48th Congress of Controllers: Expert contributions from the Exhibitors 2024

The first business intelligence application in 48 hours: zwissTEX, the Swabian manufacturer of high-tech textiles, has done it. With the help of Bissantz ERP solutions for SAP data, a comprehensive KPI landscape for controlling corporate finances was created - and integrated planning.

zwissTEX was looking for a software solution that transfers the data from SAP modules such as SD, FI, MM and PP into efficient reporting. The solution should map the financial and business goals with meaningful KPIs quickly and with little effort and measure the achievement of goals.

Trusting in Bissantz's performance promise, CEO and CFO Dr. Ulrike Neubauer made the project a top priority: “SAP landscapes can be cumbersome. I was skeptical as to whether something could really be added on top of that so quickly – but Bissantz kept the promise.” The linchpin of the solution is the Bissantz ERP Solutions, a tool for the automated creation of ready-to-use, packaged business intelligence systems. Bissantz delivered the first reporting application in just two days.

Soon analytical applications for the areas of logistics, sales and finance were created: from stock levels and ranges to product groups and bills of materials to contribution margins and profitability calculations. After only six months, the desired KPI landscape had taken on a concrete form - and can also be used on the smartphone with the Bissantz DeltaApp.

One of the first applications: the receivables monitor on the smartphone, with interactive access to perspectives such as debtor, product line, due date/dunning level and even the invoice items.

Parallel to the development of the KPI landscape, Bissantz zwissTEX supported the P&L planning: In the sense of an integrated planning, the cost planning in the nature of expense method was merged with the sales planning on sales objects. The specialist departments plan their cost centers and cost types in SAP, while Sales plans sales and turnover in Bissantz. The sales plan data for customers, articles or sales areas are in turn supplemented with the article calculation from SAP. The integrated planning profit and loss statement in Bissantz results in concrete targets for corporate management - the loop of planning, analysis and reporting is closed.

"From many years of experience, I am well aware of the possibilities and limitations of software in corporate management," explains Neubauer. "I was really surprised that such increases in information quality are possible in such a short time!"

READ THE COMPLETE ARTICLE HERE (PDF DOWNLOAD, GERMAN)

About Bissantz & Company

Bissantz & Company sees itself as a think tank for the application of artificial intelligence (AI) in management. The company was founded in 1996 and is still owner-managed and self-financed today - a true family business in the German-speaking business intelligence market. With our own basic research, the software is further developed in a scientifically sound and interdisciplinary manner. We combine insights from business administration, computer science, neurobiology, psychology, statistics and artificial intelligence. Our innovations are patented and have won numerous awards, including the Innovation Prize from the Gesellschaft für Informatik (GI) and the UX Design Award from the International Design Center Berlin (IDZ).

To the company page

"It was important to us during the course of the project that the end user received the automation and logic that they needed so that they could carry out their planning as conveniently as possible"         
Merle Schmidt-Brunn, 
Senior Director Financial Analytics
Hapag-Lloyd AG

The company Hapag-Lloyd AG

Hapag-Lloyd is one of the largest liner shipping companies in the world with an EBIT of EUR 1.315 billion (2020). The company has a clear claim: “Number one for quality”. This claim is also reflected in the effort to accurately plan global business. Hapag-Lloyd is one of the leading providers in the Transatlantic, Middle East, Latin America and Intra-America trade areas and is characterized by one of the world's largest and most modern refrigerated container fleets. Using driver-based, integrated and globally distributed corporate planning, Hapag-Lloyd wants to adhere to the strategic goal of high reliability and quality and continuously develop further.

KEY FACTS
Industry: Transport & Logistics
Locations: 418 offices in 137 countries with headquarters in Hamburg
Employees: Around 13,900 worldwide
Software: Board
Advantages & benefits: Consistent solution in the company | Standardization of processes | Greater data transparency | Better management over data | A single environment for planning | Better and more detailed operational planning | Greater user-friendliness.

The challenges of the project

The challenge of integrated corporate planning at Hapag-Lloyd still lies in the closely interlinked partial planning within the group. Starting from the allocation of ship capacities to the planning of global sales volumes and the associated direct costs of the containers through to the general costs, the individual planning steps are very coordinated and interlinked. In the past, the existing solution in Excel and other tools led to a lot of manual and time-consuming effort and a high susceptibility to errors. Changes to assumptions and adjustments to planning could only be made at aggregate levels. A particular challenge in the project are the direct container-related costs (Cargo Related Cost, CRC), especially for the terminal and transport costs. In this planning, the costs are planned for each individual data point of a trip. The starting point is the upstream volume planning with a transport volume of 12.0 million TEU (standard containers) per year as well as 121 liner services that ensure fast and reliable connections between more than 600 ports on all continents worldwide. This partial planning generates a large amount of data points with high complexity due to the large combination logic and was therefore individually planned locally and regionally using standard costs in the past. However, with the previous approach, it was not possible to achieve better transparency in the creation and management of costs worldwide, as the costs incurred could not be directly assigned to the volumes and regional cost drivers.

"In the end, our goal is to develop a process in which all partial planning is mapped in the same way as it is carried out by the planner themselves. The tool should then take over the transition."
Merle Schmidt-Brunn
Senior Director Financial Analytics
Hapag-Lloyd AG

The implementation of the project

Hapag-Lloyd's objective was to set up an efficient solution for corporate planning that, in particular, integrates operational partial planning and accelerates the planning process. Employees should also be relieved and automation and flexibility should be increased. It was particularly important to Hapag-Lloyd that the new solution was extremely user-friendly and that the design of the planning masks could be designed in the look and feel of Hapag-Lloyd in order to achieve the highest possible buy-in from employees.

After reviewing and selecting providers, including carrying out a proof of concept, Hapag-Lloyd ultimately decided on the Board planning platform and celver as a partner for implementing the solution. “Our focus was on operational planning. Most tools on the market, however, only focus on planning financial packages. Our choice of Board and the experts from celver as implementation partners was not least because of the good cooperation in analytical projects with Qlik,” said Merle Schmidt-Brunn.

