How to lead an AI revolution
Diana Schildhouse, Chief Analytics and Insights Officer, Colgate Palmolive relayed her four-part strategy for leading the AI revolution. She emphasised a business-first approach, involving understanding team needs and clearly communicating a vision through a concise strategy that explains the why behind actions.
Promoting data literacy is crucial, Schildhouse added, and data leaders should make this as inclusive as possible by avoiding jargon and simplifying information to drive engagement and adoption. This can be achieved through improved storytelling skills.
Creating value is the final piece of the puzzle, according to Schildhouse, which entails articulating the value of what you do and telling the story internally, as well as identifying your priority areas so that you can focus on achieving your goals. Be sure to share the glory and the story of how this has benefitted the business, she stressed.
Schildhouse stated the importance of quantifying value, but emphasised that data provides insights, not actions: “Our goal is to deliver impact, not outputs. Ask what actions were taken based on the recommendations and aim to quantify the expected benefits, ensuring alignment with the derived value.”
Continuing the theme of leading with AI, but in a different way, Shashank Kadetotad, Global Director and Head of Data Sciences, Mars Wrigley, delved into how the data team at Mars Wrigley has taken a “business problem first mindset” to tackling value creation with AI.
Kadetotad explained that there are numerous wins to be had with AI – top- and bottom-line growth, speed and savings, flexibility, connectivity, and more – but these can only be achieved with a connected core foundation. “Data strategy should be core to any company in this digital age,” said Kadetotad, and this digital strategy is what leads on to the connected data foundation.
It is through a connected data foundation that the Mars Wrigley team can understand and make connected decision with a clear sight into different impacts, and this in turn generates interconnected insights. Following this, the data team can then create an enterprise level analytics platform and subsequently a genAI engine.
Leading through AI transformation
Sathish Muthukrishnan, Chief Information, Data, and Digital Officer, Ally Financial, emphasised the importance of building your business to accommodate emerging technologies. At Ally, where 93% of enterprise data now resides in the cloud, they successfully transitioned from a hardware-defined to a software-defined organisation, enabling them to leverage the genAI trend effectively.
Muthukrishnan’s advice for establishing a robust technological foundation get a head start on new advancements included developing an AI Playbook to act as a translator for the new language you will be speaking, agreeing on the non-negotiables, such as human in the loop, and the things that will create value for you.
This message then needs to be communicated, team members trained, and engagement should be sought through events and speaker sessions. Data leaders need to bolster this by sharing insights and use cases from outside of the tech sector. The focus must be on delivering value and moving tech from a cost centre to a revenue generator, Muthukrishnan encouraged – genAI offers significant opportunities to deliver exactly this.
Being a data narrator
Effective data storytelling is all about absorbing complexity and delivering simplicity, according to Avinash Tripathi, Vice President of Analytics at the University of Phoenix. But while every organisation will claim to be data driven, McKinsey reported that only 8% are able to draw value from their data investments, and an IDC report on data culture found that only 30% of decisions are driven by data.
To improve this crucial skill, Tripathi proposed a three-pronged approach:
- Align your analytics resources with an overarching strategy.
- Close the widening skills gap.
- Change the way your organisation communicates with data.
As for the data story itself, the solution should not be more complex than the problem, Tripathi stressed. “Start with a clear understanding of the problem itself, identify the success criteria for the use case, focus on two data metrics at most, and identify and analyse your stakeholders – this part is critical as it’s all about collaboration.”
Innovation does not have to use completely new or unique ideas, Tripathi continued; it is the way the idea is implemented and the value it adds that matters most. One thing to avoid, however, is going in reverse – trying to make an existing solution fit the problem is not the answer.
Scaling AI: Fireside with Blend and NBC Universal
Moving beyond proof of concept is essential if your organisation is to truly leverage the potential of genAI. John Lee, CDO at NBCUniversal, identified seven main challenges to navigate to achieve scale:
- AI talent
- Lack of strong data foundations
- Technical challenges
- Speed of innovation
- Business strategy integration failures
- Change management
- Trust
Media and entertainment giant NBCU successfully navigated these challenges to scale genAI to better understand the connection between its consumers and the content. The team broke down human decision making into basic values and motivations and fed this data into the machine to map the content, assign motivational values, and assign genAI segments to all NBCU customers. In doing so, they tied emotional responses to business outcomes, and are using this data to inform the future of their content strategy.
