Local AI: Governance Before Hardware Purchase
Benedikt Langer
10 min readFour developments over two weeks show that locally operated AI goes far beyond the tech stack. ...
Read Article
8 min read
98.8 percent of Fortune-1000 companies invest in data initiatives. Only 37.8 percent have succeeded in building a data-driven organization. The gap between investment and outcome isn’t a technology question – it’s a culture question. And 78 percent of surveyed executives name exactly that: people, processes, and organizational structure as the biggest barrier.
TL;DR
- 78 percent fail due to culture – not technology: People, processes, and organization are the greatest barrier on the path to becoming data-driven (NewVantage Partners Executive Survey 2024).
- 80 percent of governance initiatives will fail by 2027: Gartner forecasts that four out of five data governance programs will miss their goals due to insufficient urgency or pressure to act.
- CDO role under mounting pressure: Success rate stands at 51 percent; failure rate is 6.3 percent – the highest recorded across five survey cycles. Gartner warns: 75 percent of Chief Data and Analytics Officers (CDAOs) who cannot demonstrate organizational influence risk absorption into IT.
- 61 percent are reshaping their Data & Analytics (D&A) operating model because of AI technologies (Gartner 2024). The AI wave is forcing a cultural shift many organizations have postponed for years.
- Data culture starts at the C-level: Without visible, active commitment from top leadership, every data initiative remains an IT department project – not an enterprise-wide transformation.
The Culture Trap: Why Technology Alone Solves Nothing
Over the past decade, most companies have invested heavily in data infrastructure: data lakes, data warehouses, BI tools, machine learning platforms. The technology is in place. What’s missing is adoption. According to the NewVantage Partners Benchmark Survey 2025, the majority of organizations continue to struggle with adoption and transformation. Cultural challenges have barely improved over five years.
The problem is systemic. When the head of sales keeps pipeline information in their head rather than updating the CRM, when production exports machine data to Excel instead of feeding it into the central platform, and when the board makes decisions based on experience – not data – the best infrastructure becomes worthless.
Source: NewVantage Partners, Data and AI Leadership Executive Survey 2024
The CDO Under Pressure: A Role Searching for Legitimacy
The Chief Data Officer (CDO) should serve as the bridge between data and business value. Reality is sobering. The CDO role’s success rate stands at 51 percent. Simultaneously, its failure rate is 6.3 percent – the highest figure measured across the past five survey cycles.
Gartner intensifies the pressure: By 2026, 75 percent of Chief Data and Analytics Officers (CDAOs) who cannot demonstrate organizational influence and impact risk being absorbed into the technology function. The CDO role won’t disappear – but it loses autonomy if it fails to deliver tangible business value.
For executive leadership, this means: A CDO without board access and without authority to change processes is merely an expensive title. The role only works when treated as strategic – not as extended database administration.
“80 percent of Data and Analytics Governance initiatives will fail by 2027 – because there is no real or manufactured sense of urgency.”
Gartner, Predicts 2024: Data and Analytics Governance
AI as Catalyst: Why Data Culture Can No Longer Wait
According to Gartner, 61 percent of organizations are evolving their Data & Analytics (D&A) operating model due to AI technologies. That’s the good news: AI is forcing the cultural shift data teams have demanded for years.
The mechanism is simple. AI models only work with clean, accessible, and well-documented data. An LLM trained on unstructured silos delivers garbage. Suddenly, data quality becomes a business risk – not just an IT concern. The board, which previously waved off data governance, pays attention the moment its AI project fails.
Gartner states the point bluntly: Governance initiatives fail when urgency is absent. AI supplies that urgency. Companies aiming to scale AI must first build their data culture. The sequence is non-negotiable.
Five Steps to Building Data Culture at the C-Level
1. Make data-driven decisions visible at the C-level. Every C-level meeting includes at least one agenda item where data forms the foundation of the decision – not as a presentation slide, but as a traceable, logical decision process. When the CEO uses data, the organization follows.
2. Equip the CDO with mandate and board access. The role must be strategically embedded – with direct access to the board and explicit authority to redesign processes. Without organizational influence, the CDO becomes an expensive advisor with zero impact.
3. Define data literacy as a leadership competency. Not every leader needs to write SQL. But every leader must understand what data their area generates, what questions those data can answer, and where critical gaps exist. This isn’t a technical skill – it’s leadership competence.
4. Anchor data quality as a KPI. As long as data quality remains unmeasured and unreported, it stays lip service. Define it, measure it, and make it visible in board reporting. What isn’t measured won’t improve.
5. Prioritize use cases – not strategy documents. Data culture doesn’t grow from an 80-page strategy paper. It grows from success. Identify one business unit, implement one concrete use case, and visibly demonstrate results – then scale. The mid-market needs outcomes – not frameworks.
Conclusion
Data culture is not an IT project. It’s a boardroom responsibility. The numbers are unequivocal: 78 percent of organizations fail due to culture – not technology. The CDO alone cannot drive this cultural shift – especially when the role itself is under pressure. AI delivers the urgency data teams have sought in vain for years. Those who delay action now will soon lack both a functional data strategy and a scalable AI initiative. The order is clear: culture first, technology second, scaling third.
Frequently Asked Questions
What exactly is data culture?
Data culture describes an organization’s ability and willingness to make decisions systematically based on data – not experience or hierarchy. It encompasses data literacy across all levels, transparent data quality standards, and organizational embedding of data-driven decision-making processes.
Why do so many data governance projects fail?
According to Gartner, 80 percent lack sufficient urgency or pressure to act. Governance initiatives are perceived as bureaucratic rulebooks – not business-critical necessities. Without a genuine or engineered crisis, there’s no impetus to change processes or behaviors.
Does every company need a dedicated CDO?
Not necessarily as a standalone C-level role. But every organization needs someone empowered to own data quality, data access, and data usage across functions. In smaller mid-market firms, this may fall to the CIO – or an expanded CIO mandate. Crucially: board access and authority to drive change must be present.
How do you measure data culture?
Three key indicators: First, usage rates of central data platforms (e.g., how many decisions demonstrably rely on data). Second, data quality scores (completeness, timeliness, consistency) for core datasets. Third, data literacy rate: what share of leaders can independently interpret data analyses and translate them into decisions?
Is AI the lever for better data culture?
Yes – paradoxically. AI projects fail immediately and visibly due to poor data quality, whereas traditional BI projects often appear functional – even with mediocre data. This highly visible failure creates the urgency data teams have requested for years. Already, 61 percent of organizations are adapting their D&A operating model specifically because of AI.
Read More
Board Reporting for IT: 5 Metrics the Board Actually Understands
AI Is Not a Tech Problem: Why 27 Percent of CEOs Fail
The Digital Operating Model: How CIOs Are Restructuring Their IT Organizations
More from the MBF Media Network
MyBusinessFuture: AI in the Mid-Market – Why Companies Hesitate
cloudmagazin: AI Cloud Costs Spiraling Out of Control
SecurityToday: NIS2 in Germany
Header Image Source: Pexels / Yan Krukau (px:7691673)