Where is the Line? When outreach becomes research

Not everything that involves people is research — until it is.

Across Aotearoa New Zealand, we see projects start as conversations, workshops, or co-design activities. But as soon as those interactions are recorded, analysed, or used to generate findings, they can move into the realm of human research. Knowing when that shift happens is one of the most common — and important — challenges in research ethics.

There is no single rule that clearly defines this boundary. In reality, the distinction between outreach and research is often blurred, and projects can move between the two over time. What matters is recognising when that shift is occurring and responding appropriately.

The following outlines a few practical tips to help navigate this grey area with confidence.

A practical way to draw the boundary

These activities typically inform design rather than generate new findings. Even at this stage, it is important to lead conversations based on ethical principles (such as respect, minimising harm, transparency, and privacy) and keep activity proportionate to risk.

• General outreach (stakeholder meetings, relationship-building, exploratory conversations).

• Co-design workshops to shape a service/tool/process before formal evaluation.

• Feasibility planning: refining questions, measures, recruitment approach, logistics.

• Internal information-gathering where you’re not systematically collecting participant data for analysis.

The “shift point”: signs you’ve moved into research (or research-like evaluation)

It is likely you are shifting into the “research realms” when you start doing one or more of the following:

1) Systematic data collection from people

Planned interviews/surveys/focus groups/observations with an intent to analyse patterns/themes.

2) Participants are being “studied” rather than consulted

You’re asking questions to produce findings (not just to improve the design).

3) Additional burden or risk appears

Discussion of sensitive topics, presence of power dynamics (e.g., staff/contractors), reputational or employment impacts, or collecting identifiable/potentially identifiable information are indicators that an ethics review may be warranted to assess risk.

4) Clear dissemination intent

You plan to publish/present externally, generalise beyond the local context, or create transferable knowledge (even if “pilot” is in the title).

5) You start testing or comparing

Piloting an intervention/tool with outcomes, pre/post measures, or comparisons across groups/sites.

At this point, ethics review is usually the right move (or at minimum, seeking advice from an ethics committee), even if the project began as co-design or scoping. No one has ever been told off by a committee for seeking guidance!

How to prevent “scope creep” into research without noticing

The following four guardrails can help to manage your projects as they evolve in this space:

Guardrail A — Define the activity prior to discussion

Write a 3–5 line statement on the purpose (design vs generate findings), who is involved, what information will be captured, and how it will be used.

Guardrail B — Limit data capture during scoping

During early outreach/co-design, avoid:

• recording “data” for later analysis (audio transcripts, formal coding)

• collecting identifiable/sensitive personal info unnecessarily

• formal recruitment scripts / participant information sheets that imply a “study”

Instead: capture high-level, non-identifiable design notes.

Guardrail C — Pre-commit a “decision trigger”

Document a trigger such as:

“If we begin systematic collection of participant data for analysis, or intend external dissemination of findings, we will stop and seek ethics review before proceeding.”

Guardrail D — Separate “advisory input” from “participants”

Treat co-design contributors as advisors, not research participants:

• be transparent about what will be noted and why

• manage confidentiality expectations

• don’t attribute quotes or identifiable views without explicit permission

A quick gut check

Are you learning about people to generate insights you’ll analyse or share? You’re probably doing research.

Are you learning from people to shape what you’re building? You’re likely still in co-design.

If it’s not clear, that’s your cue to stop and sense-check.