In the following weeks, the planning process was analyzed in workshops and the future planning process was defined. Hapag-Lloyd's solution now consists of holistic capacity, sales and cost planning with around 450 users. The annual budget planning, which results in a strategic five-year plan, serves as the basis for further updates as part of the forecasts carried out three to four times per year. The individual planning steps are connected to each other through a workflow and linked to each other through partial plan integration. Through this integration and the underlying calculation logic as well as the integration of historical data, planners can usually use suggested values as part of their plan inputs. An important part of the project was data connection and preparation. Here, various preliminary systems with different levels of detail had to first be harmonized and merged in order to ultimately obtain a detailed planning solution including the required actual data. During the course of the project it became clear once again that an agile approach, retrospectives and good collaboration bring significant advantages on complex topics.

The new planning of Cargo Related Costs (CRC), which occupies a special place in integrated corporate planning, uses the worldwide planned container volumes and their distribution on the routes as input. The modeling of this network planning enables Hapag-Lloyd to adjust and simulate costs using a wide variety of parameters and thus to determine global or local effects on the routes and in the ports.

The use

The goal of standardizing the processes and creating a “one-system-for-planning” environment was achieved in the project. This ensured global data transparency (single point of truth) and management over the data at all times. The project has already come closer to the goal of reducing global planning by two weeks, but here too, further expansion stages and refinements in the planning process are necessary in order to achieve “pace and perfection”. The new planning environment, with its focus on operational planning, also has a direct impact on end users beyond the actual planning. Since the operational logic and, above all, the detailing of the data is close to the daily business of the planners, the information is also used for operational management using plan/actual comparisons.

In an increasingly volatile and unpredictable business world, companies are faced with the challenge of being able to react quickly and flexibly to changes. The key lies in the ability to obtain precise and timely information that enables informed decisions. Ad hoc reporting is an indispensable controlling tool. It enables the creation of tailored queries for specific questions and is particularly valuable for effectively responding to short-term requests from management, stakeholders or regulatory bodies.

BASICS OF AD HOC REPORTING

Ad hoc reporting plays a central role in modern controlling. It enables companies to react flexibly and quickly to individual information needs. Unlike standardized reports, ad hoc reporting provides tailored analysis for unforeseen or specific requests and supports data-driven decision making.

CHALLENGES IN AD HOC REPORTING

However, conducting ad hoc reporting can also pose challenges:

• Time-intensive data collection: Compiling and analyzing data manually can take significant time.
• High demands on data quality and integration: Ensuring high data quality and the efficient integration of data are essential.
• Complexity of data analysis: The need to effectively analyze and interpret complex data requires specialized skills and tools.

EXAMPLE: Analyzing the impact of raw material price fluctuations and supply chain bottlenecks on production costs

Initial situation

A manufacturing company faces two parallel challenges: significant fluctuations in the prices of key raw materials and supply chain bottlenecks that affect the availability of those raw materials. This situation threatens production stability and could lead to significant cost increases. In order to respond adequately, management needs a quick and precise analysis of the current situation and the potential impact on production costs.

Challenge
The required data is scattered across various internal departments and external suppliers. In addition, current market data on raw material prices must be taken into account. Manual collection and consolidation of this data is required to perform a comprehensive analysis of the impact on production costs.

Solution

1. Data collection: colea is used to provide a central platform for collecting relevant data. This includes current and forecast raw material prices, information about existing and potential supply chain shortages, and details about alternative suppliers and raw materials.
2. Structured data entry: Appropriate input masks are created and made available to the relevant internal and external stakeholders to ensure standardized data collection. This structure makes it easier to capture complex and diverse information.
3. Preparation for external analysis: Although colea itself does not offer specific analysis functionality for this scenario, the platform supports the organization and preparation of the collected data for further analysis. By providing a structured and consolidated database, the company can use external tools such as Excel or specialized analytics software such as Power-BI or Qlik to model and evaluate the impact on production costs.
4. Development of recommendations for action: Based on the analyzes carried out, strategies can be developed to minimize the negative effects on production costs.

Result
By effectively leveraging colea for rapid data collection and consolidation, the company is able to make informed decisions to address the challenges caused by raw material price fluctuations and supply chain shortages. This helps control production costs and maintain production despite external uncertainties.

The Staatliche Porzellan-Manufaktur Meissen GmbH realized a comprehensive digitalization project together with Comarch. Meissen, one of the internationally best-known luxury manufacturers, chose Comarch as its software service provider. Comarch ERP Enterprise and Comarch Financials Enterprise were introduced within eight months. Most recently, Meissen started implementing the document management system Comarch DMS.

The customer

Meissen, the oldest porcelain manufacturer and registered brand in Europe, draws on over 300 years of styles and creates new designs using the finest craftsmanship. Every Meissen creation today becomes, so to speak, a modern interpretation of European art and cultural history. Modern opulence conquers interiors in the zeitgeist. Fine things for the table, masterpieces, figures, pieces of jewelry and home deco porcelain delight people all over the world who appreciate the luxury of sustainable production and exquisite handcraft. Meissen porcelain can be recognized by the crossed swords signet and is a testament to the quality of the manufacture. In 1710, Meissen invented European porcelain. Owned by the Free State of Saxony, it is one hundred percent manufactured in Germany.

Data & Facts

Name   Staatliche Porzellan-Manufaktur Meissen GmbH

Industry  Porcelain, luxury goods industry

Headquarters   Meißen

Project progress

Andrea Zanke, ERP Implementation Project Manager at Staatliche Porzellan-Manufaktur Meissen GmbH, explains how the company switched from a highly customized previous solution to the integrated Comarch ERP Enterprise system. The goal of this extensive project was to stay as close to the standard as possible and thus be able to map the processes surrounding a highly individualized product.

