90% Use AI for Recruiting — But Fewer Than 5% See Transformational Results. The Problem Is Process, Not Technology.
The News
More than 90% of companies surveyed by ManpowerGroup Talent Solutions said they are using artificial intelligence tools for recruiting. But fewer than 5% reported seeing 'transformational' outcomes on any key metrics, according to research released Tuesday by ManpowerGroup and research firm Everest Group. The survey of 80 C-suite, CHRO, and senior talent acquisition leaders across the US and UK paints a picture of widespread AI deployment delivering incremental rather than structural change.
Close to 4 in 10 organisations said they saw 'significant impact' on operational efficiency — faster sourcing, quicker screening — but improvements to decision quality and workforce agility were described as 'limited.' AI is predominantly deployed for sourcing, resume screening, and candidate engagement, the report found. The problem, researchers concluded, is not access to AI tools but how talent operations are designed around them.
'What the research makes clear is that the constraint is no longer access to AI tools. It is how talent operations are designed around them,' said Caroline Pfeiffer Marinho, global senior vice president of Talent Solutions RPO and Right Management. Sailesh Hota, vice president of Everest Group, added: 'As AI becomes embedded into workflows and decisions, organisations are discovering that adapting workforce models, leadership practices, and operating structures is proving equally important.' The report noted that fragmented systems, isolated tools, and siloed data are the primary blockers — and that more than half of those surveyed said candidate use of AI has made it harder to assess applicant capability.
Why It Matters
This is the most important piece of TA technology research published this quarter, and not because it reveals something surprising. It is important because it names the actual problem. For two years, the TA vendor market has sold a story that AI adoption equals AI transformation. The ManpowerGroup-Everest Group data says otherwise: 90% adoption, fewer than 5% transformational. The gap between those two numbers is where most TA leaders actually live.
The finding that fragmented systems and siloed data are the primary blockers should refocus procurement conversations. If your organisation is considering adding another AI sourcing tool, AI screening tool, or AI scheduling tool to a stack that is already fragmented, you are not buying transformation — you are buying another isolated capability that will deliver the same sub-5% result. The research suggests the higher-return investment is in process redesign, data integration, and operating model change — the unglamorous work that sits between the AI tools and the outcomes they are supposed to produce.
Taking Action
All action is situated in your corporate context. These are starting points, not prescriptions — choose what fits.
Connect
Who to talk with in your enterprise
Your head of TA operations — the report's core finding — that process design matters more than tool access — maps directly to your remit. Ask whether your AI procurement process includes an operating-model review, or whether AI tools are purchased, deployed, and expected to deliver results inside existing (often broken) workflows.
Your chief information officer or head of HR technology — 'fragmented systems, isolated tools and siloed data' is a technology architecture problem before it is a TA problem. Share the ManpowerGroup findings and ask for a data-integration audit of your TA stack. How many of your AI tools can actually talk to each other?
Your procurement lead — fewer than 5% transformational is a benchmark worth using in vendor conversations. When the next AI vendor pitches 'transformational results,' ask whether their reference customers achieved transformational outcomes inside a fragmented stack or only after process redesign — and whether the vendor can support the redesign, not just the software.
Inspect
What to take a closer look at
Map your current TA technology stack on a single page. For each tool, document: (1) what data it produces, (2) what data it consumes from other tools, and (3) whether the data flows are automated or manual. Count the number of tools that operate in isolation (no automated data in or out). If more than two of your AI tools are islands, the ManpowerGroup research suggests you are on the 90%-adoption, sub-5%-transformation path. Specific tool to look at: your ATS's integration marketplace or API documentation — how many of the AI tools in your stack have native, bi-directional integrations with your ATS, and how many rely on CSV exports and manual uploads?
Reflect
What to think about more deeply
The most interesting question isn't 'should we buy more AI tools for recruiting?' — the ManpowerGroup data suggests that answer, for most organisations, is no. The interesting question is 'what would have to be true about your recruiting operations for AI to deliver more than incremental efficiency gains?' If your process is already fragmented, adding AI to it is like putting a faster engine in a car with a broken transmission — you will make more noise but not go much further. The organisations that break through to the sub-5% club of transformational outcomes will not be the ones that bought the most AI. They will be the ones that redesigned the process around the AI. Is your organisation willing to invest in the redesign, or is it shopping for another tool?