Automation in UK workplaces: What's really holding back productivity?

Gartner analysts project that global IT spending will reach US$6.15 trillion in 2026. The proportion of that spend now going towards automation tools is also increasing. Research from the Office for National Statistics reveals that around 25% of UK businesses reported using some form of AI at the end of 2025, an increase of 15% since late 2023. 

Despite the widespread investment in technology, including software, workplace productivity isn’t seeing similar gains. In fact, the Office for National Statistics reported that output per hour worked in the UK was just 0.4% higher in the first quarter of 2026 than a year earlier, and output per worker actually decreased.

So, if AI and automation are expected to boost productivity, why aren’t we seeing that reflected in the statistics? The numbers suggest that something is wrong. That ‘something’ is often unreliable data, caused by fragmented tech stacks. In this article, we'll explain how disconnected systems hold your business back, and how unifying your data can deliver the productivity gains automation promises.

Key takeaways

• UK workplace productivity has barely moved, despite record spending on software and automation. The problem isn't a lack of tools; it's what those tools are built on.

• Many businesses are running automation on top of fragmented, disconnected systems. The adage ‘garbage in, garbage out’ is a good way to explain the lacklustre results.

• The solution is to unify the underlying data first. By adopting a single source of truth for data, businesses can finally get more productivity from automation.

The software buying spree

Gartner expects global software spending to reach US$1.44 trillion in 2026, a 15.1% increase from last year. For Europe (including the UK), software spending will grow from US$290.2 billion in 2025 to US$335.4 billion this year, an increase of 15.6%.

It’s clear that investment in software continues to increase, but to understand how it’s helping (or not helping), we need to look at what businesses are actually buying. Our own research has shown that 36% of HR teams are using seven or more applications to perform their tasks, suggesting that many businesses have simply been adding more tools to solve specific problems.

From both a human and an automation perspective, this is where many businesses are failing to see the productivity gains they expected. Let’s look at how this starts and the impacts it has.

The wrong starting point: Fragmented systems and data

Our examples above focused on HR teams, but the problem isn’t confined to HR alone. According to Gartner’s Magic Quadrant for SaaS Management Platforms, the average organisation uses more than 125 SaaS applications. here's a name for this: software sprawl.

A business using 125 applications also has 125 separate data entry points, creating challenges for efficiency, accuracy, and the way automation uses that information.

Why software sprawl happens

For most businesses, software sprawl is an unintentional and gradual process, with applications being added over time as new requirements emerge. For example, an HR department may start with one software package to manage payroll, then add another to handle recruitment, and a third to look after international hiring.

Rather than being a company’s initial plan, it reflects the different stages of growth they’ve gone through. An early-stage business may be fine with one or two applications, but as that business grows and faces new challenges, additional software solutions are needed to solve them.

Finally, software purchases can also be driven by wider business trends. For example, AI-based tools have been widely promoted for their ability to automate repetitive tasks and improve efficiency. Many businesses purchase these tools hoping that they’ll boost productivity.

However, if they do that without a solid foundation, they’re only adding one more tool to an already overloaded tech stack.

The cost of fragmented systems

Fragmented software can have both a direct and an indirect impact on employee productivity. According to Harvard Business Review, the average employee switches between different apps and websites up to 1,200 times per day. Every switch costs time to refocus, and it adds up.

Our own research reveals that 53% of HR teams and 68% of finance teams use seven different applications to perform their duties each month. At the same time, 96.9% of HR leaders and 97.6% of finance leaders are in favour of consolidating their existing tools, pointing to growing frustration with this type of setup.

As a result, many businesses are investing in automation, system integration, or both, hoping that this will solve their problems.

Garbage in, garbage out: Automation is only as good as the data it’s using

As we’ve discussed, approximately 25% of UK businesses reported using AI in some form as of late 2025, and that number increases to 44% among businesses with 250 or more employees.

Our own research in the HR space shows more widespread adoption, with 88% of HR leaders saying they’re already using or testing AI in their workflows, and 67% rating themselves as intermediate or advanced users.

