Turning data into knowledge: Data Analytics

Data is the oil of the 21st century. But while oil is becoming increasingly scarce, the flood of data in many companies can barely be managed. Good monitoring and analysis strategies are needed to ensure that the right conclusions are drawn.

Data-based decisions instead of gut feeling – “Let the data speak”

The applications of data analytics are diverse and range from calculating simple key performance indicators (KPIs), to creating dashboards, to more complicated machine learning approaches. The goal is always the same: Turn data into knowledge. However, before the analysis can begin, data must first be collected, processed and merged.

Collect data systematically

Unorganized raw data is of little use without targeted preparation. The list of data that accumulates in a company is long: transaction data, customer messages, inventory data, performance data from machines or reference data. Often, this data is badly managed and rarely used efficiently. Data must be checked, cleansed, validated, adjusted and corrected before it can be profitably processed further. Data preparation is a tedious and time-consuming task, but it is essential because all analyses and conclusions are based on it – master data maintenance and data cleansing are therefore essential.

Ensure data quality

In the case of aggregated data, the data origin should always be traced in the data warehouse system and processing and transformation steps should be documented. Optimally, this is flanked by group-wide guidelines and standards of data governance. The enrichment of internal data with external sources, such as event, weather and geodata, leads to an even better basis for planning. Due to the continuous increase in the volume of data, even smaller companies can now use complex analysis methods for pattern recognition.


Different methods of data analytics

Various methods are available for analyzing the data, depending on the target project:

Descriptive Analytics

First, it is about describing the status quo and understanding what happened in the past - this is the core of Descriptive Analytics.

Diagnostic Analytics

Based on this, connections can be analyzed and understood; "Why is something happening right now?"

Predictive Analytics

Predictive analytics involves looking from the past and present to the future: "What will happen next?"

Necessary systems

While Excel is often sufficient for descriptive and diagnostic analytics, predictive analytics relies on business intelligence solutions with dashboards. In addition, methods from statistics and computer science can be used to recognize patterns and make predictions about future events – machine learning and artificial intelligence are the key points here.

Data analytics pays off

Models can be trained on huge amounts of historical data and the success of the methods can be verified directly. By looking in the rearview mirror, it is possible to move into the data fast lane with greater certainty. The comparison between human gut feeling and trained algorithm makes it clear that the latter is not only faster, but also more accurate.


Data Analytics Applications in Pricing, Sales and Marketing

The possible applications in the areas of sales, marketing and pricing are very diverse and have long been ready for the market. Three concrete applications are mentioned below.

1. Customer segmentation

The customer base is rarely homogeneous. The differentiation of customers into target groups is a prerequisite for specific sales and marketing strategies. The aim is to classify customers according to similar characteristics so that they can be addressed together. The size and number of segments depend, among other things, on industry, sales, product portfolio and sales channel.

Cluster analyses can range from simple one-dimensional methods – such as differentiation by purchase frequency – to multivariate methods such as conjoint analysis (LINK) and factor analysis.

2. Control of sales activities

Sales platforms can be used to better manage acquisition activities and leverage optimization potential. The focus here is on “next best action” and “next best offer” so that sales staff can submit individualized product and service proposals and choose the right time to approach customers. Which customers should be addressed and in which order (prioritization)? When should an existing customer be visited again? Which sales channel is most promising?

3. Simulations of pricing changes

Simulations are a proven means of estimating price changes. Using a digital twin – the virtual image of a company – algorithms can simulate and test the effects of price adjustments even before discounts and list prices are actually adjusted. This “dry run” serves to uncover problems and inconsistencies before the company-wide rollout takes place.

Deploying tools in Power BI, Qlik Sense or Tableau

A data dashboard provides a quick overview of KPIs, handling processes, and potential for improvement and growth. Here, business-critical questions can be answered and interactive data analysis can be started. In tools such as Power BI, Qlik Sense or Tableau, raw data can be easily visualized and changes identified in real time. In addition, dashboards can be flexibly managed, shared, updated, and unlocked. Through lean and targeted dashboards, crises are identified and addressed in time.

Project experience

A proven starting point for projects is the 360° data audit, in which the data basis and current analysis tools are evaluated in a three- to four-week check and any need to catch up is identified. After this quick check, concrete fields of action are named and an implementation plan is drawn up.
Prof. Roll & Pastuch – Management Consultants has many years of experience in the successful implementation of complex analysis projects: We develop for you

Together, we are creating the transformation from Big Data to Smart Data.

Learn more about your path to successful data analytics

We will be happy to answer your questions and provide you with further information.