How to Accelerate Human Data Collection in Biosensor Development

Developing a new biosensor can be an expensive and time-consuming journey. Many companies spend years developing custom sensors, electronics, software, manufacturing processes, and packaging before collecting the human data needed to determine whether their concept actually works in practice.

The problem with this approach is simple: what if the answer is "no"?

By the time that conclusion is reached, substantial amounts of time and money may already have been invested.

At Zimmer & Peacock (ZP), we often encourage clients to think differently. Rather than waiting until a complete product exists, it is often possible to generate meaningful human data much earlier in the development process.

In many cases, that early data can be obtained in months rather than years.

The Traditional Development Path

A common development strategy follows a familiar sequence:

Develop a sensor
Develop electronics
Develop software
Optimise performance
Freeze the design
Establish pilot manufacturing
Conduct validation studies
Generate human data

While this route is logical, it can also be extremely costly.

The challenge is that the most important question is often not answered until very late in the project:

Does the analyte provide useful information in the biological matrix of interest?

Many projects reach advanced stages of development before this question is fully resolved.

Why Human Data Matters

Investors, partners, customers, and regulatory teams ultimately want evidence.

They want to know:

Does the measurement correlate with the intended outcome?
Can the analyte be measured reliably?
Is the chosen biological sample useful?
Can the data support a meaningful application?

These questions cannot be answered through benchtop testing alone.

Eventually, human studies are required.

The question therefore becomes:

Can we obtain that data sooner?

A Different Approach

For many applications, the answer is yes.

Rather than developing a completely new sensor from scratch, it may be possible to leverage existing commercial sensor technology and use it as a research tool.

The key insight is that the sensing chemistry often already exists.

What may not exist is a validated solution for your specific application, sample type, or commercial opportunity.

Instead of building everything from the ground up, it can be advantageous to start with an existing sensor platform and focus on understanding whether useful information can be extracted from real-world samples.

The Role of Alternative Biological Matrices

One common challenge in biosensor development is moving from an established sample type to a different biological matrix.

A sensor might have been designed for one application, but a developer may wish to apply it elsewhere.

At first glance, this seems straightforward.

However, biological matrices behave differently.

Factors such as:

Analyte concentration
Interfering compounds
Viscosity
Sample composition
Signal intensity

can all significantly impact performance.

A sensor that performs well in one matrix may appear ineffective in another.

But appearances can be deceptive.

The Hidden Problem with Commercial Meters

Many commercially available biosensors are supplied with dedicated electronics and proprietary algorithms.

These algorithms have usually been optimised for the original intended application.

This becomes important when developers attempt to use the sensor in a different context.

The meter may provide a numerical output, but that output reflects assumptions built into the original algorithm.

Those assumptions may no longer be valid.

A common mistake is to conclude that the sensor itself does not work when, in reality, the issue lies in how the meter is processing the signal.

The chemistry may still be generating valuable information.

The algorithm simply isn't designed to extract it.

Why Raw Signals Matter

At Zimmer & Peacock, we often look beyond the processed output generated by a commercial meter.

Instead, we focus on the underlying electrochemical signal.

Many biosensors generate rich raw data that can reveal significantly more than the final number displayed on a screen.

By accessing that raw signal, developers can:

Evaluate lower concentrations
Understand matrix-specific behaviour
Develop application-specific algorithms
Assess feasibility earlier
Generate human data more rapidly

This approach can provide insight that would otherwise be hidden behind a proprietary calculation.

Understanding Instrument Error

Another important consideration is measurement uncertainty.

Every instrument has an inherent level of error.

When working in a new matrix, the signal may be smaller than the signal originally expected by the instrument.

As a result, the instrument's uncertainty can become a much larger proportion of the reported result.

Without accounting for this effect, researchers may mistakenly believe they have demonstrated performance when the apparent changes are actually within the instrument's error range.

Understanding these limitations is essential when interpreting early-stage feasibility data.

Accelerating Learning Through Existing Technology

One of the most powerful strategies in early biosensor development is leveraging technology that already exists.

Instead of spending years creating a sensor platform from scratch, developers can:

Identify suitable existing sensors
Acquire real-world samples
Capture raw sensor signals
Develop custom analytics
Generate preliminary human data

This approach provides valuable evidence while development costs remain relatively low.

It also allows teams to fail earlier if a concept is not viable, avoiding years of unnecessary investment.

The SenseItAll™ Approach

For non-wearable biosensor applications, platforms such as SenseItAll™ can help accelerate development by providing a ready-made ecosystem for sensor evaluation and data collection.

📱 SenseItAll™ (SIA Generation 4)
https://shop.zimmerpeacock.com/en-gb/products/senseitall-sia-generation-4-device-only

Rather than immediately investing in bespoke hardware development, teams can focus their efforts on understanding the biology, the chemistry, and the signal.

That often leads to faster decisions and better outcomes.

Fail Fast, Learn Fast

The goal is not simply to shorten development programmes.

The goal is to answer critical questions earlier.

Does the analyte provide useful information?
Does the matrix work?
Can meaningful human data be generated?
Can investors and stakeholders gain confidence sooner?

By combining existing sensor technology, raw signal analysis, data science, and thoughtful experimental design, it is often possible to obtain answers far earlier than traditional development models would suggest.

And in biosensor development, earlier answers are often the most valuable ones.