Workday Q&A: Leaders Must Fix ‘Architectural’ AI Problem

In May, Workday released research revealing that UK employees lose close to a full work day each week in time spent switching between disconnected AI tools and systems. 

The company’s Copy/Paste Economy report finds that, while employees do see AI as a useful tool in the workplace, 62% of workers spent at least half their time translating and coordinating between systems and teams instead of creating value for their organisation, and more than 80% of respondents reported spending “significant” time moving information across conflicting tools. 

Discussing these findings with HR Chief in more detail is Daniel Pell, Vice President and Manager of UKI Workday.

Could you break down what The Copy/Paste Economy means in practice, and how it manifests in the daily workflows of UK professionals?

It’s the part of work that doesn't show up on an organisation chart but dominates people's calendars: chasing approvals, reconciling reports from different systems and manually moving information from one tool to another. 

Our research shows more than eight in 10 employees spend significant time coordinating across teams, moving information between tools and reconciling conflicting data. In the UK specifically, 60% of employees say their days are busy but unproductive – well above the global average.

In practice, that looks like HR leaders pulling data from three systems to answer a simple workforce question, finance teams spending evenings lining up numbers from spreadsheets and ERPs or managers copying context from email into AI tools just to get a sensible answer.

People tell us they like their jobs and feel connected to their organisation's goals. They're just spending far too much of the day acting as the glue between systems, instead of doing the work they were hired for.

Workday's research finds that UK employees lose nearly a full work day each week managing disconnected AI tools. When looking at the data, were you surprised by the sheer volume of time being swallowed up by tech intended to make work more efficient?

The UK headline is stark, but unfortunately, it doesn't surprise me. I regularly meet organisations with hundreds of separate HR and finance applications that were never designed to work together.

What did strike me was the contrast. Employees are highly engaged and optimistic about AI, yet they're sacrificing nearly a full work day a week just to make the tech talk to itself. For HR leaders, that's a clear signal: we don't have a motivation problem and we don't have an AI-awareness problem. We have an architectural problem – and one that leaders can fix.

You said that too many employees are serving as the human middleware between disconnected AI systems. What does being ‘human middleware’ look like on a practical level, and why is this an unsustainable model for enterprise growth?

Being human middleware is what happens when people are asked to bridge gaps the systems should be handling. Our data shows 77% of employees are reconciling conflicting data from different tools, and 70% are re-entering the same information into multiple systems. 

Entire roles, especially in IT, are now defined by this invisible integration work – a quarter of IT professionals say these efforts effectively define their workday. It's unsustainable because it scales linearly with complexity. Every new tool, every new AI assistant, adds another set of copy-paste handoffs for a human to manage. 

From a growth perspective, you can't keep hiring people just to move data around and double-check outputs. From a people perspective, it leads to stress and burnout, even when engagement scores look healthy on the surface.

For HR and IT, the only viable model is to push that integration into the systems themselves. This means AI and workflow engines become the glue, and employees can focus on judgement, creativity and relationships.

Over half of UK employees say that AI reduces individual task times, yet total time saved isn’t budging. Why are isolated micro-efficiencies failing to translate into macro-level productivity gains?

Most organisations have started their AI journeys at the edge: tools that help draft an email, summarise a document or generate a first version of a slide. Over three quarters of employees already rely on AI to remove friction in these individual tasks, and in the UK more than half say AI is already reducing their task times.

The problem is that the work between those tasks is still fragmented. Every generative draft still has to be checked, copied into the right system, reconciled against the system of record, and pushed through a manual approval chain. We’ve made the micro-tasks faster, but left the end- to-end process unchanged. So the productivity we gain at the edge is taxed away by coordination in the middle.

Our data shows that when AI is embedded directly into core systems and workflows, 60% of employees report time savings of 25% or more. In organisations where AI sits outside core systems, less than a quarter see that level of benefit. That’s the difference between speeding up busywork and actually redesigning how work gets done.

How could the Copy/Paste Economy worsen if businesses don't fix their underlying data architecture first?

As more teams adopt specialised AI apps without a unified data foundation, employees will spend even more time manually feeding each tool with context, translating outputs back into core systems and resolving contradictions between what different AIs tell them. 

That increases risk as well as inefficiency – because AI operating on partial, inconsistent data is more likely to hallucinate or conflict with the policies and approval chains encoded in your core platforms. Over time, you end up with a widening gap between where decisions are proposed – in peripheral AI tools – and where they're executed, in HR and finance systems. That isn't a recipe for compliant, scalable AI deployment. 

The only sustainable path is to fix the foundation – clean, connected, trusted data – and then let agents operate inside that environment, not bolted on around it.

If employees are freed from acting as “middleware,” their roles naturally shift toward oversight and strategic judgment.  What steps should HR and IT leaders take together to prepare the workforce for this skills shift?

As we move integration into the systems and give AI agents clear ownership of routine steps – monitoring, routing, first drafts – human roles naturally skew toward orchestration, oversight and problem-solving. That has big implications for skills.

Internally, we're already hiring differently. In sales and marketing, for example, we're actively looking for people with stronger technical fluency who can understand and adapt how we operate with these tools, not just execute traditional tasks. 

We're also putting a premium on critical thinkers and problem solvers – people who can take data, tools and workflows and assemble them into outcomes for different clients and scenarios.

For HR and IT leaders, the practical steps are clear: build joint programmes to upskill employees on how AI actually works in your environment; redesign roles and performance metrics to value systems thinking and experimentation and create space for teams to build and share their own workflows and agents.

When people are empowered to shape how AI is used, they stop feeling like middleware and start behaving like designers of the operating model.