Being able to scale the technology hinges on setting up strong foundations to scale first and getting all the data engineering and management in order, which NBCU did with the support of Blend.
DataIQ 100 on how to lead business transformation
In this session, panellists from the DataIQ 100 pondered the true potential of AI in transforming business and industry performance. With all of us surrounded by genAI solutions, it is up to us to maximise the value and change the way we operate. But who should lead this transformation?
Meaghan Ferrigno, CFO and Chief Data and Analytics Officer, Destination Canada, suggested it is whole a team effort, and it is necessary to upskill everybody in the business, moving outwards in concentric circles.
Data culture is key to this buy-in. Figure out how to bring the change from the boardroom and establish a structure to tie it all together under the right strategy, Ferrigno said.
If you are scaling AI, expect hurdles along the way, the panel agreed. T-Mobile’s Vice President Data, Ronke Ekwenski, warned that leaders should be mindful of the difference between policing and enabling, while emphasising the need for human centricity in responsible AI.
“There’s a reason it’s called co-pilot – it’s not meant to abdicate the role of the human in the business process,” Ekwenski asserted. Meanwhile, the faster you are working to get a solution production-ready, the more you can expect roadblocks, such as legal compliance issues, to stall progress.
Data and AI are the ultimate collaborative sport, it was concluded, and so organisations with collaborative cultures are the winners in this game. A top-down and a bottom-up approach is needed to be successful, with buy-in of the leaders and the engagement of the teams on the ground needed to drive true organisational change and adoption of AI.
Panellists were divided when asked what they they wish they knew about AI sixth months ago. While Delek’s Ido Biger labelled AI a bigger wave than first thought, stating that businesses should have been quicker to adapt strategic plans, Satya Choudary of Creduit Suisse contended that the trend has not accelerated as expected: “The traditional ways in which we chase value are changing and we need to keep up with that and the fact that ROI may be harder to quantify than before.”
Corelogic: AI transforming Real Estate
Sachin Rajpal, Chief Data and Analytics Officer, Corelogic, presented the effect AI is having on the entire real estate value chain. Every actor involved from estate agent to mortgage broker, home improvement seller data and analytics was driving value.
Some of the use cases involved demonstrating how data visualisation could indicate the degree of risk a property faced due to climate change over several years. Others showed how using smart data platforms and images helped with insurance which coupled with weather data over several years would enable the best insurance premium to be derived.
Responsible AI
The former CDO at Chevron, Ellen Nielsen, presented a four-prong structure for addressing responsible AI:
- Principles
- Assessment
- Governance
- Communications
In terms of principles, Nielsen encouraged companies to consider their risk profile and what guidelines they should provide as key principles for any AI projects. She highlighted Microsoft’s AI principles as a useful framework.
For assessment, Nielsen highlighted the importance of impact assessments with the desire to ensure the organisation has an inventory of AI projects.
Regarding governance, Neilsen encouraged companies to ensure there was a human in the loop and to ensure AI maturity assessments were regularly completed as its important to view AI over its life cycle with regular checks to assess current state. She championed the audience to view the assessments from the Responsible AI institute and the need for oversight of AI from a high-level board.
Finally, about communications, Neilsen stressed the importance of regular compliance training for AI and the need for strong policies. The NIST website and frameworks are useful for the metrics companies should use.
Ultimately, the 2024 DataIQ 100 Summit gave attendees the opportunity to learn from, connect with, and develop lasting relationships with some of the most prominent minds in the industry. This was just a quick insight into some of the essential learnings from the conference, so make sure you are registered for the 2025 DataIQ 100 Summit to make the most of the opportunities presented.
To keep up to date on registration for the 2025 DataIQ 100 Summit, register your interest here.
