CSV · UTF-8 · stable IDs
Export Instagram comments to CSV for your CRM
A plain CSV that imports into a CRM, a database or a script: one comment per row, a stable numeric user ID, UTC timestamps and the mentions already pulled out. Free for posts with up to 1,000 comments.
What the CSV contains
Comma separated, double-quoted where needed, UTF-8 with a byte-order mark (so Excel shows emoji and accents correctly and most importers detect the encoding). Header row:
#,comment_id,username,user_id,comment,posted_at_utc,likes,is_reply,reply_to_comment_id,reply_to_username,mentions,hashtags
A comment with a line break stays one record: the line break is inside quotes, as the CSV standard allows. Importers that read CSV properly handle it; a naive line-by-line script will not, so use a real CSV parser.
Mapping it into a CRM
| CSV column | CRM field | Why |
|---|---|---|
user_id | External ID / unique key | Never changes, even when someone renames their account. De-duplicate on this. |
username | Instagram handle | Human readable; may change over time. |
comment | Note or activity body | Keep it as an activity on the contact rather than overwriting a field. |
posted_at_utc | Activity date | UTC; tell the importer the time zone is UTC. |
mentions | Custom text field or tag | Useful for referral or tag-a-friend campaigns. |
comment_id | Activity external ID | Re-importing the same file will not create duplicate activities. |
There are no emails or phone numbers in the file: the export holds public comment data only. A CRM that requires an email for every contact needs a field for the handle instead, or a separate list for social contacts.
One row per person before you import
Most CRMs want contacts, not comments. Two ways to get one row per commenter:
- Tick one comment per person on the export page before pressing CSV. The file then holds each commenter once (their first comment).
- Import the full file as activities and let the CRM merge on
user_id.
In a script: pandas.read_csv("file.csv", dtype={"comment_id": str, "user_id": str, "reply_to_comment_id": str}) keeps the IDs as text, and df.drop_duplicates("user_id") gives one row per person.
Use the data the way commenters would expect
Commenters wrote in public, on your post or a post you are studying. That does not make them subscribers. Use the list to reply, to measure, or to follow up on a question someone asked; do not add people to marketing email or message them in bulk. If someone asks to be removed, remove them, and tell us too: email [email protected] and their comments are deleted from stored exports.
Questions
Is the CSV comma or semicolon separated?
Comma. If your Excel uses a semicolon locale, open it with Data, From Text/CSV and choose comma, or download the XLSX file instead.
Why does the file start with invisible characters?
That is the UTF-8 byte-order mark. It tells Excel the file is UTF-8. Most importers skip it; in Python use encoding="utf-8-sig".
Can I get JSON instead?
Yes. The JSON download has the same fields per comment, plus the post URL, the count Instagram reports and when the comments were read.
Are display names or follower counts included?
No. Only the public comment data: username, user ID, text, time, likes, reply links, mentions and hashtags.