This leads directly to digital sovereignty to avoid vendor lock-in (keyword: US hyperscalers), which is no longer a niche topic for AI-ready data platforms. 64 percent of companies rate their technical sovereignty and 60 percent their data sovereignty as high to very high. Over half (52 percent) classify it as business critical.
When IT and business talk past each other
Responsibility for data and AI is highly fragmented in German companies. Formally, IT sets the tone at C-level (data: 24 percent, AI: 23 percent), but business managers claim a strategic say (data: 18 percent, AI: 20 percent), and the IT department takes the lead in 19 to 20 percent of cases. Internal competence centers play a surprisingly small role at ten percent each. This unclear responsibility matrix makes a coherent data strategy significantly more difficult.
In general, six out of ten respondents describe the interaction between IT and departments in data management as good to very good: the exception is the departments, of which only 52 percent rate the cooperation positively, compared to 65 percent at C-level. This gap of 13 percentage points suggests that management perceives a harmony that does not correspond to day-to-day operations.
Unclear responsibilities may also be the cause of gaps in data and AI skills. Although 81 percent of companies rate their know-how as comprehensive or sufficient, there is a lot of catching up to do in specific areas. The largest skill gaps are in data science / machine learning (ML) and data strategy / product thinking (38 percent each) and in data engineering (33 percent).
Data quality and maturity as AI brakes
Given all of this, it is not surprising that a lack of data quality (30 percent), fragmented data landscapes (28 percent) and sluggish IT processes (25 percent) are the three biggest brakes on data and AI initiatives. In addition, only 67 percent rate the maturity level of their data landscape as high, 63 percent are satisfied with the data accessibility and 62 percent with the data quality. The situation is even more critical when it comes to data timeliness (58 percent), data integration and metadata availability (57 percent each), where the values are below the 60 percent mark.
This does not detract from the enthusiasm to invest, because almost a third of those surveyed (32 percent) are planning very high or high expenditure on data platforms and AI for the next two years. According to Dr. Emmanuel Klinger “many companies start with AI where it quickly leads to visible results, such as in marketing and sales, and the data connection is less important. As soon as it comes to automating the backend processes and replacing classic RPA, the data connection becomes a key factor again.”
What is significant is what is considered a top priority. This is not the introduction of an AI-ready data platform (25 percent), but the automation and operationalization of AI (39 percent), the modernization of the data architecture (32 percent) and the building of a data governance framework (28 percent). The platform is therefore a means to an end. The real goal is automated and data-driven processes as a driver for digital business transformation.
The new study “AI-ready Data Platforms 2026” from CIO and Research Services
Research Services: Christine Plote
Study check letter
Editor: CIO, CSO and COMPUTERWOCHE
Study partner: Hyland Software Germany GmbH; Lufthansa Industry Solutions; T-Systems International GmbH
Population: Top (IT) managers in companies in the DACH region: those involved in strategic (IT) decision-making processes in the C-level area and in the specialist departments (LoBs); Decision-makers and experts from the IT sector
Participant generation: Personal email invitation via the CIO, CSO and COMPUTERWOCHE decision-making database as well as – to meet quota requirements – via external online access panels
Total sample: 315 completed and qualified interviews
Examination period: January 10th to 21st, 2026
Method: Online survey (CAWI) Questionnaire development & implementation: Custom Research Team from CIO, CSO and Computerwoche in coordination with the study partners
