Follower & Connection List Checker

Find the differences between two supplied lists

The Follower & Connection List Checker answers a practical question: which identifiers appear in one supplied list, which appear in the other, and which appear in both? Paste your lists or choose supported local files, select the right platform and comparison mode, then review three separate result groups. The starting point is data you already have, not a username search against a live account.

A follower list checker is useful when totals hide the detail you need. Two lists can have equal sizes but contain different people. A growing list can also lose earlier entries while adding more new ones. Comparing membership reveals those differences, while the source counts show how much valid, distinct information actually entered the comparison.

Use the result to prepare a review list, reconcile records or summarize changes for a teammate. It is especially helpful when a spreadsheet contains repeated rows or a connection snapshot needs comparison with an earlier copy. Decide what question you want to answer before choosing files: current reciprocity and change over time require different pairs of lists.

Processing happens locally in your browser. There is no social login, authorization, upload of your list contents or saved comparison history. The checker does not scrape profiles, fetch the pasted URLs or verify account activity. It also performs no follow, unfollow or connection action. Keep an export yourself when you need a lasting record, because the page does not persist your inputs or results.

Choose the platform and the right comparison mode

The nine supported platform contexts are Instagram, TikTok, Threads, X (Twitter), LinkedIn, Facebook, Snapchat, YouTube and Reddit. Select the platform before preparing the two sources. That choice determines which identifier formats are recognized and which comparison modes are available. A profile address accepted for one platform is not a substitute for an address from another, even when the visible account names look similar.

Relationship mode supports four platforms: Instagram, TikTok, Threads and X. Put Following in source A and Followers in source B. Entries found only in A are not in the supplied follower list. Entries found only in B are followers not in the supplied following list. The intersection contains identifiers in both lists. This describes the supplied relationship pair, not a history of who stopped following.

Snapshot mode supports all nine platforms. Put an earlier list in A and the corresponding later list in B. Compare followers with followers, following with following, or the same type of friend, subscriber or connection list across time. LinkedIn, Facebook, Snapchat, YouTube and Reddit use this mode rather than reciprocal follower comparison. A LinkedIn connection snapshot, for example, should be compared with another connection snapshot for the same account.

For YouTube channel-list comparisons, prepare matching channel IDs or matching handles. A visible subscriber sample has its own coverage and is not the complete audience. For Reddit follower-list comparisons, prepare usernames or profile URLs from the same list type. The checker compares supplied lists; it does not fetch private subscribers, generate a follower archive or treat community members and commenters as profile followers.

Prepare text and CSV sources with one identity per record

For line input, use one handle or recognized profile URL per line. Paste the list directly or choose a TXT file. Avoid copying an entire profile screen with display names, biographies and counters. Those extra lines are not relationship records. Keep each source focused on one account, one platform and one list type so a successful comparison also has a clear meaning.

For CSV, explicitly select the identity column after loading the file. A table may contain a username, display name, email address and profile URL beside each other; the checker must know which field represents the profile. Use consistent headers when combining CSV files within a source. If valid rows unexpectedly disappear, inspect the selected column before assuming the original list is empty.

For a LinkedIn connection CSV, select the profile URL column containing personal /in/ addresses. Names and email addresses cannot reliably identify the same LinkedIn profile across exports. For other supported contexts, choose the column containing the accepted handle, profile URL or explicit account identifier. Preserve the original file separately while preparing a smaller comparison copy, so you can trace a questionable row back to its source.

The limits are 5 MB per file, 10 MB per source, 100,000 candidate rows per source and 30 files per source. The row limit applies before duplicate removal, not just to distinct results. ZIP and HTML imports are not supported. Extract archives locally first, then select relevant supported list files. Splitting one oversized source into many files does not remove the combined source limits.

Use supported JSON lists and read X archives as data

Prepared JSON can be a simple array of identifier strings. Each string still needs to match the selected platform, just as a pasted line would. A JSON file containing arbitrary account settings, message objects or a complete archive is not automatically a usable list. File format and list structure are separate checks: valid JSON alone does not establish that the file contains comparable identities.

For Instagram list JSON, the adapter recognizes supported follower records with string_list_data values or profile links, and following records inside relationships_following. Choose the actual list files, including relevant split parts when needed. In relationship mode, keep following files in A and follower files in B. Combining both relationship types within one source defeats the distinction the comparison is meant to measure.

An X follower archive comparison can read supported follower and following records containing accountId. The reader accepts validated window.YTD.follower.partN or window.YTD.following.partN JavaScript archive wrappers as data only; it never executes the script. Numeric archive identifiers become id:NUMBER. Use account IDs on both sides, or handles on both sides, because the checker cannot resolve an ID into a handle offline.

Other native archives need preparation first. TikTok, Threads, LinkedIn, Facebook and Snapchat data must be reduced to a compatible identity column, line list or prepared JSON array; their full native archives are not automatically extracted. For split files, include only parts belonging to the intended list. Leave unrelated archive material out, and check that all parts describe the same capture rather than combining different dates into an accidental hybrid.

Understand matching and cleanup before reading the totals

Matching uses normalized identifiers rather than the appearance of the original row. In Instagram, TikTok, Threads and X contexts, a recognized profile URL and its handle can produce the same lowercase handle key. A leading at sign is standardized and profile tracking details are removed. X and Twitter profile URLs are treated as equivalent. Snapchat accepts handles and recognized Snapchat profile URLs as well.

LinkedIn matching uses a canonical personal profile URL, not a person’s name. Facebook uses a recognized profile URL or an explicit id: identifier. Numeric account IDs need the supported explicit form; a bare number is not a universal username. YouTube channel IDs preserve their case and cannot be mixed with handle keys in one comparison. Reddit usernames and recognized profile URLs normalize to u/username. The checker has no live directory that could resolve different identity types into the same person.