Initial situation and motivation for the ERP project

The starting point for the ERP project at Meissen was one typical of medium-sized companies. There was a relatively heterogeneous IT system landscape that had grown over the years and consisted of a mix of standard and proprietary software. The individual systems required a lot of effort to maintain the interfaces. The business processes sometimes ran beyond the system boundaries, which made it difficult to conduct business efficiently.

Goal

The introduction of the new ERP Systems was about bundling a wide variety of IT systems and mapping the business processes in an integrated IT system with integrated end-to-end processes. The overarching goal was to stay as close to the standard as possible and thus be able to map the processes surrounding a highly individualized product.

“The manufacturer has set out to optimize its processes related to craftsmanship. While the craftsmanship and handwork in the production of porcelain are our unique selling point, we wanted to put the management and administration of our processes on an efficient footing,” says Andrea Zanke, ERP Implementation Project Manager | State Porcelain Manufactory Meissen GmbH.

Results

Greater data transparency in the area of finance & controlling

The finance and controlling area benefits most from the integration of the systems, as this is where most of the data from all areas comes together. The Comarch ERP system proved to be very performant and flexible. The many functions in the area of evaluations are particularly pleasing. The cockpits are very flexible and offer a high degree of data transparency.

Centralization and standardization of procurement processes

From a purchasing perspective, the new ERP system should help centralize and standardize procurement processes. It was also important to map the various business processes (tourism, production, direct & indirect purchasing). Furthermore, Comarch DMS, a new document management system, was introduced, which brings additional advantages:

• automated recording of receipts (automatic reading of PDF invoices and offers)
• an automatic certified filing
• The very good workflow system leads to long-term relief from additional work and avoidance of errors

About Comarch Financials Enterprise

The modern, internet-based accounting solution Comarch Financials Enterprise is used in many companies in Germany, Austria and Switzerland as well as internationally. As a stand-alone solution for financial accounting, controlling and asset accounting, Comarch Financials Enterprise forms an integrated platform for a company-wide customized application. As an integral part of Comarch ERP Enterprise, it enables full transparency in all areas of the company and is modularly expandable. The strengths of the accounting software include its ease of use, an open system architecture for data exchange with third-party systems, and many years of experience with a four-digit number of installations of Comarch Financials products for customers such as Victorinox, Weitkowitz and KiK.

To the company page

Authors: Mike Schulze / Felix Broßmann / Bernd Wallraff

The launch of ChatGPT (Generative Pre-trained Transformer) in late 2022 sparked a wave of interest in generative artificial intelligence (AI). It is a chatbot that uses AI to communicate with users via text-based messages. In response to user input, the user is able to generate specific texts. Within days, millions of users around the world experimenting with this newly available technology had discovered and shared countless ways to use it effectively. In the weeks and months that followed, companies also tried to keep up with this immense development. Some companies have already taken a formal approach and formed special teams to investigate how generative AI can be used to benefit business models or companies and improve efficiency in the company.

What is generative AI?

Generative AI refers to machine learning models that can, for example, generate texts, images, audio sequences or synthetic data by imitating human-like language patterns and structures. These models are trained on various internet texts and are incredibly creative in generating stories, poems and subject-related articles. In addition, they can perform various tasks, answer questions, and provide detailed explanations, making them valuable tools for a variety of applications.

There are generative AI tools for different modalities, such as text, images, music, code and voices. Examples of different AI content generators include the following:

• Text generation tools such as ChatGPT, Jasper, AI-Writer, Bard and Lex.
• Image generation tools include Dall-E 3, Midjourney and Stable Diffusion.
• Music creation tools: Amper, Dadabots and MuseNet.
• Code generation tools: CodeStarter, Codex, GitHub Copilot and Tabnine
• Speech synthesis tools include Descript, Listnr and Podcast.ai.

Basic capabilities of generative AI

Generative AI can be used extensively in many areas along the company value chain. It can make existing content easier to interpret and understand and automatically create new content. Some of the fundamental capabilities and resulting benefits of using generative AI are:

• Automating the manual content creation process,
• Reducing the effort required to respond to e-mails,
• Improving the response to specific technical queries,
• Summarizing complex information,
• Simplifying the process of creating content in a specific style.

Opportunities for controlling and finance

Some specific use cases have already been discussed for controlling and finance areas:

Identification and assessment of risks: Generative AI models can process large amounts of text information and analyze and evaluate them in relation to different risk categories. They can answer questions and provide personalized recommendations. This technology allows financial professionals to provide better customer service and streamline communications.

Financial analysis and forecasting: Generative AI models are ideal for analyzing finance-related data such as financial statements, balance sheets, and profit and loss statements. You are able to identify trends, predict results or create detailed financial analysis.

Fraud detection: By analyzing historical transaction data, generative AI models can detect patterns and highlight potential anomalies that indicate fraudulent activity. By using AI-powered fraud detection, companies can improve their security measures and protect their assets.

Extracting data from financial documents: Generative AI can automate the extraction of specific data from various financial documents such as invoices, contracts, and regulatory filings. This automation reduces manual effort, improves accuracy and accelerates processes, which ultimately increases efficiency in the controlling and finance areas.

Creation of financial reports: Generative AI models can automate the creation of financial reports, saving time and effort for controlling. By using natural language features, these models can also produce annotated reports that enable better decision making.

Limits and risks

Although generative AI offers exciting and diverse possibilities, there are also limitations and risks. Data protection and future regulatory requirements regarding the use of AI are particularly important.

Data protection: Ensuring data protection is crucial in the age of regulations such as the EU General Data Protection Regulation (GDPR). Companies must handle customer data responsibly and comply with data protection regulations to avoid legal consequences.

