How trust accelerates data transformations
It is clear that the data revolution is still under way, unleashing vast opportunities for businesses to prosper and create a true impact with their products and services.1 Startups, which developed into major players over the last decade, embody one common characteristic: they are data-aware organizations that excel at embedding AI in their organizational DNA.2 However, our latest AI survey reported that only 13% of organizations were able to scale AI at large while 72% of organizations struggled with deploying one single application over the past year.1
When we engage with clients on data-driven transformations, we often find that any doubts they may have stem from the organizational culture rather than the technological setup of AI projects. Hesitations range from trust issues and fear of job loss caused by new technologies (“We cannot (only) trust algorithms …”; “Algorithms threaten our jobs …”) to reticence towards the use of data in decision making and a reluctance to embrace new ways of working (“Our own gut feeling is better than the data …”; “Everything has always worked fine without data and AI, so why change now?”). These hesitations emphasize that it is not the technology per se that is hindering the success of data projects, but the attitude and enablement of people involved.
To ensure a sustainable and successful AI journey, it is key to create trust in AI and focus on the people side of transformations. This can be achieved with three simple steps led by data-savvy colleagues – the so-called data influencers. Data influencers are well connected colleagues within organizations that live and breathe AI and can support in closing the gap between AI, business, and people, building the foundation for trust in data transformations.
Step 1 – Build trust with an inspiring AI vision
The foundation for an inspiring AI vision that explains the reasoning behind a data transformation is a transparent and bias-free project setup. Above all, AI needs to be explainable, transparent, and auditable, from a customer and co-worker perspective.1,3 This starts with asking the right question that is aimed to be answered with the help of algorithms. The project question should be precise enough to capture the challenge at hand and likely to be correctly answered so that the base for a bias-free algorithm is set. Successful organizations have shown that it is mainly the trust set in the predictability of results that drives their AI performance and their overall ethical governance.1 This helps shape a transparent and exciting AI vision that translates the project goals into an inspiring story – creating both a common sense of urgency and understanding among stakeholders and defining the way forward. Internalizing the transformation purpose (why and where to go) characterizes the starting point for trust and engagement of the stakeholders involved in the data transformation.
Step 2 – Mobilize individual data influencers
To bring the AI vision to life, it is key to identify and mobilize individual data influencers early on and involve them in the data transformation project. This represents a major obstacle in data projects since technological development without integration in the business will not succeed as reported in the recent AI survey.3 Employees who already feel motivated to test and use AI technologies and who find joy in supporting peers in the adoption of new technologies make especially good data influencers. When they are mobilized, they will build the bridge between the AI vision and the data, and animate their internal network to drive technology adoption from inside the company. By acting as role models for their peers they build trust and confidence in AI technology and the opportunities that come with new ways of working.
Step 3 – Build an open data community
The third step towards successful AI adoption is building an open data community within the company that encourages co-workers to experience new technologies and their benefits firsthand with the help of data influencers. The project team and data influencers should jointly work on developing a forum of constant exchange with company-specific formats that will evolve into a continuously developing exchange platform over time. This should also include dedicated training sessions to upskill peers and boost both understanding and knowledge of AI. Participating peers will have the opportunity to test, discuss, and educate themselves about AI and data-based ways of working. Open exchange with dedicated data influencers lowers the barriers to understanding AI and data, while new knowledge fosters understanding and trust. All in all, the data community will solidify trust and acceptance of the new technology and help to evolve a new and positive mindset towards AI and data in an authentic manner.
At Capgemini Invent, we amplify the movement through training, workshops and knowledge-sharing to assist companies in developing the trust in data needed for successful AI transformations. For our latest insights on data-powered enterprises, subscribe to the latest research from the Capgemini Research Institute.
- Capgemini Research Institute, The AI-powered enterprise: Unlocking the potential of AI at scale, 2020
- Capgemini Invent, Change Management Study 2019: Leaping forward: Paths to Organizational Dexterity, 2019
- Capgemini Research Institute, AI and the Ethical Conundrum: How organizations can build ethically robust AI systems and gain trust, 2020
Lead of Data Science Chapter