A definition of a verbatim: where verbatims come from, how teams analyze them, and why they persuade. Full disclosure: unitQ publishes this guide and appears as one example among the tools that do this work. 1
A verbatim is a customer’s feedback in their own words, unedited and unsummarized: the open-text comment under a survey score, the body of an app review, a support ticket message, a quote from an interview transcript. The term marks the difference between what the customer actually said and everything derived from it afterward, such as scores, tags, sentiment labels, and summaries. In research and CX work, “the verbatims” usually means the collected open-text responses attached to a survey or feedback stream.
Where verbatims come from, and how teams analyze them
Any channel that lets a customer type or speak freely produces verbatims. The classic source is the open-ended survey question (“Why did you give that score?”), but many others qualify:
- Open-ended survey responses
- App reviews
- Support tickets & chats
- Social posts & forums
- Interview transcripts
Together they form the unstructured layer of a feedback program 2, and at any real scale they vastly outnumber the structured ratings. Analysis has two traditions.
Coding
Researchers read each verbatim and tag it against a codebook of themes. Rigorous, slow, and impractical past a few thousand items.
AI theming
AI clusters verbatims into themes, attaches sentiment, and quantifies volume, then humans audit the output. Modern feedback platforms, unitQ included, are at their core machines for doing this continuously.
Whichever method applies, one rule holds: keep every derived tag linked back to its source text, because an unverifiable theme count convinces nobody.
Why verbatims matter
Metrics compress; verbatims explain. A CSAT drop of nine points tells you when something broke, and the verbatims tell you what broke, for whom, and how it felt. That diagnostic role alone would justify the attention, but verbatims carry a second power that numbers lack: persuasion. A roadmap slide claiming “18% of negative feedback concerns checkout” lands politely. Three raw customer quotes describing abandoned carts land in the gut. Experienced insight teams pair every statistic with the words behind it for exactly this reason.
Verbatims also protect against a subtle failure: a survey only asks what you thought to ask, but an open-text box lets customers raise what you never anticipated. Unprompted themes in verbatims are frequently the earliest record of an emerging problem.
A worked example
A fitness app’s quarterly NPS falls six points. The score itself supports several theories: a recent price increase, a redesigned home screen, seasonal mood. Theme analysis of detractor verbatims settles the argument in an afternoon. Price complaints are present but flat versus last quarter; what grew is a cluster of comments about workout history disappearing after the latest sync update, a bug engineering had classified as rare based on ticket volume alone. The verbatims, drawn from reviews and survey comments as well as tickets, showed the true blast radius. The fix shipped in the next release, and the team now reads sampled verbatims weekly rather than waiting for a score to move.
See how unitQ compares on your data
A short demo, run on your own feedback.
Related terms
FAQ
See what your customers say, in their words
Look up any app's free public unitQ scorecard, or take a demo to theme your own verbatims at full volume.
Sources 2 references
unitQ, "Continuous AI theming of verbatims across feedback channels, with every theme cited to source text." unitq.com. Accessed August 2026.
IBM, "What is unstructured data processing? — free-form text such as verbatims." ibm.com/think/topics/unstructured-data-processing. Accessed August 2026.