Future regulatory landscape: New regulations relating to AI are currently being coordinated at the European level. For example, the European Union's Artificial Intelligence (AI) Act aims to reduce the risks associated with the use of AI by requiring developers to test their systems, document use and take security measures. The effects of these legislative initiatives cannot yet be conclusively assessed - what is certain is that there will be restrictions in the usability of non-European solutions if they do not meet the then applicable regulatory requirements.

Notes on practical application in a corporate context

Despite the risks outlined, generative AI tools can already be used in company practice in a comparatively legally secure manner, provided that data protection and the protection of business secrets are taken into account.

Use of non-European service providers: The technologies of non-European AI providers such as OpenAI (the company behind ChatGPT) should only be used in a corporate context if it is ensured that the data entered is both data protection compliant and does not represent any business secrets. In this regard, it is not problematic to use it to generate marketing and information materials. In the context of controlling and financial processes, however, it will only be possible to find use cases that meet these requirements to a limited extent. In any case, technical security should be used, for example by checking the input data before sending it from your own IT infrastructure, which is relatively easy to implement.

Use of GDPR-compliant AI models: The German company Aleph Alpha is focused on providing explainable and trustworthy AI models that are GDPR-compliant. The use of these models ensures compliance with the GDPR and offers flexibility in possible uses. In this case too, you should refrain from entering data that represents trade secrets. For such use cases, Aleph Alpha offers the option of integrating the language model completely independently into your company's infrastructure. This ensures that the data does not leave your own servers.

Use of open source models: To protect trade secrets and avoid data leaks, companies can use open source models that can be customized and also installed on the company's own infrastructure. This approach allows companies to maintain control over their protected information while benefiting from generative AI capabilities. For example, Bosch would like to provide its own BoschGPT for its employees by the end of the year. Other companies such as dm, Samsung, McKinsey and SKAD are also working on such solutions. The basis for BoschGPT is Bosch's in-house database. This procedure makes it possible to use current company data for the results created. This means that the system cannot currently be directly connected to solutions such as ChatGPT or Bard, otherwise data could leak. For this reason, the capabilities of these applications, despite immense improvements, do not yet come close to proprietary solutions such as ChatGPT or Bard. However, the rapid development of the past few months suggests that this will be the case in the medium term.

How do you achieve good results with generative AI?

Link statements, questions and instructions: In order to achieve good results, you have to give AI systems like ChatGPT instructions that are as accurate as possible. A crucial aspect that significantly influences the efficiency and accuracy of the results is so-called prompting. A prompt represents an input that is provided to the respective model to produce a response or prediction. This input can take various forms to direct the output in the desired direction. For example, questions can be asked, specific instructions can be given, or a chain of statements, questions or instructions can be strung together.

Examples of prompts

• How many inhabitants does the capital of Denmark have?
• Name the five most commonly practiced sports in Europe.
• Is there a difference between “causality” and “correlation”? If so, please explain it using examples.

An effective prompt should generally be formulated as clearly and precisely as possible. Instead of “What can you do with ChatGPT in controlling?” the query leads to “I am planning to use ChatGPT in the area of analyzing business data for a medium-sized company. What areas of application are there?" leads to significantly more specific results. The structure of the prompt gives the AI models a clear context and direction, which ultimately leads to more focused and relevant answers.

Sample Prompt Construction: Creating an effective prompt is an essential step to getting accurate and nuanced answers from a generative AI. Here is a simple guide that walks you through the process step by step:

1. Facts: You should start with a clear definition of the central topic or question. In this step, it is important to understand and convey the exact context of the question. For example, a request could be: “I would like to develop a concept for optimizing cost center accounting in my company.” This gives the AI a clear starting point that allows it to focus on the specific details that are relevant to optimizing cost center accounting.

2. Framework conditions and restrictions: The next step is to determine the specific framework conditions and restrictions that the AI should take into account when answering the query. This could e.g. include paying particular attention to certain departments or business areas. By precisely defining these parameters, the AI can create a response that is tailored to the exact requirements.

3. Format: Finally, the format in which the answer should be presented is defined. This could be a textual representation that describes the optimized cost center accounting processes in detail. Clearly stating the desired format helps the AI structure the response in a way that is easy to understand and interpret.

Example of a prompt in this simplified design:

• “I would like to develop a concept for optimizing cost center accounting in my company (fact).
• Pay particular attention to the sales and production departments to ensure efficient allocation of costs (framework conditions and restrictions).
•  Present the results in a textual representation that describes the optimized cost center accounting processes in detail (format).”

Advanced Techniques: The basics needed to create effective prompts are relatively easy to learn. At the same time, there are many advanced techniques in the field of prompt engineering that make it possible to further refine and optimize the output of generative language models.

Some of these techniques are based on the idea of "training" the model with different amounts of information or examples before producing an answer. These include approaches such as:

Zero Shot Prompting (standard case): In this case, the model is asked about a topic without prior specific examples or information. Example: “Explain the use of the balanced scorecard in controlling.”

One Shot Prompting: The model receives a single, relevant example before answering a query. Example: “Just as a company uses key performance indicators (KPIs) to measure success, how might an educational institution measure its performance?”

Few Shot Prompting: The model is provided with several relevant examples to prepare it for a specific query. Example: “A company might use a SWOT analysis to evaluate its strategic position. A sales team might use CRM software to manage customer relationships. How could an HR department use big data?”

Another example of advanced approaches is the so-called “chain of thoughts” approach. This is a method in which a language model is encouraged to explain its thinking or solution step by step when solving a task. The idea is that by showing examples that explain the thought process, the model learns to show its own thought process when answering questions. This type of prompting can make the model's answers more accurate because it is forced to think about each step of its response, which can be particularly helpful for complex tasks. This representation also makes the results of the AI more transparent for you.

In addition, a number of advanced prompt techniques such as Tree-of-Thoughts (https://arxiv.org/abs/2305.10601) or Graph-of-Thoughts (https://arxiv.org/abs/2305.16582) are currently being developed, which use different reflection techniques to improve the outputs generated by language models many times over.

