Survey

Most organisations can send a questionnaire. Far fewer can say what it actually revealed. The questions are written quickly, the answers come back as a list of averages, and the interesting part — where people disagree, where an internal team and an external customer describe the same thing differently, where someone hints at a problem without ever naming it — is lost before anyone reads the results properly. And nobody sits down with sixty respondents to ask the obvious follow-up question.

Survey is built around that gap.

Write up to thirty questions, each with a short title, a longer description that explains what you are really asking, and an image where a picture makes the question clearer. Decide who should hear from: internal colleagues, external stakeholders, or both, added individually or pasted in as a roster. Each person gets their own link, no account required, and anyone who has not answered after 24 hours gets one automatic reminder.

Switch on the smart interviewer in Setup and choose how many follow-up questions it may add, up to ten. When someone saves their answers, the tool reads what that person specifically said and writes personal follow-up questions in the same flow — the kind an experienced senior interviewer would ask: you mentioned the handover causes delays, what happens in practice when it goes wrong? They answer immediately, in the same sitting. Those answers are stored in their own section and are read as part of the deep analysis, so the investigation goes deeper than the questions you thought to ask.

The dashboard shows participants, questions, answers, participation rate, completed surveys, completion rate, responders and average answers per participant. It breaks down response by question so you can see where people stop answering, by organisation, and over time. It also compares internal with external participation, which is often the first sign that one group is not really engaged.

• The headline and key findings — what the data says overall, with the evidence behind it.
• Themes with counts — which issues recur and how often, rather than which ones were loudest.
• Question-by-question insights — where respondents agree and where the tensions sit.
• Internal versus external — how the two groups differ, and what each mentions that the other does not.
• Recommendations — concrete improvement suggestions with next steps, plus the Upgraded.today tools that would strengthen the investigation further.

Once the analysis has run, ask the collected answers anything in plain text: what do external stakeholders complain about that internal ones never mention, which theme comes up most often among the respondents who wrote the longest answers, what should the next round of questions focus on. Replies come only from what you gathered, and the tool says plainly when the data cannot support an answer.

Invite up to three colleagues: Contributors who can run analyses, Guests who read the initiative without changing anything. Everything exports as CSV — answers with participant names and question titles, and the personal links if you need to re-run the collection.