Skip to main content
All CollectionsAdd or upload data to SKU Science
💾 How to prepare your first file for upload
💾 How to prepare your first file for upload

This tutorial gives instructions for uploading your first historical sales file to SKU Science

Updated over 3 years ago

As you will see below, there is no need to collect a lot of data for this first upload. Note however that to benefit from forecasting models with seasonality, you must load at least two years of historical data per item, otherwise, detection will still be automatic but with less sophisticated models. If you upload at least three years of historical data, the platform will in addition compute KPIs for the last 12 periods and get even more accurate forecasts.

Define useful fields for your upload

First, our goal is to easily load historical sales/demand data to generate item level forecasts. You can then define other fields specific to your activity which will be used in particular to obtain aggregated data in a few clicks, or even generate forecasts on a combination of fields (eg article by customer).

At this stage, it suffices to provide in your file at least a column containing your articles.

As an example, you can download the file used for the demo account from this Google spreadsheet or from this Excel file. If you don't know how to copy a Google sheet, you can get help here.

Then you need to provide your sales history (or demand if you are able to differentiate your actual sales from the demand 👍). In order to do that, two options are available to you.

The different options for managing dates

The wizard knows how to manage 2 types of file.

Option 1: File with dates in the header

The dates may appear in the header of your file (Dates in the header option in the upload wizard), so each line in the file will match an article while columns will match to quantities (date in the header, quantities in the cells).

Option 2: File with dates in a column

Each line can also correspond to a transaction, or a sale.
In this case, we find on each line of the file, an article, a transaction date and a quantity (Dates in a column option in the upload wizard). The wizard will aggregate the data for each item in order to present it correctly on the platform for each forecast period (month, week or even quarter depending on the chosen option).

Date formats

Depending on the source of your data, your dates will have a specific format.

The wizard will try to detect the correct format. Note that several options are available when uploading for the date format including customizable formats. An article explains how to directly use the format present in your file.

Save your file in a XLSX format once satisfied with the content

  • If you use Excel: you can use your file directly to upload it to the platform.

  • If you use Google Sheet: just save your file in this format File> Download> Microsoft Excel​​(XLSX).

Let's now move to the upload process.

Did this answer your question?