> ## Knowledge Base Index
> Fetch the complete knowledge base index at: https://help.booksgpt.ai/sitemap.xml
> Use this file to discover available pages before exploring further.
> Pure-Markdown content can be obtained by appending a '.md' suffix to the content URLs listed in the sitemap (without the trailing slash).

# Import your point of sale's sales history (Clover and other files)

Your point of sale already knows what left the till. Your books know what you invoiced and deposited. Those are two different stories. Importing a sales history file keeps the till's story **beside** the books so you can compare them. It never posts income, never moves stock, and never changes a product.

|| Nothing on this page changes your books, your stock, or your products. **Sales by product/service** shows the file in its own section, **Point of sale history (not in your books)**. That section is never added to **Total income**, cost of goods sold, or margin.

A period has to be whole. Only a file whose first and last day both sit inside the report's dates is counted. A file that only touches those dates is named and left out. That is why you enter the days when you exported, and why a longer file from the same point of sale replaces the shorter ones it covers.

## Import a file

1. Open **All pages** in the left sidebar, then click **Inventory** and [**Products & stock**](https://app.booksgpt.ai/go/products-and-stock).
2. Click **Import sales history**. (You need a role that can change the books.)
3. Name the **Point of sale**. The field starts with the last name you used. Empty on a first file; the placeholder is **Clover**. Use the same name every time so matches are remembered.
4. Enter **First day the file covers** and **Last day the file covers**. The file does not say which days you exported.
5. Choose an **Item sales file (CSV or XLSX, up to 5 MB)**.
6. Clover's item sales export is recognised by its headings; its columns are matched for you. Any other file is matched column by column. **Item name** is required, and **Quantity sold**, **Net sales**, or **Gross sales**.
7. Click **Preview sales history**.

Optional column labels you may see when matching by hand: **Category**, **Gross sales**, **Discounts**, **Refunds**, **Net sales**, and **Cost, as the point of sale reported it**.

## Match each item

Review shows one line per item (category and item name). Money columns appear only when the file has them: **Gross sales**, **Discounts**, **Net sales**, **Refunds**, and **Net after refunds**.

* Match the item to a product, or mark it **Not a product** (a studio fee, for example).
* **Matched before** means an earlier file from the same point of sale used that product. Confirm those together with **Confirm N remembered matches**.
* **Same name** is only a suggestion. Confirm one at a time with **Use ...**.
* Changing a remembered match asks why. Earlier imports keep the product they were imported with.
* **Finish import** appears only when every item is decided.

![Review of a Clover item sales file with matched products, remembered and same-name suggestions, and a Not a product line](https://booksgpt.ai/help/import-sales-history/review-match.png)

## Periods, replacements, and Undo

* A file for a longer period replaces the files of the same point of sale it covers. Those files are kept. **Undo this import** brings them back.
* A file that only partly overlaps another is refused, and the refusal names that file's dates.
* The same file, dates, and matches imported again changes nothing.

| After a quarterly year-to-date export, import it with the dates you chose in the point of sale. The new file replaces earlier ones inside those dates, so the report stays one continuous history.

## See it on Sales by product/service

1. Open **All pages** in the left sidebar, then click **Reports** and [**Sales by product**](https://app.booksgpt.ai/go/sales-by-product).
2. Set **Report period** so it includes every day of the file you want counted.
3. Scroll below the report to **Point of sale history (not in your books)**.

* The section names the days it covers and any gap.
* A file that only touches the report dates is named and not counted.
* It has its own **Export**. That export is never mixed into the report above.
* An item marked **Not a product** is listed under its own name with **(not a product)**.
* A cost that covers only some rows is marked with `*`. A cost the point of sale printed as text (such as N/A) stays blank, never zero.
* Clover's **Net sales** does not take refunds out, so the section shows **Refunds** and **Net after refunds** separately when the file has a refunds column.

![Point of sale history section under Sales by product/service, with Refunds and Net after refunds](https://booksgpt.ai/help/import-sales-history/report-section.png)

## A quarterly routine

Export each quarter from the first of the year, then import that file with those dates.

1. In your point of sale, export item sales from 1 January through the last day of the quarter you just finished.
2. In BooksGPT, open **Import sales history**, keep the same **Point of sale** name, enter those two dates, and choose the file.
3. Confirm remembered matches, decide any new items, then **Finish import**.
4. On **Sales by product**, set the report to the same year-to-date range and read **Point of sale history (not in your books)** beside the books.

||| A longer year-to-date file replaces shorter files of the same point of sale inside its dates. Use **Undo this import** if you need the earlier files back.