Giveaway Duplicates and Entry Rules That Match Your Draw

Use the giveaway winner picker after you have prepared your entry list and decided what earns a chance. Its review step shows the effect of your selected rules before you draw. It works with the identifiers and comments you supply; it does not inspect social accounts or decide whether two accounts belong to one person. Keep your original source beside the working list so every removal has an explanation beyond making the totals smaller.

Write the ticket rule before cleaning the spreadsheet

Begin with a sentence that another operator could apply without asking you what you intended. ‘One eligible account gets one chance’ produces a different pool from ‘Every eligible comment gets one chance.’ The pottery club offers a beginner glazing kit and has invited members to describe a project. Before importing anything, the organizer checks whether a second project comment earns another chance or merely adds to the conversation.

Do not let the presence of repeated rows answer that question for you. A second row may be another genuine submission, an accidental double paste, or a reply thanking the organizer. Each possibility needs a different treatment. Automatic exact duplicate removal can address identical supplied records, but it cannot establish why those records exist. Your entry wording and preparation notes provide the context that the browser cannot recover.

Separate conditions that this picker can apply from conditions you must check elsewhere. It can compare supplied identifiers, exclude listed entrants, look for a literal comment substring, and change how eligible rows are weighted. It cannot establish a submission time from an ignored date column, verify a follow, count genuine tagged friends, or interpret a photograph. Complete those checks while preparing the source rather than treating import as verification.

Keep a short decision sheet with the campaign name, closing point, eligible submission types, and treatment of repeats. For a one-comment offer, specify what happens when somebody edits their answer or posts a correction. For a multiple-entry offer, specify any limit before drawing. The tool has no automatic daily cap or per-account entry ceiling, so a capped campaign needs a prepared list that already implements the announced limit consistently.

Choose an identifier that survives a useful review

Use a consistent account identifier when your source provides one. On X, the username or handle is distinct from the display name, so copying the decorative name beside a reply can lose the distinction between accounts. A spreadsheet containing two rows called ‘Clay Studio’ might describe two different profiles. Preserve the actual handle in your private preparation record and select the identifier column deliberately when importing CSV.

The picker treats a supplied name or handle as a label, not as a verified person. If your event registration system issued entry references, those references can be useful identifiers, provided you understand what they represent. Two references for the same attendee remain two identities here unless you resolve them externally. Conversely, using a shared household name as the identifier can collapse several legitimate attendees into one chance.

Never merge accounts simply because their comments sound alike. ‘I would make a blue mug’ is an ordinary answer that several people could independently write. Matching email fragments, similar avatars, and familiar surnames are also outside this tool’s identity model. If your campaign requires one entry per person rather than per supplied account, establish a separate, reliable preparation process; the picker does not turn account labels into personal identity evidence.

For a giveaway spanning several platforms, consider whether identical handle text really means the same entrant. An account named ‘kilnnotes’ on one service need not belong to the owner of that name elsewhere. Prepare distinct labels where separate platform identities should remain separate, and keep an external lookup to the original accounts. Do not add labels inconsistently halfway through the list, because only some participants would then receive separate identities.

Understand the matching switches before applying them

Identifier cleanup trims surrounding whitespace and applies Unicode NFC normalization. That removes a common copying mismatch and makes canonically equivalent character sequences comparable. It does not translate a name, remove all accents, repair a misspelling, or decide that visually similar letters are interchangeable. If an entrant looks duplicated but remains separate, compare the actual supplied identifiers with the source instead of assuming the normalization is broken.

Leading at-sign removal and case folding are optional decisions. With both enabled, ‘@MiraClay’, ‘miraclay’, and the same text padded with spaces can resolve to one identity. With those switches disabled, relevant differences remain. Choose settings that match the identifiers in this particular list, then inspect the resulting unique count. A lower count is not automatically an improvement if the source genuinely distinguishes the supplied labels.

