What Hundreds of Conversations With UK Service Businesses Reveal About a Hidden Barrier to Operational AI

Businesses are racing to adopt AI, but years of solving individual operational problems with separate software may have created the very fragmentation now making operational AI harder to implement effectively.

CQ Business Management Software has spent years holding detailed discovery and software demonstration conversations with hundreds of UK service businesses, understanding how their operations work and where existing processes and technology create problems.

These conversations examine how information moves through a business — from sales and quoting through projects, scheduling and communications to finance and invoicing — giving CQ a detailed view of the operational workflows businesses use every day.

Across those conversations, the same pattern has repeatedly emerged.

As business technology developed, software was largely built to solve individual problems. Businesses adopted CRM systems to manage sales, project management software to manage projects, accounting software for finance, scheduling tools to organise teams and cloud storage for documents.

Each solved a genuine problem and helped businesses become more digital.

But as businesses adopted more individual software, their information became spread across more places. Customer information might sit in a CRM, quotes somewhere else, projects in another system, schedules elsewhere, financial information in accounting software, documents in cloud folders and important updates buried in emails, messages or simply held in employees' heads.

The individual problems had been digitised, but the business had not been brought together.

For years, people have made that model work. Employees know where to find information, manually move it between systems, update colleagues, chase missing information and combine what they know with what exists in different software to understand what is actually happening.

AI changes that equation.

Consider a relatively ordinary question from a business owner:

“Why is this project running late, are we still making money on it, and what needs to happen next?”

A member of staff may be able to answer because they know where to look. They can check the original quote, review the project, look at the schedule, find the latest customer communication, check the hours worked, speak to a colleague and retrieve the financial information.

The employee pieces those different sources together to create the complete picture.

For AI to reliably answer the same question, it needs access to sufficiently connected, current and trustworthy operational information — and the context required to understand how those different pieces relate to one another.

Marc Mazure, founder of CQ Business Management Software, said:

“After speaking to hundreds of service businesses and looking in depth at how their operations work, I've come to believe we're approaching AI in the wrong order.”

“Businesses are rushing to implement AI because everyone feels they need to have it. But if your sales information is in one system, projects in another, scheduling somewhere else, accounts somewhere else again, with spreadsheets, emails and staff knowledge filling the gaps, simply adding AI doesn't automatically create the operational context it needs.”

“Before asking what AI they should implement, businesses need to ask whether their operations are actually ready for AI. Get the operational foundation right first, create a clear flow of information through the business, and then AI can be applied to real operational problems with the context it needs to genuinely make a difference.”

AI adoption is moving faster than operational readiness

The rush towards AI is already well underway, but growing adoption does not necessarily mean businesses are ready to use it effectively across their operations.

A government-published independent AI Adoption Plan for the Professional and Business Services sector, published on 8 June 2026, states that three-quarters of firms are not yet ready on core enablers such as data, orchestration and monitoring, while 70% report limited progress on process redesign. The report was produced by Shaheen Sayed, AI Champion for the Professional and Business Services sector.

The findings relate specifically to professional and business services and are not representative of all UK service businesses, but they highlight an important distinction between employees using AI to perform existing tasks faster and businesses redesigning how work is carried out around AI.

The report itself describes a gap between bottom-up AI adoption and firm-wide transformation, with employees increasingly using AI at task level while organisational change struggles to keep pace.

That distinction becomes particularly important as businesses move beyond isolated uses of the technology.

Asking AI to draft an email, summarise a document or create content requires relatively limited access to the wider operation. Asking it why a project is running late, whether a job remains profitable, who should be rescheduled or whether an invoice should be raised is very different.

To answer operational questions reliably, AI may need information from sales, quoting, projects, scheduling, communications, workforce management and finance at the same time.

Those functions do not necessarily have to exist within one piece of software. But the information and workflows between them need to be sufficiently connected, accessible and reliable for AI to understand the operational context.

The more businesses expect AI to understand and act across their operations, the more important the operational foundation underneath it becomes.

Marc Mazure said:

“AI can be incredibly powerful, but businesses need to start with the problem they're trying to solve. If solving that problem requires information from five different parts of the business and those five parts don't properly connect, adding AI doesn't remove the underlying problem.”

