In the rapidly evolving landscape of artificial intelligence (AI), businesses across the globe are recognizing the imperative to adopt this transformative technology. However, the journey towards AI transformation extends far beyond the mere acquisition of cutting-edge technology and talent. It necessitates rewiring of the whole organization; a shift in culture, structure, and operational norms to foster a nurturing environment for AI integration. This article delves into the significance of organization and culture in enabling AI transformation, emphasizing the need for business executives to spearhead changes that are not merely technical but profoundly structural and cultural.
At the heart of AI transformation lies the dual challenge of remodeling the organization and its culture. Traditional business models, characterized by rigid hierarchies and siloed departments, are often ill-equipped to harness the potential of AI. The transition to AI readiness requires an organization that promotes interdisciplinary collaboration, agility, and a data-driven ethos at its core. Similarly, the culture within must evolve from risk aversion and reliance on past experiences to one that embraces experimentation, adaptability, and continuous learning.
From siloed work to efficient interdisciplinary collaboration
Breaking organizational barriers has been a topic stressed for more than a decade. Still the siloed approach prevails to be sticky in many organizations. However, the complexity and scope of AI initiatives necessitate getting rid of organizational department fences, as the success in AI demands the fusion of diverse skills, knowledge, and perspectives, bridging gaps between e.g. between data & AI experts, IT and software professionals, legal advisors and business units. Diverse teams are likelier to ensure that initiatives address broad organizational priorities, not just isolated business or technology issues.
Some of the known issues:
- Communication barriers
Business and technical teams often speak “different languages,” making it difficult for them to understand each other’s perspectives and constraints. Data scientists and AI professionals may focus on the technical feasibility and sophistication of solutions, while business units might prioritize practicality, customer experience, and return on investment.
- Misaligned objectives
Business teams may have goals that focus on short-term KPIs and financial performance, whereas data & AI teams might prioritize innovation, cost-efficiency and the development of scalable solutions. This misalignment can lead to friction and inefficiencies.
- Cultural differences
The culture within technical teams often values innovation, experimentation, and rapid iteration, while business units may operate within more traditional, risk-averse frameworks. These differing cultural norms can hinder effective collaboration.
Strategies to overcome the pitfalls
A step-change forward requires first of all a common vocabulary. This means not only business experts adopting the data & AI terminology, but also vice versa. In addition, making an effort of helping others understand the context never hurts. E.g tools such as data dictionaries and simplified project documentation can also help make technical projects more accessible to non-technical stakeholders.
Secondly, it is hard to steer the ship if different teams are constantly rowing into different directions. Hence object alignment via common goals will be crucial for your endeavours’ success. Create shared objectives that require collaboration between business and data & AI teams to achieve. These objectives should be tied to both the organization’s broader strategic goals and to the performance metrics of the individual teams. Involving both sides in the goal-setting process ensures buy-in and helps align efforts towards common outcomes.
Thirdly, it is important to embrace a culture that values diverse perspectives and sees the benefits of learning from different disciplines as a key. This can involve highlighting successful cross-functional projects, facilitating mentorship opportunities between business and technical staff, and celebrating collaborative successes. Recognizing and rewarding collaborative behaviors promotes a culture that values interdisciplinary teamwork.
Last but not least, there should be a clear and shared understanding about the responsibilities and ways of working in the interdisciplinary teams (that might be also virtual). It is too easy to just throw people together and assume teams to work seamlessly together from a get-go. And even with a good understanding of the different roles and responsibilities, teams are made of people, and most (if not all) teams tend to follow Tuckman’s Team Performance Model1) about how the team performs in 4 stages of team’s lifecycle:
- Forming
- Storming
- Norming
- Performing
Hence new interdisciplinary teams should receive time to grow fully into their potential.
Although introduction of new, more diverse ways of working is not always walk in the park, interdisciplinary collaboration is almost every time worth it. It not only accelerates innovation but also ensures that AI solutions are aligned with business objectives and capable of addressing real-world challenges.
Organizing for scale
Over the years there has been a lot of debate about where data, analytics and AI capabilities should reside within organizations. It might be tempting to ask “What organizational model works best?” and expect a one-fits-all answer. However, the right answer is usually the boring “it depends”. There are several factors at play that need to be considered when deciding which organizational approach truly serves the needs and objectives of a company.
