Quick answer: Continuous monitoring does not automatically create a complete experimental record. Sensor data may show what happened, but without the hypothesis, configuration, operating conditions and decisions that applied at each stage, it may be difficult to explain why it happened or which observations belong to the same experiment. Where a technically uncertain outcome can only be determined through a systematic progression of work, the experimental context needs to be preserved alongside the measurements.
The 2026–27 Federal Budget announced R&DTI reforms for income years starting on or after 1 July 2028. Until those changes take effect, the R&DTI continues to be administered under the current legislation.
A data logger records conditions, not research intent
Field experiments often generate more data than laboratory work.
A monitoring station can operate continuously. Equipment historians can retain temperatures, pressures, loads and alarms. Remote systems can collect environmental variables without anyone being physically present.
That creates an appealing assumption: if everything is being logged, the research record is taking care of itself.
It is not.
Australian Government guidance describes observation as one part of a systematic progression that also includes a hypothesis, experiment, evaluation and logical conclusions. Observation means measuring and recording information that relates to the experiment; the records should also show when those observations were made.
A sensor stream therefore answers only part of the question.
It may tell you that temperature increased at 14:30, output fell shortly afterwards and vibration changed later in the day.
It does not necessarily tell you whether temperature was the variable the experiment intended to test, whether another configuration changed at the same time, whether the observation belonged to an experimental run or normal operation, or whether the team regarded the event as supporting, weakening or having no bearing on the hypothesis.
That context needs to be preserved elsewhere in the research record.
Capture the experimental state, not just the measurement
Consider a field trial investigating whether a control strategy can maintain stable equipment performance under variable environmental conditions.
The instrumentation records the physical response continuously.
But during the three-month trial, other things also happen.
A technician changes a controller setting. Firmware is updated. A damaged sensor is replaced. An operating limit is temporarily imposed. The team revises the hypothesis after an unexpected result. Several days of data are collected while the equipment is operating normally rather than experimentally.
If those events are not connected to the data, the final dataset can look continuous while the experiment itself was not.
For research purposes, what matters is the state of the experimental system when each relevant observation was produced.
That context needs to be preserved elsewhere in the research record.
This does not require duplicating every value already stored in the monitoring system.
It requires enough metadata to answer a later question:
What exactly was being tested when this result was generated?
The government's R&DTI record-keeping guidance similarly emphasises records showing what was done, how the work was performed and when it occurred, together with evidence of the systematic progression from experimental planning through trial results and evaluation. It states that records should be created at the time the activity is conducted, because information may otherwise be forgotten or overlooked.
Separate automatic evidence from human decisions
Machines are good at recording measurements.
They are much less useful at recording why the research team changed direction.
Suppose an early field run produces an unexpected oscillation.
The team reviews the evidence and decides that the original explanation is unlikely. It changes one parameter for the next trial and leaves the remaining configuration unchanged.
The monitoring platform may preserve both runs perfectly.
What it probably does not preserve is the reasoning between them.
That reasoning is part of the experimental progression.
A short contemporaneous decision record can preserve why the first result changed the next experiment: what was observed, which explanation became less plausible, what uncertainty remained, what the revised test was intended to distinguish, and which outcome would cause another change.
Without that link, a later reviewer may see Configuration A followed by Configuration B but have no reliable basis for knowing whether B represented a planned experiment, ordinary tuning or an unrelated operational adjustment.
This distinction becomes particularly important in commercial field research because experimental work and normal operation frequently occur on the same equipment.
The Australian Government's Agrifood sector guide provides a useful example of this principle. It lists production or field notes, sensor data logs and digital trial-management systems among the forms of evidence that may help demonstrate a hypothesis-led systematic progression of work.The important point is not the format of the record, but whether the records collectively show the planning, experimental actions, observations, analysis and conclusions.
Build a record that survives the end of the trial
A useful test of a field research system is to imagine opening the project six months later.
Could a technically competent person determine which periods were experimental?
Could they identify the configuration used in each significant run?
Could they distinguish a deliberately varied parameter from an uncontrolled change?
Could they see why one result caused the next experiment to be designed differently?
Could they reproduce the analysis using the same population of observations?
If those answers depend primarily on someone's memory, the monitoring system has preserved the data but not necessarily the experiment.
The solution does not have to be elaborate. In many projects, the strongest evidence architecture is a small number of linked records: the research question and hypothesis, the experiment identifier, the configuration or version in force, timestamps connecting the experiment to the automated dataset, deviations and interventions, and the evaluation or decision that followed.
The aim is traceability rather than paperwork.
This also creates a useful boundary between experimentation and routine monitoring. Once the technical question has been answered, the same sensors may continue operating for years. Their data can remain valuable for operations, maintenance or performance management, but continued collection does not by itself mean the experimental activity is continuing.
The research boundary is defined by the unresolved question and the systematic work undertaken to determine it — not by whether the data logger is still switched on. Current government guidance similarly distinguishes hypothesis-led experimental work from activities where the relevant outcome can already be known or determined in advance from current knowledge, information or experience.
Conclusion
Automated monitoring can preserve detailed observations, but an experimental record needs more than measurements.
The hypothesis, configuration, operating conditions, interventions and decisions that apply to each experimental period give those measurements their research meaning.
When those links are preserved, a later reviewer can reconstruct not only what happened, but why the evidence led from one experiment to the next.
Sources & Further Reading
Australian Government — Conducting core R&D activities for the R&D Tax Incentive
business.gov.au — Conducting core R&D activitiesAustralian Government — Record keeping for the R&D Tax Incentive
business.gov.au — R&DTI record keepingAustralian Government — Agrifood sector guide for the R&D Tax Incentive
business.gov.au — Agrifood sector guide
Talk to Ignition Research if a field experiment relies on remote monitoring, automated sensors or long-running operational data and you need to preserve the link between the measurements, experimental conditions and research decisions as configurations change. As a Registered Research Service Provider, Ignition Research assists businesses with research question definition, experimental design, technical investigations and contemporaneous supporting records within its registered research fields.
RSP registration confirms that the provider has met the relevant registration requirements for its registered research fields. It does not constitute government endorsement of a project or guarantee R&DTI eligibility.
We do not determine R&DTI eligibility or provide tax advice. The R&D Tax Incentive is a self-assessment program, and companies remain responsible for assessing their activities, expenditure and circumstances and obtaining appropriate tax advice.
This article provides general information only and is not tax, legal or financial advice.
Working on a project like this?
Tell us what you're building and the part you're not sure will work. A researcher from our Adelaide team (registered Research Service Provider RSP000047) reads every enquiry and replies within two business days. No documents needed.
