communication, target agreement and target control.
  • 4. Our forecasts are the most relevant input for medium-term planning and budgeting. This makes them the "indirect drivers" for the activities that are actually being tackled in the financial year.

Good planning work creates a robust basis for the pre-steering of our businesses and creates added value for all the players involved – especially in times of high complexity and dynamism.

At What Points Can We Make Concrete Use of Machine Learning?

Advanced statistics and machine learning paired with the right technology enable us to make a practicable entry into two subject areas with high potential benefits for our business.

  • 1. Critical reflection of our planning quality and reliability:
    From different perspectives, examine where we systematically over- or underestimate ourselves in the planning process, considering the respective market volatility and time.
  • 2. Active use of the existing data history for machine-supported forecasts:
    Discover patterns from past time series and use machine-generated recommendations as a " plausibility check".

How Do I Find the Needle in the Haystack?

If numerous planning units provide planning figures for volume, sales and earnings over several years at least annually for all their regions, application fields and product groups, then the existing time series and data points can quickly reach thousands. Who can look at all this in detail?

We need to find a way to automatically sift through all the data available, taking time into account. Each forecasting unit (e. g. sales development of a product group in a region) is examined with regard to its relative importance for the business, its volatility and its forecasting error.

If the condensed results from this automated filtering process are transferred into a forecast quality portfolio, it becomes clear where the relevant candidates for the business can be found for further investigation:

Forecast Error
Illustration: Schematic illustration of a forecast quality portfolio

  • „Why so good?”
    There we find planning units whose development we predict relatively accurately despite high volatility. This is a source of best practices and learning impulses for others.
  • „Why so wrong?“
    There we find planning units where we find it difficult to form reliable forecasts. And this despite the fact that the volatility of the respective unit is relatively low. This is where there are potentially effective levers for more planning security.
  • „Hot Potatoes“
    As expected, units that are located at the top right of the portfolio are more difficult to plan and require special attention.
  • „Safe Bet“
    Units on the bottom left seem to have a relatively low planning risk.

This relevance detection considers three variables over time:

  • 1. Relative significance of the planning value
    Each value type (volume, sales, earnings, ...) is standardized and relatively weighted.
  • 2. Relative forecast error
    Relative percentage deviation from forecast to actual.
  • 3. Relative trend-adjusted volatility
    Standard deviation of the difference between a time series and its trend (linear curve fitting) divided by the mean value of the time series.

By using quite simple statistical methods, we are in a position to check the plausibility of developments in our businesses, to discover overarching patterns in a large number of planning units and to sustainably improve the quality of our forecasting work. And all this without having to spend days searching in haystacks.

Use the Data History for Automatic Forecast Suggestions

Use the Data HistoryDuring the planning process we naturally focus on the future, historical developments play a subordinate role. What would happen if we had a digital assistant at our disposal that knew and always kept available all past figures, hundreds of time series, correlations and correlations?

This is precisely where the core competence of machine learning models lies. To recognize statistical correlations from a vast amount of data and to derive forecasts from them. Such models learn from our past and recommend probable scenarios. The machine becomes the perfect complement to the manager with his or her experience and fine feeling for the business. The greatest weakness of the model is at the same time the greatest strength of the human being and vice versa.

So-called predictive modeling helps us at different points:


  • An early plausibility check of manual forecasts becomes possible at all levels and for all planning units.
  • Early warning of potential non-achievement of planned values and planning risks happens in real time.
  • Alternative prognosis suggestions in the sense of a mechanical second opinion increase planning security.
  • Possible unexpected business developments can be identified at an early stage irrespective of human forecasts.
  • During the forecasting process, the manager is actively supported by a digital assistant who knows all the data by heart, takes the planner's view and offers sparring.

As Long as the Tail Doesn't Wag the Dog ...

Human-Machine-InteractionThe importance of forecasting work is too high to allow it to be completely left out of hand and for the machine to decide for itself. Ideally, humans interact with the algorithm (human in the loop). This includes interpreting the simulation results, deciding which values to include in the planning and which impulses to return to the machine in order to rotate learning loops and optimize the algorithm.

Assessing market and business developments has nothing to do with analytical-deductive science. It has to do with finding a balance between today and tomorrow, the inner and outer world of an organization. This requires experience, intuition, creativity and perception. Everything that distinguishes the human leader and no machine can replace. These intelligent, digital new possibilities will soon find their way into today's organizations. We should only make sure that the tail does not wag the dog at the end.

With a little practice and used correctly, we will make our planning work more reliable, efficient and stress-free for ourselves. We no longer speak of abstract ideas. In a current project, we have succeeded in beating the accuracy of human forecasts by a factor of 2 to 3 with mechanically generated forecasts of sales.

In the third part of the the publication series Strategic Insights we go deeper into the methodology and systematics of machine learning technologies and predictive modeling.

Credits (from top):
© Ronald Herse/Florian Dreher (own visualization)
© | Sasho Bogoev
© | Andrey Popov
© Ronald Herse
© Florian Dreher

About the Authors

Mag. (FH) Ronald HerseAs a partner at EVOLUTIONIZER, Ronald Herse is responsible for the development and integration of business content into software for strategy development and implementation.

He was managing director and partner of a leading software company for nine years. As a pioneer in the combination of integrated management concepts and modern software for anchoring excellent management in complex organizations, he implemented concrete and sustainable solutions in the entrepreneurial practice from very early on. 

Previously, he worked in more than 80 international projects for six years as a consultant and spiritus rector of the Competence Center for Strategy at the Malik Management Center St. Gallen, Switzerland.

Already at the beginning of the digital era, he founded a digital marketing agency and still teaches at the University of Applied Science in Vorarlberg, Austria, on strategic management, transformation and change processes. You can reach him via

Florian DreherAs Senior Data Scientist at EVOLUTIONIZER, Florian Dreher is responsible for everything from simple statistics to artificial intelligence.

He uses machine learning methods for topics such as predictive analytics, data mining and text mining to make EVOLUTIONIZER products even more intelligent.

Florian Dreher worked for an automotive supplier for four years. There he used machine learning to automatically forecast complex product characteristics in order to optimize product development.

As a trained engineer in mechanical engineering, he built up his data scientist skills in a joint project with the Fraunhofer Institute and from then on continuously expanded them on the job. You can reach him via