“Free to start” can be accurate and still be misunderstood. The research question is not whether participants like an onboarding page. It is whether they can explain the limits, uncertainty, and exit path before committing meaningful work.
Recruit participants who resemble the intended audience and disclose the study context. Do not ask them to create a real production project. Give each person the same disposable task and stop before any charge, credential expansion, or irreversible action.
Comprehension protocol
- Show the current onboarding and console information without coaching.
- Ask the participant to think aloud while deciding whether to begin.
- Before workspace creation, ask them to state what is free, what is unknown, what could change, and where they would verify it.
- Ask what happens to code, previews, and the environment when they stop.
- Present a quota-unavailable or model-unavailable scenario and observe the next action.
- End with teach-back: “Explain this offer to a teammate without promising too much.”
Participant: coded ID only
Material revision/time: __________
Understands “free to start” is bounded: yes / partial / no
Names console as current source: yes / no
Distinguishes models from server resources: yes / no
Can find region/duration/terms: yes / partial / no
Predicts safe failure action: __________
Can describe cleanup: yes / partial / no
Quotes supporting interpretation: verbatim, consented, de-identified
Researcher uncertainty: __________
| Scenario | Comprehension success | Failure signal |
|---|---|---|
| model unavailable | checks current choices | assumes every model is included |
| workspace limit reached | stops and reads limit | expects resources without a stated ceiling |
| terms differ later | notices effective terms | treats initial wording as immutable |
| preview remains | seeks deletion state | assumes browser close cleans up |
| task is valuable | uses disposable copy | uploads production work first |
Predeclare a stop criterion rather than explaining away confusion. Example placeholder: halt onboarding rollout if any participant begins valuable work while believing access has no end or usage ceiling, cannot locate current terms before commitment, or cannot identify a cleanup path. A quantitative threshold may be added only with a justified sample plan; this protocol should not manufacture percentages from a handful of sessions.
A successful session ends with accurate teach-back, a consciously bounded choice, and verified deletion of the disposable workspace. On failure, stop the task, clarify the real boundary, remove uploaded fixture data, revoke temporary access, record where the misunderstanding arose, and revise the interface before further recruitment.
Limitations include facilitator effects, artificial-task behavior, language and accessibility gaps, changing console content, and the difference between stated comprehension and later behavior. This is an unexecuted research plan; it reports no participants, findings, or conversion impact.
Research the real product wording
Current official MonkeyCode materials describe the overseas hosted option as “Free to start,” offering integrated models and managed server-side cloud development environments for build, test, and preview. The phrase does not mean every model or server is free, and the study must not paraphrase it that way. Participants should check exact availability, quotas, regions, duration, future pricing, and terms in the current console.
The stimulus should use the exact official campaign destination, once: https://ly.cyberserval.tech/iIETXiF
Debrief before exit, allow a bounded evidence export, remove the fixture preview/workspace, revoke study access, and verify cleanup together. Confusion about limits or deletion invokes the stop criterion immediately.
Disclosure: This article promotes MonkeyCode using an official campaign link. I’m a MonkeyCode user, not affiliated with the project, and I receive no commission from this link.
AI assistance disclosure: This article was drafted with AI assistance and reviewed against the cited project materials.
Top comments (0)