Overall, it can be said that choosing the right prompt technique depends heavily on the specific request and the desired result. The only thing that helps here is the continuous practice of different techniques by the users.

Summary and conclusion

Generative artificial intelligence, such as ChatGPT or similar models, has transformative potential for controlling and financial areas in companies. Applications include conversational finance, financial analysis and forecasting, fraud detection, document analysis and financial reporting. However, companies need to keep an eye on data protection, regulatory compliance and changing legal environments. By using solutions like ChatGPT, collaborating with AI providers specializing in data security, using GDPR-compliant AI models, and adapting open source models, companies can maximize the power of generative AI while minimizing risks and comply with regulatory requirements. It will also be crucial for companies to continuously stay up to date with the latest developments in this extremely dynamically developing field and to introduce responsible AI practices within the company in order to be able to use generative AI effectively.

“Carthago”: The name stands throughout Europe for premium motorhomes with an exceptional sense of space. When Karl-Heinz Schuler founded the company Carthago in Ravensburg in 1979 with individual conversions of panel vans according to customer requirements, he had the premium segment in the motorhome market in mind right from the start. Today, Carthago Reisemobilbau GmbH is one of the leading providers in Europe. The group of companies has operated from its headquarters in Aulendorf, Upper Swabia, since 2013.

In “Carthago City”, the owner-managed family business not only operates one of the most modern motorhome production facilities in Europe, including a sales exhibition, but also offers numerous rental vehicles. Around 1,400 people are employed here and at the Odranci production site in Slovenia, generating annual sales of around 390 million euros (2021).

The trend towards “traveling on four wheels in your own living space” was somewhat slowed down by delivery bottlenecks at the Upper Swabian premium manufacturer. Nevertheless, the Carthago Group remains on a growth path, which is underlined by the newly opened second production site in Slovenia.

Satisfy the need to catch up

The focus of Carthago's business activities has been the development and production of premium class motorhomes for over 40 years. Great importance is placed on comfort, safety and longevity, as well as on continually improving and refining the vehicles.

In view of the rapid growth, the Carthago group of companies was faced with changing planning and reporting requirements at the beginning of the 2020s, when the tools previously used - Excel and IBM Cognos Analytics - could no longer adequately reflect. Analyzes should be completed more quickly and reports should be available on a daily basis. “We had a certain amount of catching up to do,” says Christian Staud, Head of Controlling at Carthago. A new solution with BI functionality should help here.

The selection was easy thanks to existing internal system applications and experience. Carthago should make progress very quickly with the advice and technology of the service provider elKomSolutions GmbH from Tuttlingen. “I wasn’t part of the selection process yet, but I can only confirm this,” explains the qualified mathematician who has worked at Carthago since June 2019. At the turn of the year 2018/19, Carthago put the elKom solution into operation.

Prototypes simplify the introduction and operation of the software

“When I came to Carthago, the software was technically implemented but not yet fully rolled out. The roll-out then went very quickly, which, says Staud, was “particularly due to the methodology of elKomSolutions”. Developing prototypes that then serve as a template for operations leads to quick visibility and rapid implementation success. Staud also attributes this to the software, which is “completely uncomplicated to use and runs stably”. That's why there is very little need for support.

elKomBI is now in use based on the OLAP database IBM Planning Analytics TM1. Carthago uses the solution for reporting and planning. Staud emphasizes that a particular advantage is that the controlling department can operate and further develop the solution almost independently of internal IT or external consultants. IBM Cognos TM 1 carries out evaluations of large amounts of data in a very short time and is easy to use.

The BI function supports the analysis of more complex data sets and their condensation for reports. Thanks to the Excel integration, controlling can quickly and easily design reports for the various recipient groups in the company. TM1's technology then enables access via the web, so that report recipients can log in via the Internet and access their reports on a daily basis.

One keystroke instead of endless Excel evaluations

Time savings, convenient and customizable evaluations and a better overview of a wide range of key figures are the main advantages that Carthago derives from the solution from elKomSolutions. The cost center reporting goes to around 65 report recipients. Before introducing the software, Carthago had to invest at least half a year to create evaluations for all of them on their respective budget trends. Now all you need to do is press a button to send your specific report to all report recipients.

This “very great strength” also affects other specialist areas. Similar to car manufacturers, Carthago customers can configure their motorhome. A large selection of product variants is available. With the BI tool from elKomSolutions, these can be analyzed over several years and thousands of customer orders in a matter of seconds - in order to better understand customer needs and, for example, to be taken into account for future product development.

elKomBI has significantly accelerated the evaluation of detailed information from the product range over many years. elkomBI creates these reports in seconds: “We now run a data set with 20 million individual pieces of data through the software and receive the analyzes in a few seconds,” notes Staud. That really impressed him as a controller. Completely automated, the software delivers such reports not only monthly, but daily.

Outlook: even more visibility, even more functionality

“We have now rolled out the reporting to the first and second management levels for the purpose of economic management. This greatly benefits the first level of management, who have access to consistent data. The same applies to the second management level, which receives the information it needs at the push of a button,” says Staud. The Carthago Group's controlling department is constantly developing the analyzes and reports “because the benefits are so great”. After the cost center reporting and the managing director reporting, there are now individual reports for the specialist departments, for sales, production and others such as: purchasing controlling. Head of Controlling Staud summarizes: “In the area of controlling, we have catapulted ourselves into a different era. When we commissioned our new plant in Slovenia, we were particularly able to benefit from the scalability. In the Excel world we would probably have had to hire significantly more staff. Now, all you need is powerful BI software.”

elKomSolutions: Integration as the key to success

elKomSolutions GmbH has pursued an integrated approach since it was founded in 1997. From the outset, it has relied on IBM Planning Analytics TM1 database technology and connects previously separate departments in the company with a uniform database based on OLAP technology. The basic idea is to make work as easy as possible for customers from small and medium-sized businesses, but also large corporations. But meeting the customer where they are currently traveling also means starting from Microsoft Excel. 95 percent of companies continue to use the versatile tool for controlling, sales and production planning, group consolidation and reporting. However, there are limits, be it in terms of the efficiency, speed and transparency of these processes, or in their integration.

