Are You Ready to Architect the Flow of Insights Through the AI-enabled Enterprise?

For years insights teams have battled with challenge of tracking the ROI of their research on business performance. The constant effort of connecting the deeper market and consumer understanding insight teams develop with the downstream decisions and actions of the wider business can be draining.

The explosion of new AI capabilities that are now pouring into enterprises pose an opportunity to close the gap between insight development and business impact. New ways of working will emerge as a result of embedding AI into the daily processes that drive business forward, opening the door for insights to flow more freely across functions and teams. We should expect a more direct and proactive engagement with insights and data from business users who will increasingly be supported by AI assistants and advisors that will guide them through their daily tasks.

Generative AI alone is predicted by McKinsey to add up to $4.4 trillion in value annually to the global economy through productivity improvements. The transformation of how we work is going to sweep across all industries and bring with it a redefinition of the skills and capabilities we need to operate successfully in this new environment.

Emergence of the Insights Architect

Insights professionals will be no exception to this AI triggered workplace transformation. Tech literacy and competence will need to be dialed up significantly as insights leaders take on a central role in architecting the overall flow of insights through their enterprise’s digital nervous system.

While they will need to work more closely than ever with colleagues in IT and operations, it will be for the insights leader to define the end-to-end insights journey. In addition to being accountable for the quality and volume of knowledge available, in future they will have responsibility for how intelligence reaches and is absorbed by the business.

Insights leaders will need to consider in detail where insights and intelligence will be connected to business tasks — for instance, which group of AI assistants will be interfacing with other assistants to locate and draw in the relevant insight? Importantly, within this new technical architecture, how will the insights leader govern and validate the insights and underlying data that are being shared on a real-time basis? Rather than defining the end point of the insights management process as a handover to the business, tomorrow’s corporate intelligence strategist will plan how insights are woven into the everyday operations of the business.

Tomorrow’s guardian of data quality

While generative AI presents incredible opportunities, it also raises important ethical considerations. As AI generates content autonomously, there is a need for human oversight to ensure the outputs are trustworthy, and that any recommendations or suggestions align with brand values and ethical standards. A balance needs to be struck between efficiency and ensuring the core insights and intelligence continue to be robust.

The risks of incorporating poor-quality data into strategic planning are not new. Insights teams already have the responsibility for scrutinizing data before sharing it with their stakeholders. Equally, they have governance practices in place to ensure newly developed research is created by adhering to exacting standards.

In some industries, such as Pharmaceuticals and Financial Services, detailed compliance audits are required to track the entire research process. What is new about incorporating AI is the speed with which data sources are being combined to uncover new knowledge. In parallel, the volume of raw data being fed into AI-powered insights development is vast.

The insights architect will be the one to ensure that the AI tools used can faithfully share knowledge. This will again require a deeper level of technical understanding than has been required in the past. Not only will the insights architect need to be satisfied that ‘AI hallucinations’ are not present, but they will also need to govern how AI orchestrations pass data between different AI tools, each of which will have expert functionality.