Duplicates are removed within each source before the lists are compared. Input rows, duplicate rows, invalid entries and unique identifiers are different quantities. If a source contains ten candidate rows, two repeated entries and one invalid value, it contributes seven distinct valid identifiers. Review cleanup counts on both sides, especially when one file appears much larger than its resulting source count.

The checker allows up to 100 characters per handle and 500 per profile URL, subject to the selected platform’s accepted format. It does not use fuzzy matching for display names or emails. Case normalization is a matching rule, not identity confirmation. A renamed handle may appear as one absence and one addition; inspect the original records when that possibility matters to your interpretation.

Work through an example with six and five identifiers

Consider two prepared handle lists with no invalid entries or duplicates. Source A contains @alder, @birch, @cedar, @dahlia, @elm, @fern. Source B contains @cedar, @dahlia, @elm, @iris, @juniper. A has six unique identifiers and B has five. Three identifiers appear in both sources: @cedar, @dahlia, @elm. Those shared entries are counted once in the overlap group.

The three disjoint results are only A: 3, both: 3, only B: 2. Only A contains @alder, @birch, @fern; only B contains @iris, @juniper. The union is 8 distinct identifiers, calculated as three plus three plus two. Adding the source totals instead gives eleven occurrences because each shared identifier appears once in each source.

In relationship mode, A means Following and B means Followers. The three entries only in A are not in the supplied follower list; the three shared entries are in both relationship lists. The two entries only in B are followers not present in the following list. None of these categories establishes whether a relationship changed yesterday, last month or at all.

In snapshot mode, A means Earlier and B means Later. Three earlier identifiers are absent from the later list, three remain present and two are added. The net change is -1, calculated as five minus six, or two additions minus three absences. The earlier-list overlap is 50%: three shared identifiers divided by six earlier identifiers, multiplied by one hundred.

That overlap percentage uses the earlier list as its denominator. It is not three divided by the union, and it is not proof of audience loyalty. A net decrease of one also does not mean only one identifier changed: this example contains five nonshared identifiers. Read additions and absences together whenever a headline total seems too small to explain the underlying movement.

Review result groups and resolve incomplete comparisons

After comparison, check the source unique counts and cleanup figures before selecting a result group. The chart separates only A, the intersection and only B; its three parts do not overlap. Results are sorted for review, not ranked by importance or likelihood of an unfollow. A high position in the list has no special meaning beyond the identifier ordering.

Select a group, search its identifiers and move through pages of 50 rows. Search narrows the displayed group without recalculating the overall comparison. Source counts and group totals still describe the full validated lists. If a search returns nothing, clear the query and check the selected group before concluding that an identifier is absent from every category.

A missing B source is not automatically treated as an empty list. Explicitly confirm an intentionally empty source only when you know it contains zero entries. A file you have not obtained, an unreadable file and an export with no valid identifiers are different situations. With A populated and B confirmed empty, every A identifier belongs only to A; that mathematical result does not establish a mass unfollow event.

For snapshots, dates are optional context. When you supply dates, provide both and place the later date after the earlier date; the labels do not discover individual event times. Resolve source-specific parsing errors and check reversed relationship files or incompatible identifier types. Changes and rejected imports invalidate earlier results and exports, so compare again after correcting the inputs rather than relying on an older download.

Choose the right output for review or sharing

Copy uses the selected group and current search. It copies all matching identifiers in that selection, not just the visible page of 50 rows. Clear the search when you need the entire group, or keep it when preparing a focused review subset. The copied text contains identifiers, so it is suitable for detailed review rather than an anonymous summary.

CSV includes all categorized validated identifiers. It is not restricted by the selected group, search term or current page. Each row records the platform, comparison mode, category and identifier, with snapshot dates when supplied. Use it when you need the complete membership result in a spreadsheet. Category labels preserve why an identifier appears, which is more useful than mixing every result into one unlabeled list.

PDF is an aggregate report without raw identifiers. It includes the comparison context, optional dates, source and cleanup counts, the three result totals, a chart and practical next steps. The footer points back to this tool. Choose it for a meeting or status handoff where people need the scale of the differences but do not need an account-by-account review list.

A sensible handoff keeps the aggregate PDF with the reporting note and shares the detailed CSV only where individual review is needed. Record what the source lists represent so someone else can interpret the categories correctly. Local processing does not make a downloaded file disappear: copy and CSV expose identifiers wherever you paste or store them. The checker itself does not retain a backup of any output.

Make each connection snapshot useful for the next review

A repeatable comparison starts with consistent captures. Record the account, platform, list type, capture date and identity format beside each source file. Use the same preparation method next time, and keep follower snapshots separate from following snapshots. If one capture includes only a manually selected subset, compare that same scope later or describe the difference before interpreting additions and absences.

Check suspicious patterns at the source level first. A large disappearance alongside many invalid rows may reflect the wrong CSV column. An unexpectedly large combined list may contain duplicate archive parts or multiple capture dates. A clean result can still come from an incomplete source, so investigate how the lists were assembled when a change looks implausible. Successful parsing confirms usable records, not complete platform coverage.

When you compare follower lists over time, keep membership change separate from growth planning. Use the Follower Growth Calculator to explore a target from your own counts, and the Growth Health Checker for a separate assessment using its required inputs. Neither changes the membership evidence here. The Social Media Profile Audit is a relevant next step when your review turns toward the profile itself.

This free checker is separate from SMTasker account automation. It does not connect to SMTasker, import results into an automated workflow or act on the listed profiles. Finish with a concrete decision: correct a source, save a summary, review a small subset or schedule your own next capture. You can find other focused utilities in the free tools directory, without treating a list difference as an instruction to remove a connection.