As of 13 August 2026, AI can create an onboarding feedback survey.

What to actually do

1. Do it yourself

   The route this page recommends

   A chat interface, no skill needed, and roughly 15 minutes until you can act on the result.

   How to actually do it

   1. Open a blank document and write the survey audience, when it will be sent, the onboarding stages it covers and the three to five decisions the feedback should inform.
   2. Gather the current onboarding schedule, relevant systems or training names, the response platform you plan to use and your organisation's actual rules on anonymity, confidentiality and follow-up.
   3. Paste those details into the prompt, replacing every bracketed slot, and ask the chatbot to produce the survey and its assumptions.
   4. Compare every drafted question with the onboarding schedule and remove questions about activities the respondent did not experience.
   5. Ask an HR colleague to check the introduction, anonymity wording, open-text questions and proposed handling of identifiable or sensitive responses before publishing.
   6. Build the approved questions in your survey platform, test every response option and preview the survey on a phone and desktop.
   7. Send the test version to two colleagues who know the onboarding process, compare their answers with the intended objectives, then make the final edits before sending it to new starters.

   Prompt

   Create an onboarding feedback survey for [organisation or team] to send to [new starters or other audience] after [time period or onboarding milestone].
   
   The survey should help us understand [three to five objectives, such as clarity of the role, manager support, access to systems, training and sense of belonging]. It covers this onboarding process: [brief description of the stages, tools and people involved].
   
   Create:
   1. A short, neutral introduction explaining the purpose, expected completion time and whether responses are anonymous: [anonymous or confidential details]. Do not promise anonymity unless the information provided supports that promise.
   2. A concise set of questions, grouped by theme.
   3. A suitable response type for every question, using a consistent five-point agreement or satisfaction scale where appropriate, plus a not-applicable option.
   4. Two open-text questions that invite useful detail without asking for unnecessary personal or sensitive information.
   5. One optional follow-up question asking whether the respondent wants someone to contact them, without requiring contact details in the survey.
   6. A closing message.
   
   Use plain UK English. Keep questions neutral, specific and easy to answer. Do not ask for names, health information, protected characteristics, or identifiable details unless there is a clear stated reason. Avoid leading, double-barrelled and accusatory questions. Do not invent facts about our organisation or claim that the survey is legally compliant. After the survey, list any assumptions, privacy decisions and questions an HR colleague should settle before publishing. Also suggest a simple way to review the results without treating the responses as statistically representative.

   Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

2. Use a tool built for this

   Second choice

3. Hand it to a person

   The distant third

   A person who owns the outcome does this end to end, worth it when the failure is dear.