Summary (Key Takeaways)
- “Human middleware” is when skilled staff spend their day moving data between disconnected systems so AI tools can function, copying, checking and correcting by hand.
- For regulated firms, this is not only a productivity drain; it is a governance and compliance risk, because every manual handoff of client data is unlogged and unaudited.
- Firms can feel more productive on the surface while effort quietly shifts from advising clients to coordinating software.
- Bolt-on AI tools tend to arrive faster than the systems beneath them, leaving people to bridge the gaps manually.
- The fix is a governance-first review of how information moves through the firm, not adding more isolated tools.
- Warning signs include constant app-switching, manual data reconciliation, and staff routinely correcting AI outputs before they can be trusted.
The Promise Was Simple. The Reality Is Messier
Artificial intelligence arrived in professional services with a clear pitch: remove the repetitive, low-value tasks and give skilled people back their time. For a mortgage brokerage, an accountancy practice, a firm of solicitors or an IFA, that promise is genuinely appealing. Fee-earners are expensive, client expectations are high, and the regulatory burden only grows heavier each year.
Yet in a surprising number of firms, something quite different has taken hold. Rather than removing work, AI has added a new layer of it, and that layer is often invisible on any dashboard or productivity report. Skilled staff are spending meaningful portions of their day moving information between systems simply so the technology can do its job. They copy notes from a case management system into an AI assistant, paste the result into an email, then reconcile the figures against a spreadsheet because nobody quite trusts the automated output. It feels busy. It even feels productive. But it is not the work the firm hired them to do.
What “Human Middleware” Actually Means
In technology, “middleware” is the software that sits between two systems and helps them exchange information. When people take on that role themselves, manually shuttling data so that disconnected tools appear to work together, they become human middleware. Once you recognise the pattern, you begin to see it throughout the firm.
An adviser downloads a client file from one platform because a second system cannot access it directly. A paralegal pastes sensitive personal details into an AI drafting tool, generates a letter, then copies the finished text into the document management system. A member of the finance team manually checks AI-produced figures against the source records because the automated reconciliation cannot be fully relied upon. Each of these actions is small. Collectively, they consume hours, and in a regulated environment they carry a weight that goes far beyond lost time.
Why This Is a Governance Problem, Not Just a Productivity One
In most businesses, human middleware is an efficiency nuisance. In a regulated firm, it is something more serious. Every time a member of staff manually copies client data between systems, that handoff typically happens outside your logging, your access controls and your audit trail. There is no record of who moved what, when, or why. If a regulator, an auditor or a professional indemnity insurer later asks you to demonstrate how a particular piece of client information was handled, “someone copied it into a chatbot and pasted the answer back” is not the answer you want to give.
This is where an unmanaged tangle of tools quietly becomes technical debt, the accumulated cost of shortcuts and disconnected systems that must eventually be paid back with interest. The more your people bridge gaps by hand, the more fragile and opaque your processes become, and the harder it is to prove control. We explore this in more detail in our article on how outdated and disconnected systems create hidden drag on a professional firm: the true business cost of technical debt.
The Productivity Illusion
The genuinely tricky part is that AI does help. Emails are drafted faster, research is summarised in moments, and long documents can be distilled into a readable brief. On any given day it can feel as though the firm is moving quicker than ever. But new admin has grown up around the tools, and much of the apparent speed is being reinvested into coordination rather than progress.
Consider a typical morning for a senior adviser. They summarise a client meeting with an AI assistant, a real time-saving. Then they copy that summary into the CRM, reformat it for a compliance note, paste relevant sections into a suitability report, and manually correct three details the AI got slightly wrong. The two minutes saved on drafting have been swallowed by twenty minutes of stitching. The firm feels productive. The adviser feels busy. And yet the actual advisory work, the judgement, the client relationship, the regulatory reasoning, has barely moved forward.
Why AI Tools Outpace the Systems Beneath Them
Part of the reason this happens is that AI features arrive far faster than the underlying systems can evolve to support them. A firm adds an AI assistant in one application, an automation in another, and an AI-powered feature in a third. Individually, each looks like an upgrade. Together, they form an archipelago of clever tools with no bridges between them, so your people become the bridges.
