Why Optimising AI for Speed Alone is a Race to the Bottom
Organisations are racing to become more efficient using AI, without realising that once every competitor has done so, efficiency stops being an advantage. It just resets the baseline.
For instance, a company automates ticket handling with an AI agent to accelerate case assignment, so cases reach the right team faster, even as ticket volumes rise. In parallel, it also builds an agent to automate campaign reporting. Spend, click-through and conversion rates from every channel are collected as they land, so the marketing team can optimise spend in real time.
Both make the organisation faster.
The problem is that when everyone gets more efficient doing the same thing, the whole market gets faster together, and nobody actually gets ahead of the rest. Gartner reports that finance AI portfolios often focus on use cases that enhance individual productivity or streamline processes and that once a task becomes faster or lighter touch, the benefit often plateaus unless it changes a wider business decision.
The Microsoft Work Trend Index 2026 found that organisational factors account for roughly twice as much of the reported AI impact as individual behaviour does, which suggests that focusing on efficiency can be a mistake.
What is being missed?
Businesses following a broader AI strategy that links AI to the data ecosystem are the ones that aren’t just working faster. They’re also working smarter, seizing the advantage and maintaining it. As we’ve argued before, AI should be driving growth, not just efficiency, and organisations need to look beyond their own systems.
When AI agents are built on first-party data, such as ticket volumes and resolution times, they can only optimise what’s already been measured, and therefore a CX team may fail to identify the underlying causes behind the figures. AI agents that combine data beyond an organisation’s own systems can explain rather than just report.
If the organisation that automated ticket handling and campaign reporting had built AI agents with access to the wider data ecosystem, it would have gained an advantage competitors could not easily copy, grounded as it is in something more than the organisation’s own data.
That something more is market research. It comes from in-house research teams that run field surveys and analyse trends, or, as is often the case, from external agencies, which provide one-off studies or subscription-based products, like category trackers and omnibus surveys.
From reporting to driving growth
Applied to earlier scenarios, incorporating market research would make the organisation faster and smarter. An AI agent that draws on research data can capture the issue and how it was resolved. It can also connect frequent complaints to shifts in the category, such as a new product entry or a price change, as they emerge.
Automating campaign tracking is equally transformed by going beyond internal data. An AI agent cross-references the tracker with category research revealing that a dip isn’t really about awareness, but a signal that a competitor’s new proposition is reshaping what people value, something the tracker numbers alone wouldn’t explain for another quarter.
Increasing efficiency with AI agents won’t enable an organisation to do more than just keep pace with an increasingly AI-driven world. To drive growth, agents need to combine first-party data with market research, so they can explain findings and not just report them. That also means basing decisions on data a competitor can't easily replicate. Combining your own data with research is what turns AI into a growth lever.
Ipsos Data Labs
At Ipsos Data Labs, we help organisations build AI roadmaps grounded in the data ecosystems that drive success. That means combining first-party data with the market research that explains it, focusing on the opportunities most likely to help businesses grow, and helping them move from ambition to execution with confidence and clarity.
