Data Processing

The File:

We accept a wide variety of custom data formats from customers. All data files are not alike though. We prefer to see simple, easy to manage text files, with fixed field sizes or data that is tab delimited. We also welcome .DBF database format files.

Here is slice of a good sample fixed field file:

Flat File

Note how the Order rows start with "O", package rows start with "P" and as might be expected, item rows start with "I".
There should be one order line for each order, maybe multiple package lines if your client is shipping multiple box orders and probably multiple item lines, one for each SKU that's in the box.

We like to see OrderIDs not get repeated and the same thing goes for PackageIDs. Easy right?

Here is an example of what we would not like to see.

Bad File

Why is the above spreadsheet sample bad? First it's hard to tear down into its data units. You need to be able to read five lines to get the complete order record, and then go back up five rows so that you can start reading the SKUs. And what if there just happens to be more than five SKUs? Finally, what's up with the third record's quantities? Did their poor server just get tired and stop putting quantities into their files? Data files should be simple, cleanly segmented and very predictable. Rows with missing data fields are not what we want to see.

How it's sent:

We also offer confirmation emails when one of our servers receives an automated file drop.