This is closely related to a wider governance blind spot: the growing number of AI capabilities operating inside a firm without clear ownership, oversight or a defined purpose. When tools are adopted piecemeal, nobody holds the full picture of what is running, what data it touches, or who is accountable if it goes wrong. We examine how this fragmentation creates oversight gaps at board level in our piece on the governance blind spots created by unmanaged AI agents.
The Warning Signs to Look For
Human middleware rarely announces itself. It builds up quietly, one workaround at a time. Here are the signals that it may already be established in your firm:
- Constant application-switching. Staff routinely bounce between three or four systems to complete a single task, with copy and paste as the connective tissue.
- Manual reconciliation. People double-check AI or automated outputs against source records because they do not fully trust them.
- Routine correction of outputs. AI results are “almost right”, so someone always edits them before they can be used, a hidden but recurring cost.
- Shadow data flows. Client information is being moved through tools that were never formally approved or risk-assessed.
- Undocumented handoffs. The path a piece of client data takes cannot be reconstructed from your logs.
Any one of these is worth investigating. Several together suggest your technology strategy is quietly working against you.
Who Approved All These Tools Anyway?
There is an uncomfortable question hiding beneath the surface of most AI sprawl: who actually decided to adopt each of these tools, and against what criteria? In many firms, individual teams or even individual employees have quietly introduced AI subscriptions and features to solve their own immediate problems, with no central review of cost, data handling or regulatory impact. That is how you end up with a dozen partially overlapping tools and a workforce acting as glue between them. We look at how unapproved and unmonitored AI spending creeps into professional firms, and how to bring it under control, in our article on who is really signing off AI-made purchases in your business.
A Better Approach: Start With How Information Moves
The instinct, when technology feels clunky, is to buy another tool. In practice, that usually makes the problem worse. The more productive path is to step back and look at how information actually flows through your firm before adding anything new.
Begin with a simple mapping exercise. For your most common processes, such as onboarding a client, producing a suitability report, or handling a case file, trace the journey of the data. Where does it originate? Which systems does it pass through? At which points does a human have to intervene to move, reformat or correct it? Those intervention points are your human-middleware hotspots, and they are where governance risk and wasted effort concentrate.
From there, a governance-first framework helps you decide what to do next:
- Where several tools overlap, reduce to the fewest that genuinely integrate, so data moves automatically rather than by hand.
- Connect the systems that must exchange information, replacing manual handoffs with logged, controlled ones.
- Assign clear ownership for every AI tool that touches client data, with defined purpose, access limits and an audit trail.
- Build in appropriate checks so that trusted, automated outputs no longer need to be manually reconciled every time.
The goal is not to slow AI adoption. It is to make sure each tool sits inside a controlled, auditable environment, so it reduces workload and demonstrable risk at the same time.
What Your People Should Be Doing Instead
Your fee-earners and specialists are the most valuable resource the firm has. Their time should go into exercising judgement, interpreting regulation, advising clients and building the relationships that underpin client confidence and retention. It should not be spent helping one piece of software talk to another.
When a firm removes human middleware, the benefits compound. Advisers reclaim hours of genuine advisory time. Compliance gains a clean, auditable trail of how client data is handled. The board gains confidence that the technology estate is understood and controlled rather than sprawling. And clients experience a firm that feels responsive and assured rather than one that is quietly drowning in admin.
Is It Time to Review Your Technology?
If your people are constantly switching between apps, correcting AI outputs, or manually stitching workflows together, that is a clear signal your technology strategy needs tightening. In a regulated firm, the stakes are higher than lost minutes. You are also carrying unseen governance and data-protection exposure with every manual handoff.
The reassuring news is that this is a solvable problem, and solving it improves both efficiency and compliance at once. At Absolutely PC, we help professional and regulated firms review how information moves through their business, remove the human-middleware bottlenecks, and put a governed, auditable technology strategy in place. If your team feels busy but progress is slow, it may be time to look beneath the surface. Get in touch. We would be glad to help.
What next?
One of my passions is helping businesses to succeed and if I can help you save some money as well – even better. You can fill out our contact form, phone us or click on the appointment button below and let’s start a conversation to see if I can help your business. Our guarantee:
- There are no hidden charges – this is a 100% free 15 minute consultation with no hidden charges.
- We will never spam you or sell on your contact details.
- We will treat your information with absolute confidentiality.
Other Posts for you to Enjoy
