Research Plan
Defines the question, the method and the decision the research will inform, before any data is collected.
Definition
A research plan forces the useful questions up front: what you want to know, what you will do with the answer, and what result would change your mind.
Craft convention. True because the industry converged on it. Breaking it costs familiarity, not correctness.
On this plate
Fill it in
Loading your draft…
Saved in this browser only. Nothing is sent anywhere.
Purpose
If you cannot complete this section, the study is not ready to run.
One question, phrased so it can be answered. Not a topic.
What will be done differently depending on the answer. If nothing, do not run the study.
State assumptions explicitly so they can be disconfirmed rather than reinforced.
Name it in advance. This is the strongest defence against confirmation bias.
Method
Match the method to the question. Behaviour needs observation; prevalence needs a survey.
Who, how many, how recruited, and which segments. Include people who rejected the product where relevant.
Tasks are goals, not instructions. Avoid naming the interface elements.
Behavioural measures first. Satisfaction is secondary and contaminated by aesthetics.
Logistics and ethics
Dates for recruitment, sessions, analysis and readout, and who is accountable for each.
What participants are told, what is recorded, how it is stored, when it is deleted, and what may leave your systems.
Sampling limits, moderator bias, prototype fidelity, anything that constrains the conclusions.
Signs the plan is not ready to run
Each of these is cheap to fix on paper and expensive to discover after the sessions are booked.
- The research question is a topic. Onboarding is not a question; where people stall in setup is.
- No decision is named, which means no result can be acted on and the study is entertainment.
- The method was chosen before the question, usually because the team is used to running that method.
- Nothing is written under what would change our mind, so any result can be read as support.
- Recruitment is described as users, without saying which ones or how they will be reached.
- Consent, storage and deletion are unaddressed, which is a problem to solve before the first session rather than after it.
Keeping the plan useful
Do
- Write the decision before the method, so the study is shaped by what it has to inform.
- State the confidence you already have, and where it came from. It makes the gap the study fills explicit.
- Name the limits in advance: who this sample cannot speak for, and what the prototype cannot show.
- Keep it to a page, and circulate it. A plan nobody read has not aligned anyone.
Do not
- Do not plan a study whose result is already being used in a roadmap.
- Do not recruit only from your own users when the question is about people who left or never arrived.
- Do not list satisfaction as the primary measure for a question about behaviour.
- Do not skip the plan because the study is small. Small studies are where unexamined assumptions do the most damage.
Local chart
TPL·001 · Region VIII · Grid H4
This entry and the entries it is stated to connect with. Each line is labelled with what the connection means.
Continue from here
Each link says what the connection is, so you can tell a principle from an alternative from a thing people mix this up with.
Often used with
These usually appear in the same screen or the same decision.
- Research Notes to InsightsAI·003 · AI workflow
- Confirmation BiasBIAS·009 · Cognitive bias
- Survivorship BiasBIAS·010 · Cognitive bias
- Card SortingRES·005 · Research method
- Research SynthesisRES·014 · Research method
- SurveysRES·004 · Research method
- Usability TestingRES·001 · Research method
- User InterviewsRES·003 · Research method
- Interview GuideTPL·002 · Template
- Usability Test ScriptTPL·007 · Template
- DiscoveryFLOW·004 · Workflow