What many businesses fail to understand is that AI relies on the information it’s given. So, while it might analyse your information faster, if your data is inaccurate or unreliable, then the output will also be flawed. In effect, your staff will just have to go back and correct the mistakes, meaning there are few, if any, efficiency gains.

This is especially problematic when working with fragmented or disconnected systems, as the AI won’t know which source of information is correct. For example, if there is a separate record for an employee in HR, finance, and IT systems, an AI system won’t know which is the correct or most up-to-date version.

Simply put, adding AI on top of multiple, disconnected applications will only compound the problem.

System integration: Why it’s a band-aid solution

A common workaround for fragmented applications is the use of integration, which connects separate software platforms so they work as a more cohesive ecosystem. Businesses are often attracted to this approach because it allows them to continue using the software they’ve already purchased.

Here are some common challenges with this approach:

• Unreliable connections: Attempting to connect multiple software solutions from different vendors doesn’t always result in reliable data sharing, especially when legacy systems are involved.

• Accountability gaps: With multiple vendors involved, it’s not always clear who is responsible when a problem occurs. For example, if one platform is failing to integrate with others, is it the responsibility of the company producing that software or the company performing the integration?

• Ongoing maintenance: While the immediate solution may appear quick and easy, any new software will require additional integration. As systems continue to age, managing the whole network can become increasingly expensive. These rising costs are referred to as technical debt.

• AI challenges: Integrating separate software can continue to pose challenges for AI, especially if data between two platforms isn’t aligned. This is especially problematic for agentic AI, which acts autonomously and therefore requires highly accurate data to inform its decisions.

Data unification: Why it’s the way forward

Many people confuse the terms ‘data integration’ and ‘data unification’, but the two processes are fundamentally different.

What a unified platform does

A unified platform, also called an all-in-one system, is built from the ground up to handle multiple business functions. From our earlier example, that could mean having a single software package that handles your HR, payroll, training, recruitment, and international hiring processes.

In this scenario, there would be data on each employee related to how they were hired, what their pay is, what courses they’ve completed, and which IT systems they have access to. When all of this information is stored on a single platform, there’s no risk of duplicate or out-of-date information that’s common with fragmented systems.

This structure is often referred to as a ‘single source of truth’, and it’s increasingly common in areas such as human resources, customer relationship management, accounting, and project management.

How unified platforms improve productivity

Fragmented systems slow your employees down and starve AI of reliable data. A unified platform solves many of these challenges. In terms of your employees, you’ll be replacing multiple tools with a single user interface. This reduces the fatigue caused by switching between applications, simplifies the staff training process, and improves coordination between different departments.

When considering AI and automation, the single source of truth offered by unified platforms is where you’ll realise some of the greatest productivity gains. A centralised, single source of information reduces the likelihood of incorrect or contradictory data, meaning that AI will be delivering more accurate output. When your teams can trust what AI is delivering, they can focus their efforts on other tasks.

Signs your business could benefit from a unified system

Unified platforms are sometimes associated with enterprise-level requirements, rather than small or mid-sized businesses. In truth, the best systems cater to a variety of requirements, allowing you to add or remove functions as your needs change. The following table contains some common complaints from businesses of all sizes, along with how a unified platform can help:

Signs a unified platform can help

Implementing a unified platform

If you’ve decided to unify your tech stack, here’s how the transition works. One proven approach for this kind of transition is the strangler fig pattern. It involves building your new, unified system in parallel with your existing applications and moving functions across one at a time. Old systems are progressively retired once the transition has been tested and verified.

Before any migrations occur, however, you’ll need to reconcile the information that’s currently being stored on each platform. This includes identifying duplicate records, validating which is the correct version, and creating a master version to be imported into the new platform (sometimes called a golden record).

From there, you’ll need to decide on the order in which your existing functions will be moved across to the unified platform. A staged process allows you to confirm that each function works as expected, minimising disruption to your business.