For this reason, elKomSolutions has created modular software components for the areas of business intelligence (BI), corporate planning and group consolidation - and is constantly developing them further. elKomBI, elKomPLAN and elKomKONS are independent solutions for the individual sub-areas. Connected via OLAP technology, they integrate seamlessly and offer all departments direct access to all data. What they have in common is extremely fast data processing, even with extra large volumes, openness to any database of operational systems, etc. from the ERP and financial accounting environment, the high level of flexibility and adaptability - without external employees having to do this - as well as a uniform front end for controlling, planning and consolidation. And of course Excel can still be used, which reduces the training effort for departments that use Excel, for example for data entry.

The solutions form a holistic, modular software platform for analysis, reporting, planning and consolidation, in short for integrated results and financial management. Through integration into the operational world, classic performance management becomes identical in terms of approach, methodology and database. In other words: planning, forecasting, consolidation, but also actual reporting are always from a single source.

Von Prof. Dr. Ronald Gleich
Professor für Management Practice & Control an der Frankfurt School of Finance and Management / Leiter des dortigen Centres for Performance Management & Controlling

Summary

Der Vertrieb ist eine der Kernfunktionen im Unternehmen und sollte intensiv durch das Controlling unterstützt werden. Dieser Beitrag beschreibt, welche Logiken im Vertriebscontrolling existieren und was die wichtigsten Ansatzpunkte sind, um von Seiten des Controllings die Vertriebsperformance transparent zu machen sowie zu steigern. Schließlich wird aufgezeigt, welche Erkenntnisse aus aktuellen Studien gezogen werden können.

Prozessmodell des Vertriebscontrollings

Betrachtet man das Controlling-Prozessmodell der International Group of Controlling (IGC), dann ist das Vertriebscontrolling analog zum Produktions- oder dem Personalcontrolling ein spezieller Ansatz für eine Funktion der betrieblichen Wertschöpfungskette. Unter dem Vertriebscontrolling versteht die IGC im Detail „…die gezielte Steuerung der strategischen und operativen Vertriebsaktivitäten, um die Effektivität und Effizienz des Vertriebsbereichs sicherzustellen“ (IGC 2017, S. 68).

Wie Abb. 1 zeigt, umfasst das Vertriebscontrolling Teilprozesse wie

• die Durchführung einer Vertriebs-Kostenrechnung (z. B. mit Instrumenten wie der Produktergebnisrechnung, der Deckungsbeitragsrechnung oder der Auftragskalkulation),
• strategische und operative Planungsaktivitäten,
• ein Vertriebs-Reporting mit vertriebsrelevanten Steuerungsinformationen (in der Regel verschiedene vertriebsbezogene Kennzahlen) sowie
• ein vertriebsbezogenes Risikocontrolling.

Ablauforganisatorisch liegt es nahe, sich im Rahmen des Vertriebscontrollings mit Vertriebsprozessen und deren Performance auseinanderzusetzen. Dies ist sowohl Gegenstand der Vertriebs-Kostenrechnung als auch des Vertriebs-Reportings. Dabei analysiert man etwa Kosten- und Leistungsmerkmale wie

• Prozessdurchlaufzeiten,
• Prozesskosten,
• Prozessqualität,
• Wartezeiten,
• prozessinterne und prozessübergreifende Koordinierungsprobleme,
• prozessinterne und prozessübergreifende Doppelarbeiten,
• wertvernichtende Aktivitäten,
• wirkungslose bzw. wertschöpfungsneutrale Aktivitäten oder
• Fehlerbearbeitungen
   

Initial situation

Business Intelligence (BI) modernization has great potential for benefits on a strategic, tactical and operational level. The existing BI solutions in companies are often not geared towards agility and sustainability, which is why they cannot adapt to and grow with constantly changing requirements.

From conversations with experts and customers who have actively dealt with this topic, we came to the conclusion that BI would have to function like a kind of “data factory”: a flexible, automated and sustainable system where users can obtain relevant data products without having to put these together or verify them themselves.

Dimensions of BI modernization

As a result, a model was developed in a workshop with external experts that characterizes the dimensions of BI in their “outdated” and “modern” forms. Below is a list of the dimensions with a partial outlook on the “modern” form:

• Strategic importance: Data is viewed as a company asset
• Data-driven management: Actions are triggered by data products
• Financing: Long-term and recurring investments are made in BI
• Project approach / time-to-value: Agile, BI-specific development processes
• BI organization & managment: Established BI organization and defined processes
• Data literacy / transfer of know-how: users receive training and further education
• Technology / suppliers / service providers selection: Users are involved in the selection
• Platform location / scaling: Scalable (cloud) solutions are in use
• Degree of automation: Processes are (partially) automated
• Types of data sources: Relational, semi-structured, internal and external data sources
• Data Governance: Is an integral part of the system architecture
• Degree of use of BI: Data is used company-wide
• Notation standardization: Design based on notation standard (e.g. IBCS)
• Distribution: Data products can be consumed independently by users
• Advanced Analytics: Forecasting and optimization models are in use

Based on these dimensions, a procedure was developed that should allow our customers to implement BI modernization as efficiently and transparently as possible.

With the 5-STEP approach to a modern data factory

The goal of the 5-STEP approach is to keep the time-to-value for users as small as possible. The first results should be able to be used accordingly quickly and users should be empowered to make data-driven decisions. The transformation should have a sustainable character and the customer should actively participate in it and be enabled to operate the data production lines and ultimately the entire data factory themselves.

