TruVerbatim
Qualitative data analysis, from what people said to what it means
TruVerbatim codes your survey’s open-ended answers, lets you check and correct every code, and turns them into charts and reports.
Every example on this page uses one fictional study: Copperleaf Home’s customer survey, 1,240 answers to “What’s the main reason for your score?”
Animation: customers’ open-ended answers go into a spreadsheet, then into TruVerbatim, which codes them into themes, charts them, packages them as a PowerPoint deck and an Excel codebook, and they are presented in a meeting.
From raw text to a codeframe
Upload the file, and TruVerbatim reads it first
Bring a CSV, Excel or SPSS file. TruVerbatim cleans the answers, asks what the survey question was, then recommends the qualitative data analysis technique that suits your text.
1Upload
Upload verbatim fileClick button below or drag file to uploadA CSV, Excel (.xlsx) or SPSS (.sav) file with your verbatim responses.copperleaf_survey.xlsx1,262 rows × 6 columnsSelect column with verbatim dataChoose a columnQ5_reason open textUpload & processCSV, Excel or SPSS. Pick the column with the answers; the other columns stay, ready for crosstabs.
2Cleaned
Cleaning complete. 1,240 rows ready (1.7% removed).
Cleaning stats:Metric Count Original rows 1,262 Blank rows removed 15 Duplicates removed 7 PII instances protected 23 Final row count 1,240 Blank and duplicate rows go, and emails, phone numbers and IP addresses are masked before any coding.
3Recommended
I’ve analysed your data. Here’s my recommendation:
★ RecommendedThematic analysisAnswers are full sentences with varied wording.Transcript analysisFor long transcripts, coded section by section.Key term extractionFor short answers naming brands, products or places.Theme coding complete.TruVerbatim checks how long and varied the answers are, then suggests thematic analysis, transcript analysis or key term extraction. You choose.
Your themes
Every theme, ready to explore
This is what thematic analysis gives you: themes and sub-themes named for you and ranked by how many people mention them. Click a theme to see its sub-themes, check the coverage, or open the network to see which themes come up together.
| Main topics | Sub-themes | Coverage |
|---|---|---|
| 1: Delivery | Late delivery, Courier updates, Damaged on arrival, On time | 31% |
| 2: Product quality | Build quality, Matches the photos, Comfort | 30% |
| 3: Price and value | Good value, Delivery charges, Discounts | 25% |
| 4: Customer service | Slow replies, Helpful staff | 19% |
| 5: Website and checkout | Easy to order, Stock shown wrongly | 15% |
| 6: Returns | Easy returns, Refund speed | 9% |
| Main topics | Sub-themes | Coverage |
|---|---|---|
| 1: Delivery | Late delivery, Courier updates, Damaged on arrival, On time | 31% |
| 2: Product quality | Build quality, Matches the photos, Comfort | 30% |
| 3: Price and value | Good value, Delivery charges, Discounts | 25% |
| 4: Customer service | Slow replies, Helpful staff | 19% |
| 5: Website and checkout | Easy to order, Stock shown wrongly | 15% |
| 6: Returns | Easy returns, Refund speed | 9% |
| Topic ID | Topic | Sub-themes |
|---|---|---|
| 1 | Delivery | 1.1 Late delivery, 1.2 Courier updates, 1.3 Damaged on arrival, 1.4 On time |
| 2 | Product quality | 2.1 Build quality, 2.2 Matches the photos, 2.3 Comfort |
| 3 | Price and value | 3.1 Good value, 3.2 Delivery charges, 3.3 Discounts |
| 4 | Customer service | 4.1 Slow replies, 4.2 Helpful staff |
| 5 | Website and checkout | 5.1 Easy to order, 5.2 Stock shown wrongly |
| 6 | Returns | 6.1 Easy returns, 6.2 Refund speed |
| 0 | Other/outliers | None |
Your codeframe or ours
Discover the themes, or code to the codeframe you already use
Two qualitative data analysis methods in one place. Let TruVerbatim find and name the themes and sub-themes, or upload your codeframe and it codes every answer to it, with more than one code where an answer needs it.
Theme coding complete.
| Verbatim | Themes multicoded (named by TruVerbatim) |
|---|---|
| Arrived two days late and nobody told me. | 1.1: Late delivery1.2: Courier updates |
| Great sofa, much cheaper than the high street. | 2.1: Build quality3.1: Good value |
| Took three emails to get a reply about my refund. | 4.1: Slow replies6.2: Refund speed |
| Easy site, but it said in stock and it was not. | 5.1: Easy to order5.2: Stock shown wrongly |
Quality check complete. Your coding looks good.
| Verbatim | Your codes (your numbers and labels) |
|---|---|
| Arrived two days late and nobody told me. | 11 Late delivery31 No contact from courier |
| Great sofa, much cheaper than the high street. | 05 Product quality02 Value for money |
| Took three emails to get a reply about my refund. | 21 Slow response41 Refund issues |
| Easy site, but it said in stock and it was not. | 51 Website easy to use52 Stock accuracy |
Your codeframe keeps your own code numbers. Upload it as a CSV or Excel file.
Refine it in words
Review the coding with your business context
Good analysis of qualitative data still needs your judgement. Bring what you know about your market and customers, and refine the coding in your own words: TruVerbatim shows each answer it would change and why, you untick any you disagree with, and nothing changes until you confirm.
Theme coding complete. 2.1: Build quality (under Product quality) is mentioned by 17% of respondents and 1.3: Damaged on arrival (under Delivery) by 6%. Ask about your data, or type /feedback to correct the coding.