The evidence for unified platforms

A study published by Harvard Business Review highlighted some of the real-world benefits of transitioning to a unified data platform. In the report, one finance team achieved a 400% increase in time available to focus on strategic work compared to the legacy, siloed systems it had been using.

In terms of increasing the value of data, including through automation, McKinsey also identified a need for a “holistic approach to data projects”, including “Platforms for managing master data” to “provide a base for global, secure, scalable, and resilient architecture”.

Focusing specifically on AI, Gartner states, “Realizing AI ambition at scale requires new, deeply integrated engineering practices. Siloed practices for data, AI, context and software engineering will fail to realize an AI-first ambition.”

Rippling's unified HR platform: a real-world example

Rippling runs HR, IT, and finance from one underlying data model, so automation and analytics work from a single accurate record of your workforce.

A Total Economic Impact (TEI) study carried out by Forrester Consulting found that a composite organisation using Rippling achieved 137% ROI over three years, with HR, payroll, and finance operational efficiency up 42% and IT efficiency up 35%. Interviews with users revealed that most efficiency gains were achieved through a reduction in manual processes and more effective use of workforce analytics.

How Open Data Institute achieved a 65% monthly cost saving

The UK’s Open Data Institute (ODI) was using five separate platforms to run its HR department, requiring manual data entry for each system, as well as five different vendor maintenance charges.

“We had one platform doing payroll, another for surveys, another for expenses, another for recruitment and then our HRIS system”, head of IT, Kwaku, explains. “The most concerning part was the amount of data duplication—there was so much data all over the place.”

Rippling was engaged to consolidate these systems into a centralised platform, with automatic data synchronisation across each function. As a result of the consolidation, the non-profit achieved a 65% reduction in monthly costs while simultaneously improving efficiency and productivity.

Kwaku says, “Going to one platform was huge, not just for the HR team, but for all employees… Rippling takes care of the mundane tasks in your day, so you can focus on the work that actually drives productivity.”

Read the full ODI case study.

Looking to the future

Many see stagnating productivity figures across the UK and conclude that AI is a failed experiment, without identifying the underlying reasons that automation is unsuccessful.

The solution isn’t to buy more tools; it’s to create a data infrastructure that allows those tools to work. That’s where unified platforms come into their own.

Rippling brings all of your data into a single platform and allows your business to manage HR, finance, and IT in one place. And with Rippling AI, you not only get more reliable automations, but you also get insights from a system that deeply understands your organisation, can answer specific employee questions, and produce in-depth reports using real-time data. 

If you want to learn how Rippling’s platform can help your business realise the true benefits of AI, request a demo from our team.  

FAQs

What is workplace productivity, and why has it stagnated in the UK?

Workplace productivity is essentially output per hour or per employee, or how much a business gets done relative to the time and people it puts in. In the UK, growth has been sluggish since the 2008 financial crisis, and recent data shows it barely moving despite record spending on business software and automation. The problem is the poor underlying data that fragmented systems rely on.

Is automation actually making productivity worse?

Automation isn't what makes productivity worse on its own. It executes whatever data and process it's given, so if that data is fragmented or inconsistent across systems, automation can make errors happen faster and more often rather than fixing them. The solution is to address the underlying data, not to use less automation.

What's the difference between integrating software and unifying it?

Integration connects separate systems through APIs or middleware so they can exchange updates, but each system still holds its own version of the data, which can drift out of sync. Unification means the systems share a single source of truth from the start, so there's no separate data to be integrated.

How do I know if my business has an automation problem or an alignment problem?

It's an alignment problem, not an automation problem, if different teams each hold their own version of the same record. That could be a customer, an order, or an employee. If access or permissions don't update the moment something changes, or if a routine change requires several people to manually update different systems by hand, you have a data alignment issue.

What should I look for when fixing a fragmented tech stack?

Start with an audit of your current tools to find overlap, then look for a platform built on one underlying data model rather than one that connects separate products through integrations. The goal is a setup where every department reads from the same source of truth, not just passes updates to each other.