US TikTok Settlement CSV Audit Guide
A TikTok settlement CSV audit means downloading US TikTok Shop finance exports, normalizing signs and transaction types, joining every row to an order ID where possible, and running pivot checks for fee, shipping, refund, and reimbursement gaps. Statement totals alone hide the mismatches that become claims.
This guide is the spreadsheet-level companion to payout reconciliation. Reconciliation defines what “expected” means; the CSV audit proves whether your exports actually contain the rows needed to support that math.
TikTok Seller Center (US) labels and column names change. Keep the original file, record the export date, and map columns into a stable working schema before you pivot.
Which settlement CSV fields matter most?
Not every column belongs in every check. Prioritize fields that let you reproduce a USD amount for one order:
| Field (or equivalent) | Why it matters |
|---|---|
| Order ID | Primary join key for order-level audits |
| Statement / settlement ID | Ties rows to a payout period |
| Transaction / adjustment ID | Deduplicates re-exports and later corrections |
| Transaction type / adjustment type | Separates sales, fees, shipping, refunds, reimbursements |
| Settlement amount (and related amount columns) | The USD value that hit the statement |
| Currency | Prevents mixed-currency false positives |
| SKU ID (when present) | Needed for SKU-level fee checks |
| Transaction / posting date | Distinguishes timing delays from permanent gaps |
If a row lacks an order ID, keep it—do not delete it. Reconcile those rows in a separate “unlinked” bucket using statement ID, adjustment ID, amount, and date.
How do you normalize signs and transaction types?
Decide one sign convention for the working sheet and apply it to every import:
- credits (sales, reimbursements, rebates) as positive; or
- keep TikTok’s native sign and document it in a header note.
Never mix conventions across shops or months. Then map original labels into a small category set while preserving the raw label beside it:
- sale;
- fee;
- shipping;
- refund;
- reserve;
- adjustment; and
- reimbursement.
Preserve both columns (raw_type and category). When Seller Center renames a label, your historical pivots still work and your evidence pack can cite the exact export wording.
Which pivot checks catch missing money?
After normalizing, run these pivots before reading individual tickets. They surface clusters faster than scrolling the CSV.
| Pivot | Group by | Flag when |
|---|---|---|
| Net by order | Order ID | Net is materially below expected, or zero when a sale should exist |
| Fees by order | Order ID + fee category | Fee total exceeds the policy basis or appears twice |
| Shipping by order | Order ID | Shipping debit with no later offset after carrier audit |
| Refunds by order | Order ID | Refund debit exceeds approved refund, or admin fee looks high |
| Reimbursements by order | Order ID | Eligible logistics event with no reimbursement credit |
| Unlinked rows | Statement ID | Material amounts with no order ID and no explanation |
Use the same expected-versus-actual logic as the shipping fee discrepancy checklist and refund admin fee guide, but feed those formulas from the CSV pivots rather than from memory of the order screen.
How do you join orders to settlements in the sheet?
- Export the settlement / finance transaction CSV for a closed period.
- Export the matching order report for an overlapping date window (orders often settle after the order date).
- Create a unique key on order ID in both sheets.
- Left-join settlement rows to orders so every finance row keeps its amount even if the order row is missing.
- Add a helper column: has_order_match (TRUE/FALSE).
- Sum settlement amounts by order ID; compare to your expected net formula.
- For FALSE matches, triage in the unlinked bucket before dismissing.
Common join failures: truncated order IDs from spreadsheet scientific notation, leading zeros stripped, mixed shops in one file, and filtering the order export too narrowly so late settlements have no order row.
Store order IDs as text on import. A single scientific-notation conversion can break hundreds of joins and invent false “missing order” exceptions.
What export hygiene should multi-shop sellers use?
- One shop per working workbook (or one clearly labeled sheet per shop).
- Filename includes shop ID/name, statement period, and export date.
- Never overwrite an earlier export; archive the raw CSV unchanged.
- Record currency and confirm USD for US shops before pivoting.
- Note the Seller Center filter used (date range, statement ID, transaction types).
- Re-export later statements when hunting offsets—do not assume the first file is complete.
- Keep a change log when you remap transaction labels.
- Separate API/sync ingest from manual CSV uploads so you know which source produced each row.
Mixing two shops in one pivot is the fastest way to create fake gaps: one shop’s reimbursement will never offset the other’s shipping debit.
What does a practical settlement CSV audit checklist look like?
- Confirm shop, period, currency, and raw file checksum or archive path.
- Import order IDs as text; verify no scientific notation.
- Normalize signs; map types; keep raw labels.
- Deduplicate on transaction / adjustment ID.
- Join settlements to orders; count unlinked rows.
- Pivot net, fees, shipping, refunds, and reimbursements by order.
- Search later statements for offsets on every material gap.
- Classify exceptions (high / medium / low confidence).
- Build an evidence pack only for unresolved gaps.
- Track status through filed, recovered, rejected, or dismissed.
For logistics-shaped gaps, confirm eligibility with the lost-in-transit reimbursement guide before filing from a pivot alone.
How does RecoverPayout use settlement data?
RecoverPayout ingests US TikTok Shop settlement/statement transactions (via shop sync or uploads), stores them as structured rows, and runs deterministic reconciliation rules against orders. The same gaps a careful CSV audit finds—unsupported fees, shipping variances, missing reimbursements—surface as review items with expected vs. actual amounts.
You still decide what to file in TikTok Seller Center (US) or U-PIC. RecoverPayout does not submit claims for you. Browse the RecoverPayout guides or see how the product detects and tracks reimbursements.
Sources and policy note
- TikTok Shop: 1099-K and Tax Form FAQs
- TikTok Shop: Logistics & Fulfillment
- TikTok Shop: Referral Fee Updates
This guide addresses the US market. Confirm current US Seller Center export labels, statement fields, and filing instructions before relying on a column name from this article. Operational guidance only—not legal, tax, or financial advice.