CHALLENGE

Adaptable controlling solution for growing companies

The controllers at Henkell Freixenet have many years of experience with Jedox; The software was introduced into the company in 2012 and central controlling has constantly modernized the application and established it as a group-wide platform for planning and reporting. And even after the Henkell Freixenet merger, the scope of applications will continue to be expanded.

During the initial implementation, the goal was to support the group consolidation of the management results with a suitable tool and to replace Excel. Over the years, more and more departments used Jedox for planning and reporting. An important milestone was the establishment of self-service options with easy access to the fully automatically updated daily invoice via Jedox Web.

Since joining Jedox Cloud at the end of 2022, there have been new opportunities for innovative applications. The Controlling International department is currently primarily exploring the further development of predictive forecasting and driver-based planning using Jedox AIssisted™ Planning.

SOLUTION

Reporting and planning for a wide range of topics and expert areas

With the move to the cloud, the number of active users has increased to over 150 worldwide. Six departments currently use Jedox for a wide variety of topics: from sales and rolling demand planning, predictive forecasting, fixed cost controlling, technical reports, brand reporting and market data analysis to group consolidation, IC publication and a pricing database.

Dashboarding instead of rigid PDF reports

In international controlling, the focus is primarily on management reporting. Management currently receives a PDF report consisting of approximately 120 pages twice a month, which summarizes the paragraphs by region and brand. The PDF version is static and at first glance does not allow any statements to be made about action-relevant KPIs such as drivers.

Using Jedox Canvas, the project manager built a prototype dashboard that summarizes the diverse sales information in a screen window with dynamic filters. Using the canvas functions, a variety of content and visualizations can be arranged side by side using drag-and-drop and the most important KPIs can be easily pinned to the report. The development of the global icons, deviations or individual drivers in the portfolio can now be seen at first glance in the dashboard, and management can create any additional detailed views, e.g., you can drill down by region, business unit, brand, brand subgroup or product group.

Simulation, predictive forecasting and driver-based planning

Using AI-supported forecasts from Jedox AIssisted™ Planning, controlling is currently checking whether the planning quality can be further increased with a view to the group result. Simulations also help to identify key drivers of business development and integrate them into planning.

When comparing the historical actual and planned values from previous years with the forecasts that the Jedox AI model generates on this basis, the automated algorithm performed significantly better than the laboriously created traditional planning, with a forecast accuracy of almost 91 percent. This was achieved in the first step based on key figures from just four years - as the database grows, the AI module can be expected to be more and more accurate.

OUTCOME

Scalable and investment-safe analysis platform

With Jedox, Henkell Freixenet has created a stable and expandable platform for planning and performance management, which has grown steadily with the successful company in over ten years of use. A key advantage for the experienced controllers is that they can further develop their application independently and roll it out within the group as required.

For example, in the latest expansion stage, the department responsible for management reporting created the sales dashboard without the need for consultants, which is able to replace the comprehensive PDF report. The dashboard offers faster and better insights as well as more analysis options.

The department also has data management under control and is continuously automated, from direct Excel uploads to a self-developed upload center and direct access to the source systems via the SAP connector. Among other things, this has made the work in investment controlling much easier and has meant that the responsible team can concentrate more on analyzing the data on reporting days instead of uploading it. And with better data quality and fewer coordination loops: Companies can not only provide the data more easily in self-service, but can also more easily validate and release it internally in advance in the upload center. This increases the companies' process commitment and ownership of their numbers, and saves over 175 person-days of non-value-adding work steps every year.

In addition, global access with live reporting status enables seamless, global collaboration by eliminating time zone-related downtime and sometimes nerve-wracking, short-term adjustments. Henkell Freixenet will continuously expand the Jedox platform. An important goal is the further development of driver-based planning using AI-generated suggested values. By simulating and verifying possible drivers, the planning of management-relevant KPIs should be further automated and made more precise. The team also wants to further optimize the visualization - and, above all, make better use of Jedox's great strength as a planning tool.

“Until recently, annual planning took two months of work and involved entire teams. We are talking about an enormous number of working hours and the main problem: the error rate. Today, K4 Analytics can resolve the issue within hours.”

MARTIN ČEVELA, BUSINESS INTELLIGENCE ANALYST AT ESET

The customer

ESET is one of the world's largest manufacturers of security software and is one of the pioneers of virus protection. The company develops award-winning threat detection software for customers in more than 200 countries.

ESET – the challenge

• Financial and human resources planning, consolidating financial plans across the company
• Automated consolidation of plans and results
• Management of deviations
• Increase planning efficiency (top-down and bottom-up, re-forecasting, time saving)
• Versioning of plans
• Workflow

The solution

With K4, your requirements for a modern planning process can be solved with manageable effort:
“Under normal circumstances the process would take a very long time and cause a lot of stress. Working with Excel requires constant concentration, which can be overwhelming, especially in stressful situations, and can easily lead to failure. Additionally, the company is wasting money because employees are doing something they shouldn't be doing. They are demotivated and discouraged, and there is also a huge margin for error,” reports Martin Čevela.

Due to the positive experience with K4 Analytics, ESET has expanded its use from the finance department to sales. Sales teams can now improve collaboration with their partner networks and, for example, distribute their marketing budgets more efficiently depending on their performance.

The sales department now uses K4 instead of MS Excel® to deliver data to the finance department and K4 takes over a large part of the planning. Sales employees are given a tool with which they can continuously monitor their set goals. This means that retailers know immediately and at all times where they stand, whether they have to work more or whether they can look forward to a surcharge.

The use

K4 Analytics is particularly beneficial for ESET employees at lower management levels because they do not have to spend a lot of time recalculating costs due to a change from above. Now you can select an item to reduce the budget for and the changes will immediately be reflected in other parts of the plan. The procedure described above completely avoids the tedious management process, which also includes Excel files sent via e-mail, which must (additionally) be provided in a user-friendly format. This means less time wasted.

