Companies agree that artificial intelligence is the decisive key technology and will majorly impact business models and processes in the next few years. However, data availability, data quality, and a lack of data architecture often slow down AI projects. The quality and cost objectives can only be achieved with the right strategy, as a new study by Lünendonk shows.
Companies believe AI has the potential to change their entire industry disruptively. Because of this assessment, however, it is astonishing that only one in four of the large companies analyzed has a dedicated AI strategy or a definition for AI , and less than one in two of the few AI projects is in the productive operation of the companies surveyed. Reasons for this include missing or bad data, a lack of data science know-how, and hurdles within the organization. This includes, for example, the lack of interdisciplinary cooperation, which is usually required in AI projects.
For the current Lünendonk study “Artificial intelligence – fields of application, challenges and goals of AI projects in large companies and corporations,”
AI Projects: Poor Data Quality Takes Revenge
“Data is the new oil” – this statement has been heard more often in the past five years. If this analogy applies, it must be stated that the “oil quality” in companies is often poor. Data pipelines do not fit together, or the individual company areas are not ready to share their data. Legacy IT systems are the major challenge, especially for non-start-up companies.
“Existing solutions can also be integrated with clean data maintenance and a well-defined architecture. But in terms of operational implementation, this is a major challenge for companies. In addition, data architectures are an unpopular field for them, as the direct business case is difficult to calculate,”.
However, if the homework in data quality and data architecture is not done, then AI projects will deliver incorrect results and fail. Over 70 percent of companies, therefore, see data as a big or very big challenge. In addition, over half of the companies surveyed stated a lack of know-how to implement data science projects.
A Lack Of Strategy Slows Down The Implementation Of AI
For companies to be well prepared for the developments in artificial intelligence, they need a company-wide strategy to modernize the company’s IT in a targeted manner on the one hand and to build up AI competence within the company step by step on the other. In addition, the organizational prerequisites must be created so that specialists and managers have solid knowledge in artificial intelligence, interdisciplinary cooperation is possible, and an internal competence center for artificial intelligence exists that supports the departments in implementation.
“Companies can only develop into a data-driven company in the long term if they proceed according to plan. And, as the discussions show, AI initiatives can only be successful in the long term with a clear commitment from the Executive Board,”.
AI Projects: Numerous Use Cases Already Exist
Artificial intelligence is no longer as abstract as is sometimes assumed. In numerous areas of the company, such as customer services, marketing, sales, and production and development, there are already diverse use cases listed in the study with specific examples. All of the respondents assume that the future potential of AI in almost all areas of the company is very great. Customer service as well as marketing and
Sales emerge as fields of action. For example, companies expect more sales and happier customers through individual and intelligent interaction, seven days a week and 24 hours a day. In general, the interviewees stated that in many cases, AI should be used specifically to prepare decisions, but in the end, it is the person who makes the decision.