Data Lab - the centre for business innovation

from | 24 September 2021 | Basics

Data labs are future-proof, data-driven centres for business innovation. They support the development of use cases by analysing and scrutinising existing data and identifying solutions. We get to the bottom of what is actually meant by a data lab and how they are set up. In addition, we will show common challenges and best practices, and we will discuss the DataLab in the context of the [at] Data Journey. 

Data Lab - bridge between business intelligence and specialist department

IT departments in companies across all sectors are now faced with the challenge of finding classic business intelligence solutions or Business intelligence tools with the requirements of the specialist departments. There is an increasing need to make data-based decisions faster and faster. In addition, these need to be transferred into models and routines that can be monetised. For this purpose, many departments use suitable self-service tools and Cloud services - often on their own Business Intelligence (BI) past. This almost inevitably brings with it the danger of a loss of transparency - people work at cross purposes. 

The use of a data lab can counteract the emergence of a "two-class IT". On the one hand, this is because data labs create a central location for all data. On the other hand, the introduction of corresponding processes enables the basis for intelligent work with the collected data. 

The Data Lab combines traditional and modern approaches in the form of a bimodal BI. In this way, it offers a place for innovation, but at the same time it also serves as a link in communication. The overarching goal is to analyse company data quickly and in an application-oriented manner together with the specialist departments. Accelerated feedback loops operationalise the findings in prototype development.  

The processes created in a data lab encourage the generation of hypotheses and enable them to be tested for feasibility in a timely manner. Business-relevant data-based models thus find their way into application more quickly without overloading the capacities of a classic BI.   

Strategy and experts as success factors for setting up a data lab 

Implementing a data lab in a company seems complicated at first glance and may scare some companies away. However, the advantages clearly outweigh the disadvantages and with comprehensive planning, the project is easy to implement. The following success factors play a role: 

  1. Define goals and KPIs 
  1. Define the scope and budget of the Data Lab. 
  1. Provide appropriate resources 

Based on the target definition, companies can determine the individual scope of the Data Lab. For a successful implementation, the available resources are particularly decisive. These include, above all, qualified and trained employees. The following key positions must be filled in a data lab: 

Data Analyst or Business Analyst 

A data analyst understands the requirements of the users in the various departments and brings these insights into the data lab. This employee thus forms the direct link between the Data Lab and the business - and not as a silent observer, but as a proactive player. The data analyst is responsible for actively understanding business processes and identifying needs. 

Data Scientist 

The Data Scientist translates the findings about the needs and requirements from the departments into concrete data questions. For this purpose, he works out suitable answers and develops corresponding models.  

Data Engineer 

The compilation of the data for the Data Scientist is done by the Data Engineer. At the same time, he builds bridges to the developers and data architects of Corporate BI. His task is to initiate and monitor the transfer to Corporate BI and the adaptation of the models developed in the Data Lab. In this way, he also prevents the emergence of self-sufficient data silos in the departments.  

Data Architect or Developer 

The Data Architect actually belongs more to Corporate BI than to the Data Lab. Nevertheless, he plays an important role due to his direct proximity to the user in the business departments. The Data Architect implements the models, i.e. the results of the Data Lab's work, in business operations. 

What are the challenges to be overcome? 

Even if companies have a comprehensive strategy and staff with expertise, a data lab can present challenges. The most common hurdles include: 

Compatibility of all systems used 

It is advantageous to have a technological infrastructure that is equally adapted to the needs of Corporate BI and Data Lab. When selecting suitable systems, the support of experienced consultants can be helpful. They evaluate the existing setup and provide suggestions for improvements and best practices.  

Scalability of the Data Lab 

Through the use of Cloud computing solutions scalability can be significantly increased. The requirements for storage and computing power can be flexibly adapted to the daily requirements and workloads. In addition, the creation and simultaneous use of different test environments is included. 

Data security 

Sensitive data is stored in the Data Lab. Constant monitoring and protection against access by third parties are therefore indispensable. Internally, this often cannot be covered due to limited resources. This can be remedied by cloud-based data security systems. 

The [at] DataLab - from concept to prototype 

Data & AI projects are carried out at [at] according to the Data Journey Method is being carried out. The aim of the DataLab phase is to test use cases as quickly as possible based on the data strategy and to convert them into initial prototypes. The following phases are run through: 

1. Concept 

Together with the individual departments, hypotheses are created, existing data is examined and analytical concepts are developed in use case workshops.  

2. exploration 

The subsequent use case exploration, i.e. the examination of business-relevant feasibility, is realised by building a test environment during a so-called hackathon lasting several days. After a week, it is clear whether the use case can be implemented. 

3. prototype 

Depending on the result, a prototype of an application can be developed and used in the next step. For this, a first version is created in a test environment based on the existing database. This offers realistic results and added value for the users. 

The Data Lab as the key to success 

In order for a company to be in a position in terms of infrastructure, personnel and technology not only to set up a data lab, but also to achieve helpful results with it, the support of external consultants can be useful. This not only makes strategy development easier, but also makes it possible to fill important key positions. The chances of identifying valuable company data and realising a business innovation increase considerably. Ultimately, the Data Lab offers the targeted and successful promotion of the innovative drive of its specialist departments for companies of all sizes.



Our AT editorial team consists of various employees who prepare the corresponding blog articles with the greatest care and to the best of their knowledge and belief. Our experts from the respective fields regularly provide you with current contributions from the data science and AI sector. We hope you enjoy reading.

0 Kommentare

Submit a Comment

You may also be interested in