/feedback items damaged in transit are coded as Build quality. They should be Damaged on arrival.
I found 41 answers coded to 2.1: Build quality that describe damage in transit. Here is what I would change. Untick any you want to keep as they are.
Re-code 41 response(s) following your feedback.
| Apply | Verbatim | Current theme | Proposed | Reason |
|---|---|---|---|---|
| 2.1: Build quality | 1.3: Damaged on arrival | Broken in transit: the box was crushed | ||
| 2.1: Build quality | 1.3: Damaged on arrival | Damage noticed on arrival | ||
| 2.1: Build quality | 1.3: Damaged on arrival | Damaged by the courier | ||
| 2.1: Build quality | 1.3: Damaged on arrival | Broke in use, not in transit | ||
| 2.1: Build quality | 1.3: Damaged on arrival | Damaged in transit | ||
| and 36 more | ||||
40 of 41 changes will be applied; unticked responses keep their current codes.
Done. 40 responses moved from 2.1: Build quality to 1.3: Damaged on arrival. Build quality is now 14% of respondents (was 17%) and Damaged on arrival 9% (was 6%). Saved as a new version; the original coding is kept and can be restored from the Data tab.
- Add your context. Type /feedback and what you know, as you’d brief a colleague.
- Check each change. Every answer it would move, with its reason. Untick any you disagree with.
- Go back any time. Each refinement is saved as a new version, and the original coding is always kept.
Ask the data
Compare segments and pull quotes in natural language
Data analysis in qualitative research means comparing groups. Ask TruVerbatim, as you would a colleague, for themes by customer type or score. It builds the crosstab, charts it and finds the quotes that explain it. Switch to Data to read every answer beside its codes.
Here are some suggestions:
| ID | Verbatim | Theme | Themes multicoded | Sub-theme | Sentiment | Score group |
|---|---|---|---|---|---|---|
| 104 | Arrived two days late and nobody told me. | 1: Delivery | 1.1: Late delivery1.2: Courier updates | Late delivery | Negative | Detractor |
| 231 | The courier texted a slot and turned up on time. Lovely table too. | 1: Delivery | 1.2: Courier updates2.1: Build quality | Courier updates | Positive | Promoter |
| 377 | Box was crushed and one leg was scratched. | 1: Delivery | 1.3: Damaged on arrival | Damaged on arrival | Negative | Detractor |
| 402 | Great sofa, much cheaper than the high street. | 2: Product quality | 2.1: Build quality3.1: Good value | Build quality | Positive | Promoter |
| 512 | Fine. It came when they said. | 1: Delivery | 1.4: On time | On time | Neutral | Passive |
| 618 | Took three emails to get a reply about my refund. | 4: Customer service | 4.1: Slow replies6.2: Refund speed | Slow replies | Negative | Detractor |
| 733 | Easy site, but it said in stock and it was not. | 5: Website and checkout | 5.1: Easy to order5.2: Stock shown wrongly | Stock shown wrongly | Negative | Passive |
| 845 | Returned a rug with no fuss and the refund came in two days. | 6: Returns | 6.1: Easy returns6.2: Refund speed | Easy returns | Positive | Promoter |
- Crosstabs by any column in your file: score, region, customer type.
- A chart and a network map for every crosstab, so you see where each theme leans.
- Every code traces back to a quote. Read, filter and change the coding in Data.
What researchers said
It helps with story driven insight.
Backed by data and analytics consultants
The researchers who build TruVerbatim also help you adopt it: onboarding, help designing your codeframe, and a second pair of eyes on the coding and its interpretation before results go out.
Frequently asked questions
Questions about TruVerbatim
About the files it reads, how it codes, how you correct it and how we look after your data.
Which files can I upload?
CSV, Excel (.xlsx) and SPSS (.sav and .zsav). Pick the column with the open-ended answers; the other columns stay, so you can crosstab the themes by them later. A codeframe can be a CSV, Excel or Parquet file.
Can I use my own codeframe?
Yes. Upload it and TruVerbatim codes every answer to it, with more than one code where an answer needs it, and keeps your own code numbers. Or let it find the themes and sub-themes for you, and name them.
Are personal details removed before coding?
Emails, phone numbers and IP addresses are masked before any coding, and blank and duplicate rows are removed. To code the answers, TruVerbatim sends the open-ended text, masked, to the language model. Your data is never used to train AI.
Can I change the coding?
Yes. Merge, rename or split themes in plain English, or type /feedback with a correction: TruVerbatim shows every answer it would change and why, and nothing changes until you confirm. You can also change any answer’s theme or sentiment in the Data tab. Every change is saved as a version you can restore.
Does it work for interview transcripts and short answers?
Yes. Transcript analysis codes long transcripts section by section. For short answers that name brands, products or places, key term extraction pulls out the terms and brings different spellings together.
Which languages does it support?
English. TruVerbatim doesn’t translate, so answers in other languages need translating before you upload them.
What do I get at the end?
An Excel codebook (codes, key terms, example quotes, a summary, crosstabs and an audit trail of every change), the coded data as a CSV with your original columns, an AI-written PowerPoint report with charts you can edit, and every chart as an image.
How much does TruVerbatim cost?
Pricing depends on usage and the support you need. Most teams start with a single project: book a demo and we’ll set one up around your data.
Book a demo
Get started
See TruVerbatim on your open-ends
Send us a file of open-ended answers and we’ll code it with you in a demo, or talk to our data and analytics consulting team about a project.
Or call 020 3002 1700 Email contactus@trustrata.com
