When to Use It
Use this workflow when a customer sends you a questionnaire directly (e.g., a spreadsheet emailed by a prospect) rather than through a VISO TRUST collection request. Typical inputs:- A vendor security questionnaire from a prospect
- A CAIQ, SIG, or custom spreadsheet
- A short list of ad-hoc questions you want answered from your existing documentation
Prerequisites
- A Trust Profile with artifacts uploaded to the Knowledge base tab
- Artifacts have finished processing — quality of answers depends on what’s available in your Knowledge base
- The Org Admin role
Adding a Questionnaire
Navigate to Trust → Questionnaire answering, then choose one of two options from Add questions.Upload a File
1
Select Upload file
Accepted file types:
.xlsx, .xls, .xlsm, .csv. Maximum size: 100 MB.2
Wait for parsing
VISO TRUST extracts questions from the file and lists the questionnaire in the table with a Processing status. Question count appears as answers are generated (e.g., 12 / 40).
3
Open the questionnaire
Once parsing finishes, select the questionnaire name to open the answer view.
Paste Questions
1
Select Enter questions
A dialog opens with a text area.
2
Paste your questions
Enter one question per line, or separate them with commas. A live count shows how many questions were detected.
3
Select Add questions
The pasted set is added to the table alongside uploaded questionnaires.
Reviewing Answers
Open a questionnaire to see the full answer table. Filter toggles at the top let you narrow the view to All, Answered, Needs attention, or Unable to answer. A Processing indicator appears while remaining answers are being generated. Each row includes:- Status — Answered, Needs attention, Unable to answer, or Processing
- Question — the extracted question text
- Answer — the AI-generated response
- Confidence — how strongly the supporting evidence matches the question
Statuses
Source Citations
For every AI answer, select View source citation to expand the supporting evidence. Each citation shows the quoted snippet, the source document name, and the page number when available. Use this to verify an answer before accepting it.Accepting or Editing an Answer
For rows in Needs attention, you have two ways to move the row to Answered:- Select the check-mark Mark answer as good button next to the answer to accept the AI response as-is.
- Select the edit icon to modify the wording, then save.
Renaming a Questionnaire
By default, uploaded questionnaires use the file name and pasted questionnaires usePasted Questions <date>. To rename one, open the questionnaire and select the edit icon next to the title.
Exporting Answers
From the questionnaire detail page, select Download CSV to export every question, answer, status, confidence score, and source citation. Use the export to return the completed questionnaire to the requesting customer. Each AI answer includes its source citations in a Source column, formatted asfileName (p.N). When an answer has more than one citation, the first appears on the answer’s row and each additional source is written on its own row below it. Manually-authored answers have no source column populated, matching the detail view where citations are hidden once you supply your own answer.
Improving Answer Quality
Answer quality depends on what’s in your Knowledge base. If too many questions come back as Unable to answer or Needs attention, or confidence scores are consistently low:- Add missing documents (policies, SOC 2, DPAs, sub-processor lists) to the Knowledge base tab
- Replace outdated artifacts so the AI doesn’t cite expired evidence
- Include URLs to your public security and privacy pages