“Fewer errors and less time spent on Excel means that employees can work on something meaningful. They can save up to twenty days per month and invest the time in strategy, resulting in real added value,” concludes Martin Čevela.

“With Excel the options were limited, now we can plan flexibly and quickly. The new planning system integrates all partial plans and will be rolled out to our branches worldwide. What we like best about it is the ability to plan continuously and issue flexible forecasts,” says Daniel Olthaus, Head of Financial Analysis and project manager for the ROSEN Group in Germany.

About the ROSEN Group

The ROSEN Group is a world-leading provider of highly developed technical solutions. Active in over 120 countries, the group implements innovative technological solutions for the inspection of industrial systems in the energy industry to ensure safe and reliable operation according to the highest standards to protect people and the environment.

Facts about ROSEN
Founded: 1981
Employees: 3.900
Headquarters: Stans, CH

“We have worked very Excel-heavy in the past. As many people know it – preparing and merging data. Some macros had opening times of over a minute, which made user handling very difficult. Overall, the planning took too long. Since introducing the system with smartPM, our planning and forecasting cycles have been massively shortened.” Daniel Olthaus, Head of Financial Analysis, ROSEN Group in Germany.

INITIAL SITUATION

Initial situation and challenges of the ROSEN Group before the start of the project:

• Time-consuming and laborious merging and maintaining large Excel sheets.
• Ensuring good data quality and data timeliness was challenging.
• Different partial plans continually led to coordination efforts. The focus was on comparing partial plans and analyzing deviations.
• The analysis was neglected and the derivation of management decisions was challenging because data interpretation was discussed instead of analyzes and recommendations for action.
• The planning cycles were long. There was a lack of flexibility and responsiveness.
• The call for a professional, integrated system with high user acceptance was loud.

Daniel Olthaus, Head of Financial Analysis:

“We planned the implementation of the fully integrated planning project in several implementation steps. We were aware that we would not have a fully developed planning system from the start and that we would not be able to use it across the group. That's why we have planned 5 stages or project phases:

1. Legal planning of the German units: cost centers, project, personnel, CAPEX, investment, P&L with different planning cycles, so that we don't just plan on annual slices, as in the past, but continuously, throughout the year. The plan is continually updated and scenarios can be differentiated in the flexible system.
2. Based on this, the hourly rates are derived from the planning and the balance sheet figures (intra-year development of working capital).
3. Integration of legal entity planning and sales planning with order intake planning and sales and profit center planning.
4. Automating the linking of subplans for a fully integrated system. A change in the subplan shows the impact on the other subplans (sales, cost planning, etc.).
5. At the end there is the fully developed xP&A with a coherent picture of the future development for the entire group.”

Project goals

• A fully integrated system with connection to all data sources and networking of all sub-plans that impact financial planning (sales/sales, procurement, production, delivery/service provision) for the entire group.
• Acceleration of the budget process: Continuous planning with the ability to incorporate adjustments directly. This allows management decisions to be optimally supported.
• Improving data quality and timeliness
• Quick response to changing partial plans including adjustments during the year and flexible, easily configurable scenarios/bandwidth planning with value drivers.
• Mapping and merging different planning dimensions: sales view vs. management view of the legal units.

Daniel Olthaus, Head of Financial Analysis at ROSEN Group in Germany:

“We can now plan and make decisions continuously - with current data. The planning process is standardized. All sub-plans are linked and the need for coordination has decreased significantly. The users or user groups have their own cockpit with their own planning steps. The responsibilities are clearly assigned and the submission of partial plans is scheduled in the system. This allows the status of the planning to be monitored. Thanks to the Teams and Office integration, you can contact the planners directly. This saves a lot of time. For example, planning scenarios, cost types compared to previous year's values, etc. can also be carried out in the standardized planning masks. Value changes can be easily adjusted and distributed automatically. Deviations can also be easily identified and comments for any special effects can be entered by the planners. Planning can be done at the account level and parent level with automatic distribution.”

Project implementation

• 2022: Development of a 5-step plan: Go Live planned for 9/2023
• December 2022 to March 2023: Software selection with exact requirements lists and demo dates with 5 providers on the longlist.
• January 2023: Implementation of prototype capacity planning with ROSEN company data through smartPM, another 4 providers show their solutions
• April-May 2023: Proof of Concept with 2 providers (shortlist): Testing data model and connection, user management, dashboards, license model, support, flexibility
• May 2023: Unanimous decision for smartPM.solutions. Start implementation capacity & resource management, cost center planning, data connection
• June 2023: Implementation of financial reporting
• July/August 2023: Implementation of cash forecasting and test phase of the system, implementation check whether all requirements are met
• September 2023: Planning go-live exactly according to schedule for all planners/users

Daniel Olthaus, Head of Financial Analysis, ROSEN:

“Support was a key point in the decision for smartPM.solutions. The software for most providers is quite identical. Our focus was more on: What is the implementation team like, what is the support like and how does it work after the project. We had a good feeling that the implementation with smartPM would also be successful. This was an important point for us and is now an important factor, even after we have completed the first implementation step on time, because points keep changing during the project or implementation phase, where we notice that we can cope with it in the short term Providers must react together and adjust their priorities. Ultimately, what is crucial is that we have a partner who is as flexible as possible and with whom we can implement things quickly. So far we have made the right decision with smartPM.”

Conclusion & Learnings

• Step by step introduction is highly recommended – the proven 5-step plan
• Decisive success factors for the BI and FP&A implementation project: Not the software, which is very similar from many providers, but the support even after the project and the experience of the implementation partner.
• Massive improvement in data and decision quality by networking data and partial plans in one system. Significantly simplifying the monitoring of key financial indicators. Time savings: Fast and continuous budget planning.
• Very clear personnel planning and investment planning with planning masks per cost center. Complete reporting with very good visualization displayed in the system.
• High user acceptance, responsibilities can be adjusted centrally very easily – for entire units or functions, good user guidance with an individually definable home page – cockpit.