For organizations that use Viva Glint to understand employee feedback, the distinction matters. A spreadsheet with assigned topics can speed up follow-up work, but it also creates opportunities to overread labels, compare unlike survey cycles, or confuse a Comments report export with the more restricted raw-survey-response export.
What the Comments report export provides
Microsoft’s current Viva Glint documentation describes an Export Comments to XLSX Spreadsheet option in the Comments report. The exported data includes a topic column, allowing reviewers to move beyond a long list of individual comments and work with an associated categorization field.
In practice, that can make several ordinary tasks easier:
- Filtering feedback associated with a particular standard topic.
- Grouping comments for review meetings.
- Identifying recurring themes that merit a closer read of the underlying text.
- Building a controlled internal follow-up workflow without manually copying comments from the report.
The key point is that the spreadsheet is not simply a text dump. It carries both the original comment and topic-related information. That makes the export more useful for analysis, but it also means the topic column must be understood as a product feature with defined limitations—not as an infallible interpretation of every employee’s meaning.
A topic label can help a reviewer prioritize material. It is not, by itself, proof that all comments sharing a label express the same concern, agree on a cause, or require the same response. Teams should still read a representative set of the comments behind any apparent trend before treating the label as a management conclusion.
Why some exported comments will show N/A
The presence of a topic column does not mean every row will receive a named topic. Microsoft states that comments which do not align with Viva Glint’s standard topic set appear as N/A in that column.
That is an important caveat for anyone using filters or counts in a spreadsheet. An N/A value should not automatically be treated as irrelevant feedback, poor-quality feedback, or a blank technical failure. Based on the documented behavior, it means the comment did not align to a standard topic.
This creates a practical review requirement: N/A comments need their own sampling and reading process. They may include highly specific issues, emerging concerns, mixed feedback, or language that does not map cleanly to the standard classification set. Excluding them from a review simply because they cannot be easily grouped could make an analysis less representative.
A sensible workflow is to separate the export into at least two review paths:
- Use named topics to organize broad, recurring feedback.
- Review N/A comments as a distinct queue, looking for issues that the standard topic set does not surface well.
That approach avoids two opposite mistakes. The first is assuming that the categorized rows tell the whole story. The second is dismissing topic assignment altogether because it does not classify every comment.
Multilingual feedback: original text, English-based analysis
The treatment of non-English comments deserves particular care. Microsoft says that comments exported to a spreadsheet are not translated: the comment text remains in its original language. At the same time, the documented topic assignment reflects analysis performed after translation to English.
For a global organization, this has a clear benefit. People reviewing feedback in its source language can retain the original wording rather than working solely from a translated export. It also means a multilingual spreadsheet may contain comments in many languages while still offering a common topic field.
However, the combination calls for measured interpretation. The topic value can support cross-language grouping, but it should not be treated as a substitute for a fluent review of the original comment when nuance matters. Tone, local terminology, idioms, and ambiguity can be significant in employee feedback. A topic may be directionally useful while still failing to capture all of the context a human reader would identify.
Teams working across regions should consider a simple validation practice: when a topic becomes important to a decision, ask reviewers who understand the relevant languages to examine a sample of the original comments. This is especially valuable for small clusters of feedback, sensitive organizational changes, or comments that appear contradictory despite sharing a topic.
Copilot-enhanced assignment is conditional, not automatic everywhere
Organizations should also avoid assuming that Copilot-enhanced topic assignment is active merely because Viva Glint is in use. Microsoft says the capability is automatically available when Copilot in Viva Glint is enabled for the organization and when the applicable survey program or report is supported.
The documented scope currently includes ad-hoc and recurring surveys, with support for always-on surveys planned to follow. As a result, two programs in the same organization may not necessarily offer identical topic-assignment behavior if they use different survey models.
That creates a basic governance question for HR, people analytics, and IT teams: before comparing exports from separate programs, confirm that they were produced under comparable conditions. Differences in survey type, feature enablement, and reporting setup can affect what reviewers see and how they should interpret it.
The practical takeaway is not that topic assignment lacks value. Rather, it is that the presence and quality of a topic field should be checked as part of the report-review process. A repeatable checklist can help:
- Confirm the survey program type.
- Verify whether Copilot in Viva Glint is enabled for the organization.
- Check that the relevant program or report is supported.
- Inspect the proportion of N/A values before relying heavily on topic-based filters.
- Read original comments behind important topic clusters, particularly across languages.
Do not confuse the Comments report with raw response export
Viva Glint has a separate capability for exporting raw survey responses. It can include fields such as comments, comment topics, survey-cycle details, and a sensitive-comment flag. But it is not the same as exporting from the Comments report, and its controls are more restrictive.
Microsoft documents that raw-survey-response export is available only to administrators. It must also be enabled for a survey program before the relevant survey cycle goes live. If that option was not enabled in advance, the raw export is not enabled for that survey cycle.
This difference is easy to overlook because both features can involve comments and comment topics. Yet their purposes are distinct:
- The Comments report export is the report-level workflow that provides an XLSX export with a topic column.
- The raw-survey-response export is a separately controlled administrative capability, configured before a survey cycle begins and capable of including a broader set of response fields.
For Windows and Microsoft 365 administrators, this means export expectations need to be set early. A team should not assume that access to a Comments report spreadsheet implies that raw-survey-response data will be available later. Conversely, enabling a raw response export is a more consequential configuration choice because it is tied to the survey program before the cycle starts.
Organizations should align those choices with their internal privacy, employee-feedback, and data-retention practices. Even when a spreadsheet improves analysis, copying it into unmanaged locations, distributing it widely, or combining it with unrelated employee data can create avoidable handling risks. The appropriate recipients, storage location, and retention period should be decided before exports become routine.
How to use topic exports without overstating the result
The strongest use of topic data is as a review aid rather than a final verdict. A management team might use it to locate clusters of comments about a recurring workplace issue, then read those comments and decide whether further listening, a policy response, or clearer communication is needed.
A weaker use is to treat the count of a topic as a precise measurement of workforce sentiment without examining the comments, the N/A population, survey context, or language mix. Topic assignment can bring order to unstructured feedback, but categorization necessarily compresses meaning. The spreadsheet gives teams a starting point for analysis, not an automatic explanation of why employees feel as they do.
There is also an unresolved implementation detail worth keeping in mind: Viva Glint’s feature description refers to assigned “topic or topics,” while the Comments report documentation refers to a topic column. The available documentation does not establish the exact spreadsheet layout where a comment has multiple assigned topics. Administrators and analysts should inspect an actual export from their own tenant before designing formulas, dashboards, or downstream imports around a presumed column format.
A more disciplined spreadsheet workflow
For organizations making regular use of the export, a disciplined routine will produce better outcomes than simply sorting the spreadsheet by topic. Start by preserving the survey cycle and report context alongside any working copy. Then review named-topic groups and N/A comments separately. Where comments are in multiple languages, retain the original text and involve appropriate language expertise before escalating a conclusion.
When sharing findings, summarize the patterns and include enough underlying qualitative review to explain what the labels do—and do not—show. Avoid presenting a topic field as though it were an employee’s own selected category. It is an assigned analytical aid.
Viva Glint’s comment-topic export can make feedback review faster and more structured. Its real value lies in helping people move from a large body of written feedback to a careful human discussion. Used with attention to N/A values, multilingual context, feature scope, and export permissions, it can improve the path from survey comments to informed action without creating false certainty from a spreadsheet column.