July 31, 2026 AI’s Read-world Impact in Category Intelligence

How has AI moved from hype to real-world impact in category intelligence in the last few years?

By Madhukar Irvathraya, Co-founder and Managing Partner at Oraczen

Key Takeaways:

• Traditional category intelligence is struggling to keep pace with rapidly changing markets, leaving procurement teams to make decisions using outdated or incomplete information.
• Critical procurement knowledge often remains trapped in siloed systems and individual expertise, limiting collaboration and creating long-term organisational risk.
• Agentic AI enables continuous category intelligence by combining live market signals with internal organisational knowledge to provide timely, actionable insights.
• Subscription-based agentic AI platforms are making continuous procurement intelligence more accessible, helping organisations improve decision-making without lengthy AI development programmes.

Artificial intelligence has dominated procurement conversations for a while. While the technology promised better sourcing decisions, greater efficiency and deeper insight, many teams are still struggling with the same challenges: limited bandwidth, fragmented knowledge and intelligence that is already outdated by the time decisions are made.

But, the next stage of AI adoption is beginning to look very different. Rather than generating more reports or dashboards, agentic AI has the potential to continuously monitor markets, connect fragmented organisational knowledge and provide procurement teams with intelligence that evolves alongside changing market conditions.

“The future of category intelligence won’t be defined by how much data procurement teams can access, but by how quickly they can turn changing market conditions into better decisions,”

argues Madhukar Irvathraya, Co-founder and Managing Partner at Oraczen.

Why traditional category intelligence is struggling to keep pace

Category management has always required procurement teams to balance supplier intelligence, spend analysis and market knowledge to make decisions that can significantly influence organisational spending. The challenge today isn’t a lack of expertise or flawed processes. Procurement teams know what good category intelligence looks like. Instead, the issue is that the procurement environment is moving faster than the traditional approaches designed to support it.

A sourcing strategy completed last month was likely accurate when it was developed. However, markets move on, never staying static for long. For example, disruption to shipping routes through the Red Sea forced procurement teams across manufacturing, retail and consumer goods to reassess suppliers, transport routes and inventory strategies within days. Organisations with access to continuously updated intelligence were better positioned to identify emerging risks and alternative sourcing options. Incidents such as these all contribute to supplier landscapes, logistics routes and risks evolving.

Category managers are expected to oversee a growing number of categories, yet few have the bandwidth to analyse each category in sufficient depth. In fact, less than 20% of available procurement data is actually used to make decisions, according to McKinsey. Procurement teams are facing several competing priorities and therefore focus on their largest or high-risk categories, while maintaining the others more reactively. Decisions today are increasingly being made using incomplete or outdated intelligence, not due to a lack of capability, but instead a resourcing challenge and an inability to keep pace with the speed and complexity of the market.

The challenge of stale and siloed intelligence

An additional challenge procurement teams face is that valuable knowledge often lives outside of formal systems – in siloed software or individuals’ heads.

Knowledge such as supplier relationships, negotiation history and category-specific insights can often be found stored in emails and spreadsheets, rather than structured repositories that the whole team can access. Additionally, knowledge from individual experiences is often stored in that person’s head. As they don’t have the time to note everything down, or didn’t realise at the time how valuable the information would be in future, years of commercial intelligence can disappear when experienced employees move on or retire. The organisation may hold a goldmine of intelligence within siloed software and managers’ heads, but much is often lost, limiting both responsiveness and long-term value creation.

From static to continuous intelligence

Traditional approaches to category intelligence are becoming increasingly difficult to sustain in today’s fast-moving and volatile markets. Alongside the limited capacity of category managers, intelligence is overlooked, outdated or lost altogether.

While some in procurement have invested in AI-powered tools and analytics platforms, the issue remains that they are still relying on periodic analysis and disconnected data sources. They may be able to access more data, but not necessarily more intelligence. Agentic AI offers a fundamentally different approach. Rather than producing one-off analysis, AI agents continuously monitor market developments, synthesise external signals with internal organisational knowledge and surface relevant changes as they happen.

This results in procurement teams having access to live, structured intelligence that evolves alongside changing market conditions and can be shared, searched and reused across the entire organisation. Additionally, this allows far more categories to be monitored continuously, meaning that intelligence becomes comprehensive. Equally important is how this capability is now being delivered. Increasingly, organisations can now access agentic AI through subscription-based platforms, allowing them to adopt continuously updated intelligence without having to embark on complex development programmes.

Conclusion

The industry is now entering a new phase which will likely be defined not by the quantity of information it generates, but rather how effectively it can enable businesses to respond to changing markets in real time. Agentic AI will increasingly be measured and valued by its ability to deliver continuously updated, structured intelligence. By helping teams make better decisions faster, this is clearly where AI begins to move beyond the hype and deliver tangible operational benefits.

About the Author:
An Electrical Engineer from IIT Madras with an MBA from IIM Calcutta, Madhukar is a third-time serial entrepreneur and technology leader. His journey spans senior leadership roles at Unilever and Infosys, and founding successful ventures including Trryst and Vusiem. He is passionate about building scalable AI-led products that solve complex human and business problems.

Subscribe to Industry Today