Use a small representative sample to test the effect before reviewing the full campaign. Include an identifier with surrounding spaces, a mixed-case handle, and a real accented name if your list contains one. Compare the expected matches with the review. This is especially helpful when several volunteers collected entries with different copying habits. Agree on one preparation convention instead of repeatedly changing switches until a preferred winner becomes possible.

Preserve recognizable display text in the source archive even when matching uses a cleaned identity. The important distinction is between the label a teammate needs to find the submission and the comparison rule used to count it. If you manually correct a transcription error, record that correction outside the picker. Changing the actual input later invalidates the reviewed pool and any result, so resolve ambiguous spellings before the draw session.

Remove exact copies without deleting valid participation

An exact duplicate row has the same normalized entrant and the same comment after trimming and NFC normalization. It is not merely another row from that entrant. In the pottery list, Mira’s ‘A blue mug’ and ‘A green bowl’ are different supplied comments. Two copies of ‘A blue mug’ for the same normalized identity are exact duplicates. That definition keeps identity grouping separate from the decision to retain individual entry rows.

Exact duplicate removal defaults to on. This is useful when two collectors pasted an overlapping set of comments or a CSV was assembled twice. However, repeated identical comments might represent separate submissions under an offer that explicitly allows them. Turning removal off lets repeated eligible rows count in entries mode. Decide from the announced participation rule and the source evidence, not from a general belief that repetition is always wrong.

Comment comparison is not the keyword filter. Exact duplicate detection does not make comment text case-insensitive merely because identity case folding is enabled. ‘Blue mug’ and ‘BLUE MUG’ are not the same trimmed comment under the stated duplicate definition. Internal punctuation and spacing differences also need attention during preparation. Do not describe the feature as a similarity detector or expect it to remove paraphrases and slightly rewritten comments.

A plain list has no selected comment content to distinguish repeated lines. Repeating an identical normalized identifier therefore creates the same entrant with the same empty comment. With duplicate removal on, those copies do not provide extra eligible rows. For a genuine multiple-entry campaign, use the setting and source structure that preserve the intended entries, then verify the eligible row total rather than assuming every pasted line survived unchanged.

Calculate what the two weighting modes actually change

Consider eight eligible rows representing five identities: Mira has three, Owen has two, and Pia, Ravi, and Sana have one each. In unique mode, all five identities have one chance at the first winner position. Mira’s first-position probability is one fifth, or twenty percent, just like Sana’s. Keeping her three valid project descriptions in the prepared source does not multiply her chance when the selected mode is unique.

In entries mode, each eligible row contributes to the first selection. Mira has three of eight chances, or thirty-seven and a half percent; Owen has two of eight; each remaining identity has one of eight. This is an intentional weighting difference, not a defect in random selection. State the mode in your handoff so a teammate does not interpret a row-weighted draw as an equal-account draw merely because the final names are distinct.

Once an identity is selected, all its rows leave the pool before the next rank. If Mira is first, five rows remain across four identities. Owen then holds two of those five chances in entries mode. He cannot receive a second position after being selected himself. Winners and reserves therefore remain distinct even when one account originally contributed many rows; this picker is not designed to award several prizes to the same supplied identity.

Do not multiply the initial first-position percentage by the number of prizes and call that the final chance of winning. Later selections depend on which identities and their rows have already left. The simple fractions above explain the first position and one stated next-step condition, not every possible outcome of a multi-winner draw. For a campaign needing repeated awards to one account, choose a different process rather than trying to force that behavior here.

Apply eligibility without turning text into identity evidence

The keyword filter searches the selected comment for a literal, case-insensitive substring. Searching for ‘glaze’ can match ‘GLAZE workshop’ and ‘glazed bowl.’ It does not require a whole word, understand the answer, or apply a regular expression. A short search such as ‘art’ can also appear inside ‘cart.’ Test your actual entry phrase against plausible comments before deciding that the filter expresses your campaign requirement.