Three main approaches are following:
- Consolidate the majority of AI and analytics capabilities within a central organization, usually under a “neutral” ground (like digital, strategy or IT) to ensure equal and fair treatment of all company units using the services of the central analytics/AI team. Central organization can push towards harmonization in ways of working and technology use, in a cost-efficient manner. On the flipside, increased bureaucracy can occur, as well as more effort needed in making the cross-unit collaboration work, as the teams working for AI solutions then tend to be only virtual by nature
- Decentralize analytics/AI capabilities, and place them mostly in the business units, where more close and stable interdisciplinary collaboration can emerge, but on the other side of the coin looms an adverse impact on cost efficiency, where resources are multiplied several times.
- Distribute them across both, using a hybrid (“hub-and-spoke”) model, where there is a central team (“The Hub”) that provides the foundational services and acts as a center of excellence/enablement, and then there are “Spokes” that reside near the business, hence usually retaining good agility and responsiveness because of the direct connection between business stakeholders and analytics experts. Spokes follow certain principles set by the Hub, as well as use certain Hub services if needed.
Budgeting as much for integration and adoption as for technology
A common pitfall in AI transformation efforts is the disproportionate focus on acquiring technology at the expense of facilitating its integration and adoption. Business executives must recognize that the true value of AI is realized not through technology alone but through its seamless integration into workflows and its adoption by users. What was already addressed before, interdisciplinary teams can think through the operational changes new applications may require—they’re likelier to recognize, say, that the introduction of an algorithm that predicts machine maintenance needs should be accompanied by a total re-thinking of current maintenance workflows, technology interfaces and roles/reponsibilities. In addition, end-user adoption should not be taken for granted. Sometimes old habits die hard, and if there is little or obscure communication about the benefit of new solution, users might use silent resistance and just stick in the old ways. Hence significant resources should be allocated towards change management, training, and user support to ensure that AI tools are effectively embraced and utilized.
From rigid and risk-averse to agile, experimental, and adaptable
The pace of change in the AI domain requires organizations to adopt a more flexible and responsive approach. Transitioning from a rigid, risk-averse mindset to one that values agility, experimentation, and adaptability is crucial. This entails embracing agile methodologies, fostering a culture of continuous testing and learning, and being prepared to pivot strategies in response to new insights and market dynamics.
From experience-based, top-down decision making to data/AI-driven decision making at the front line
One of the most transformative aspects of AI is its ability to democratize decision-making processes. By shifting from a paradigm where decisions are predominantly based on experience and hierarchical position to one where data/AI-driven insights empower front-line employees, organizations can enhance responsiveness and precision in their operations. Cultivating a data-driven culture requires not only the right tools and technologies but also a shift in trust and empowerment towards those closest to the operational realities.
Reinforcing the change and nurturing organization to be nimble and resilient
The role of leadership is important in supporting the organization to grow resilient and flexible. AI transformation is not a one-time project but a continuous evolution, a non-linear function that is hard to depict, especially in the fast-changing environment. Organizations must be committed to reinforcing change, fostering a culture of resilience, and learning from failures. This involves celebrating successes, learning from setbacks, and remaining open to evolving AI strategies in response to emerging trends and challenges.
A couple of topics should be emphasized in particular:
- Embrace adaptability as a core value
Cultivate a mindset of adaptability among your organization. Encourage them to view change as an opportunity for growth rather than a threat. Highlight stories of successful adaptations despite setbacks on the way within the organization to inspire resilience
- Invest in continuous learning
There is a strong need to promote a culture of continuous learning and development among all levels of organization. Employees should not be left alone with mere wishes that they would automatically be on top of of the most topical industry or technological trends, but should be supported with targeted, tangible training and learning resources to acquire new skills and enable them to apply newly learned knowledge in practice. A workforce that is constantly evolving is better equipped to handle challenges.
The journey towards AI transformation is complex and multifaceted, requiring much more than technological prowess. It demands a profound rethinking of the organization and its culture to create an ecosystem where AI can flourish. Business executives play a pivotal role in driving this transformation, championing changes that foster interdisciplinary collaboration, agility, a data-driven mindset, and scalability. By aligning organizational structures and cultures with the demands of the AI era, businesses can unlock the full potential of AI to drive innovation, efficiency, and sustainable competitive advantage.
Sources:
- Bruce W. Tuckman: “Developmental Sequence in Small Groups”.
Psychological Bulletin 63 no. 6, 1965.