A list of
100 problems.
This didn't begin with a client brief or a workplace problem. It began with a list — a personal exercise in systematically identifying where industries are slow, inefficient, or unnecessarily dangerous before deciding which problem is worth solving.
Approximately 100 operational problems were documented. The one that stood out was water sampling.
Utility companies needed regular water quality data from reservoirs. The standard method involved a three-person crew, a boat, days of planning, access coordination, risk assessments, and approximately £200 per hour in operational cost. All of that to collect a water sample.
"The logistics were slow. The costs were significant. And every time someone went out onto the water, there was a health and safety exposure that didn't need to exist."
Three problems.
One assumption.
Cost — £200/hour, every visit Industry benchmark boat hire for reservoir sampling ran at approximately £200 per hour. A three-person crew with specialised vessels, insurance, and access logistics made every sampling visit a significant operational expense — regardless of how much data was actually needed.
Mobilisation — days, not hours Getting a crewed boat to a reservoir required advance planning, access permissions, crew coordination, and weather assessment. A simple sampling request could sit in a queue for days before deployment. Reactive responses to water quality events were effectively impossible at speed.
Health & Safety — unavoidable exposure Every crewed sampling visit placed people on open water. Regardless of precautions taken, the exposure existed. Falls, vessel incidents, access hazards — the risk wasn't catastrophic, but it was present on every single visit, by design.
All three problems shared the same root assumption: that a person needed to be there to collect the sample. Remove that assumption and the problem looked completely different.
Built from
scratch.
The question was simple: could the human element be removed from the water entirely?
The answer took 2.5 years to prove — starting with trial and error and independent research into how to collect an uncontaminated water sample remotely, then building the platform to do it.
Three unmanned vehicles were developed and equipped, each designed for different scenarios, water depths, and weather conditions:
Aerial drone Purchased and equipped with custom sampling hardware. Capable of reaching open water locations rapidly, collecting surface and near-surface samples without vessel deployment.
Submersible ROV — self-built A remotely operated underwater vehicle, built from components, capable of collecting samples at depth — accessing water quality layers that surface methods couldn't reach at all.
Remote control surface vessel For conditions where aerial deployment wasn't optimal. Equipped with sampling equipment for mid-water collection with weather robustness the drone couldn't match.
This wasn't a software solution or a process tweak. It was a ground-up methodology built for a problem nobody had formally solved before.
UK's first.
UKAS accredited.
Working in partnership with a major UK utility company, the methodology was submitted for formal accreditation through UKAS — the United Kingdom Accreditation Service.
UKAS assessed the process in action, focusing primarily on sample integrity — specifically ensuring no cross-contamination occurred during collection or the logistical chain that followed. Water sampling standards are exacting. The accreditation process made no exceptions.
The UK's first UKAS-accredited unmanned aerial and subsurface water sampling methodology. The utility company holds the accreditation certificate. The methodology is the one that earned it.
The numbers
changed.
eliminated entirely
per service request
on the water
A single operator with an unmanned platform could mobilise within hours of a service request. The previous model — three crew, a boat, access coordination — took days.
On health and safety, the outcome was categorical. Nobody went out onto the water. The H&S exposure that existed on every crewed visit was eliminated at the point of methodology — not managed after the fact, removed entirely.
On sample quality, the unmanned platform revealed something the traditional method couldn't. By accessing different depths within the reservoir, the methodology identified variation in water quality layers — data that directly informs treatment intensity. Better data at source means less strenuous treatment downstream.
The assumption
was the
constraint.
The cost wasn't the water. It wasn't the regulation. It wasn't even the technology — the components already existed.
It was the assumption that a person needed to be there. That assumption had never been challenged because nobody had bothered to challenge it. The industry had simply absorbed the cost, the delay, and the risk as unavoidable features of the work.
The constraint wasn't the work. It was the assumption the work was built on.
That's usually where the real friction is.