A tag appearing in comment text does not become another entrant. If Mira writes ‘Making mugs with @OwenPots,’ Mira remains the identifier selected from that row. The picker does not add Owen to the pool, verify the mention, or count how many different friends were tagged. If your offer depends on valid mentions, review that requirement in the source preparation stage and supply only the rows that should qualify under it.

An excluded entrant is matched using the same normalized identity rules and loses all associated rows. This is appropriate when the prepared exclusion list identifies an organizer account or another account outside the campaign. It is different from rejecting one nonqualifying comment while retaining that account’s qualifying entry. Before using an entrant exclusion, check whether the intended decision concerns the whole identity or just a particular supplied submission.

Review missing and invalid records alongside exclusions and keyword failures. An empty identifier cannot become an eligible entrant just because its comment contains the right phrase. Nor should a malformed source row quietly become evidence of misconduct. Keep reasons factual, such as missing identifier or outside the prepared cutoff, and distinguish technical import problems from participation decisions. Correct a source problem first when the underlying submission can still be established reliably.

Reconcile the denominator with the collection boundary

The starting comment count on a post is not the denominator for every giveaway. A conversation may include organizer responses, late replies, and contributions that never attempted to enter. YouTube, for example, threads replies and offers different comment ordering views. When preparing a pool from that context, define whether replies qualify and document the collection boundary. A top-comment view alone is not a documented list of every eligible submission.

Suppose a club spreadsheet starts with sixteen collected rows representing ten identities. Three rows are overlapping copies from two collectors, two comments arrived after the closing point, and one row is the organizer’s response. The meaningful summary is not simply ‘six duplicates removed.’ State each preparation decision separately, then report the final eligible rows and identities. Which identities disappear entirely depends on which rows those removals affected, so calculate rather than guess.

Dates, platform labels, and evidence links in unselected CSV columns are ignored by the picker. They can help your external preparation, but importing them does not activate time filtering or account verification. If the cutoff was Friday evening, resolve late submissions before selecting the entrant and comment fields. Keep the full working spreadsheet privately and import a minimal supported source; do not expect an ignored column to influence the denominator invisibly.

For limited bonus entries, reconcile the prepared rows against the cap before import. If a workshop attendee may earn one attendance entry and one project entry, three project comments should not silently become three project bonuses. Keep an external explanation of which qualifying events generated the retained rows. The final review should let you answer both ‘How many chances are there?’ and ‘Why does this account have that many?’ without reconstructing decisions afterward.

Freeze a pool another operator could explain

Before drawing, ask a second reviewer to explain the settings using the prepared list. They should identify the matching options, duplicate choice, weighting mode, exclusions, keyword, and eligible totals. If they cannot explain why Mira has three rows but one unique-mode chance, pause to clarify the rule. This check is useful even for small giveaways because the confusing part is usually the relationship between rows and identities, not the number of participants.

Review freezes the eligible pool for the next explicit draw. Editing the source or a rule invalidates that review, existing results, and exports immediately. Treat that reset as a meaningful boundary: you have changed what is being selected from. Return to the picker to review the final entry rules only when the source is settled, then draw without making opportunistic changes after seeing an unfamiliar name.

Save the full draw record for your private campaign file and use a smaller result for public communication. The record includes settings, counts, the eligible pool, and excluded row reasons as well as winners and reserves. It can explain what you ran, but it is operator-generated rather than tamper-proof proof. Keep unrelated contact details out of imported comments, and inspect any record before forwarding it to a group or attaching it to an announcement.

Write down the replacement approach before you need it. The reserve and redraw record guide explains how an ordered backup list can preserve the original selection. Keep this entry decision sheet with that handoff, and browse the free tools directory for other preparation work. The completed outcome is a pool with a defensible denominator, not merely a spreadsheet with fewer repeated names.

Questions about Duplicates and Entry Rules

Continue with another giveaway workflow

Return to the Giveaway Winner Picker to review your entries, or explore another guide below. Browse all free tools for other preparation tasks.