“That's why I think there's too much focus at the moment on implementing AI and not enough on getting businesses ready for it. The technology sitting on top can only be as useful as the operational information and processes underneath it allow it to be.”

From AI assistance to operational AI

Much of today's business use of AI still happens within individual tasks. An employee provides the information or prompt, asks AI to draft, summarise, analyse or create something, and then decides what happens next.

The human is still providing much of the context and judgement.

The challenge changes as AI moves from assisting with individual tasks towards understanding and eventually taking action across business operations.

Rescheduling a job, allocating staff, notifying a customer, identifying a delay, approving a change or triggering an invoice requires more than access to an isolated piece of information.

AI needs to understand what has already happened, what should happen next, which information is current, how different parts of the operation relate to each other and what effect an action in one area may have elsewhere.

This is why adding AI to individual pieces of software does not necessarily solve the wider operational challenge.

Businesses can connect separate systems through integrations, APIs and other technology, or bring more of their operations into unified platforms. The architecture can vary. The underlying requirement remains the same: AI needs reliable access to the operational context required to understand the problem it is being asked to solve.

Marc Mazure said:

“The mistake is thinking AI itself is the solution. It isn't. AI is an incredibly powerful tool, but first you need to give it an operation it can actually understand.”

“At the moment, people fill the gaps. They know that something happened in an email, something else is in the CRM, another part sits with accounts and somebody on the team knows what happened yesterday. As AI begins to take on more operational responsibility, we can't expect it to rely on that same unwritten knowledge.”

“If we eventually want AI not just to answer questions but to take meaningful action across a business, we have to give it the operational context to do that reliably. That's why getting the foundation right comes before operational AI, not afterwards.”

Getting the foundation right first

Despite the pressure on software companies to rapidly add AI features, CQ has deliberately focused first on building the connected operational foundation that it believes meaningful AI will ultimately depend on.

Rather than asking where AI can be added to the software, the company believes the more important question is what real operational problems AI can solve — and whether the information and workflows required to solve them are ready for it.

Marc Mazure said:

“We could add AI features to CQ tomorrow and put an AI badge on the website, but that isn't innovation if we're not solving a real problem.”

“I think the potential of AI in business is enormous. That's exactly why we're taking the opposite approach to rushing it in. We've concentrated first on building the operational foundation — because when we introduce AI, I want it to understand enough of the business to genuinely change how people work, not simply perform an isolated task that happens to have AI attached to it.”

“The businesses that ultimately get the greatest value from AI may not be the ones that adopt it first. They may be the ones that make sure their operations are ready for it first.”

The question for businesses may therefore be changing from “How do we implement AI?” to “Are our operations ready for AI?”

Notes to Editors

Basis of CQ's observations

CQ's observations are based on hundreds of detailed discovery and software demonstration conversations with UK service businesses conducted over several years.

As part of these conversations, businesses discuss their existing operational workflows, software and processes and the problems they are seeking to solve. Discussions can cover how work moves from sales and quoting through project or job management, scheduling, communications, workforce management, financial visibility and invoicing.

The observations presented in this release are qualitative operational observations based on recurring patterns identified through those conversations. They were not collected as part of a designed research study and are not presented as a statistically representative survey of UK businesses.

Businesses engaging with CQ are self-selecting and may be more likely than the wider business population to be experiencing operational or software-related challenges.

CQ uses these observations to provide qualitative insight into how and why operational workflow problems occur, rather than to make statistical claims about how prevalent those problems are across the wider UK market.

External source

AI Adoption Plan: Professional and Business Services, published 8 June 2026. Independent report by Shaheen Sayed, AI Champion for the Professional and Business Services sector, published by the Department for Science, Innovation and Technology on GOV.UK.

View the Professional and Business Services AI Adoption Plan on GOV.UK

About CQ Business Management Software

CQ Business Management Software is a UK business operations platform designed to bring the different areas of a service business together within one connected system.

The platform combines CRM and sales, estimation and quoting, project and job management, scheduling and field operations, team management, communications and financial visibility, helping businesses reduce fragmented processes and gain greater visibility across their operations.

Media enquiries

Marc Mazure
Founder & CEO
CQ Business Management Software
[email protected]

0204 548 4000

www.cq-business-management-software.com

Marc Mazure is available for interview and further commentary on operational AI readiness, fragmented business systems, service-business technology and the operational challenges businesses face as they adopt AI.