By tokviewer.app editorial team
Comparing two TikTok comment exports starts with preserving both original files and matching records by a stable comment identifier when that field is available. A row missing from the later collection does not, by itself, prove that a comment was deleted. Collection limits and missing replies can also change the file.
Key Takeaways
- Keep the original exports and label their collection times.
- Compare the same video and the same record types.
- Treat unmatched rows as observations requiring explanation.
Define the comparison before collecting
Write down the question you want to answer. "Which comments appear in the second collection but not the first?" is a question about your files. "How many people changed their minds?" requires evidence that those files usually cannot provide.
Record the public video URL, collection time and time zone, collection method, file format, and any known limits. Note whether replies are included. Keep these notes beside each export rather than relying on a filename alone.
The export guide explains the collection workflow. This article describes a manual file comparison; it does not claim the tool automatically tracks changes.
Preserve identifiers as text
Open copies of the files for analysis and keep the originals unchanged. Import comment identifiers as text so a spreadsheet does not round long numbers. Once an identifier has been rounded, converting that cell back to text does not restore the lost digits. Reimport from the original file instead.
Follow the Excel identifier guide before matching rows. If an export lacks a stable ID, say so. Matching on text and username alone can confuse separate comments with identical wording or accounts whose displayed details changed.
Separate matches from unmatched rows
Create a comparison sheet with the identifier, first collection text, second collection text, reply reference where present, and a comparison label. Use a small set of labels:
| Label | Meaning within these files |
|---|---|
| In both | The same preserved identifier occurs in both collections |
| Second only | The identifier appears only in the later file |
| First only | The identifier appears only in the earlier file |
| Needs review | Missing IDs, duplicate IDs, or inconsistent fields prevent a clear match |
"Second only" does not necessarily mean newly posted. It may be an older comment that the earlier collection missed. "First only" does not establish deletion or moderation. Keep the labels literal until you can support a stronger explanation.
Check comparability before reporting a change
Compare replies with replies and top-level comments with top-level comments. A later file containing more reply rows can grow even when the top-level conversation has barely changed. Different sorting, collection depth, or retrieval failures can also affect overlap.
A platform's displayed count is a separate observation. Do not force the export totals to equal that count by inventing missing rows. The comment count versus export guide explains why that difference matters.
Inspect a few records directly
Review examples from each label, especially ambiguous rows. Preserve enough context to distinguish a reply from its parent. If a match looks suspicious, return to the original files before changing the matching rule.
For a fictional example, an identifier in both exports with a different like count establishes only that the exported values differ. It does not establish the exact time or cause of that change. Keep volatile counts separate from the identity of the comment.
Write a bounded conclusion
A useful report says which collections were compared, how identifiers were matched, how replies were handled, and how many records remained uncertain. Describe changes in the collected files rather than claiming complete historical coverage of the video.
Keep personal identifiers out of a shared report unless needed for a justified purpose. For theme analysis after matching, use the comment coding workflow. Preserve the raw evidence so a later collection can be added without rewriting the history.
Top comments (0)