The Scenario
You are a UX researcher. The research ops review is in three days. You manage 25 Tally feedback forms spread across product, support, and onboarding. Leadership wants to know whether the field names and question types are consistent across the library, because three different teams built these forms over 18 months and nobody ever did a systematic review.
You have all 25 form IDs in a Google Sheet, column A. You need one flat table: one row per question, with form name, question label, field type, and required flag, so you can sort and filter for inconsistencies.
The manual version:
- Open Tally form one
- Click into the form builder to see the question structure
- Copy each question label and type into the sheet by hand
- Repeat for form two through twenty-five
- Lose your place on form fourteen and have to start that one over
- You spend a full day on data entry and still miss two forms.
The fast version is one prompt.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent inside your spreadsheet that reads the form IDs and pulls the question structure from Tally for each one.
Open the SheetXAI sidebar and type:
For each form ID in column A of this sheet, fetch the full list of questions from Tally. Write them to this sheet, one row per question, with these columns: form ID, form name, question order, question label, field type, and required flag. Keep the rows grouped by form ID.
SheetXAI calls Tally for each of the 25 forms, flattens all question metadata into a single table, and writes it into the sheet grouped by form. You now have a sortable, filterable audit table built from actual form data, not from memory.
What You Get
A flat question table, one row per question across all 25 forms:
- Form ID and form name — so every row is traceable back to its source form
- Question order — the position of the question within the form
- Question label — the exact text as it appears in Tally
- Field type — short text, long text, multiple choice, email, number, etc.
- Required flag — true or false per question
The table is the audit. Sort by question label and you can immediately see whether "Company Name," "Company," and "company name" are all being used for the same concept across different forms. Sort by field type and you can find places where a date question was built as a short-text field instead.
What If the Data Is Not Quite Ready
Tally form audits surface messy patterns. SheetXAI handles the common cleanup steps inline.
When question labels have mixed case and you want to normalize for comparison
Three different teams capitalized things differently, making it hard to spot duplicates by sorting.
For each form ID in column A, fetch the question list from Tally. Write one row per question with form ID, form name, question order, question label, field type, and required flag. Add a Normalized Label column that converts each question label to lowercase so you can sort and identify duplicates.
When you only want to see questions of a specific field type
The review is focused on whether any free-text questions should have been multiple choice.
For each form ID in column A, fetch the question list from Tally. Write only the rows where field type is "TEXTAREA" or "INPUT_TEXT." Include form ID, form name, question label, and required flag. Skip all other field types.
When you want to flag questions that appear in fewer than half the forms
If a question appears in only 3 of 25 forms, it might be inconsistently used or missing from the others.
Fetch question lists for all 25 form IDs in column A. Write one row per unique question label showing how many forms include it, what field types it uses across those forms, and whether the required flag is consistent. Flag any question label that appears in fewer than 13 forms.
When you need the full audit plus a summary of inconsistencies ready to present
The research ops review needs a table of raw question data and a separate summary of the top five inconsistencies.
Fetch question structure for all 25 form IDs in column A. Write the full question table to the Audit tab: one row per question with form ID, form name, question order, label, field type, and required flag. Then write a separate summary to the Findings tab listing the five most common inconsistencies you detect: label variations, field type mismatches, and missing required flags on critical questions.
The pattern: the raw audit table and the summary both come from one prompt. You walk into the review with both.
Try It
Get the 7-day free trial of SheetXAI and paste your form IDs into a column, then ask it to pull the question structure for each one. The Tally integration is included in every SheetXAI plan. For related workflows, see how to export a form inventory or the Tally in Google Sheets overview.