Author: Anya Skvortsova is a Certified Master Anaplanner and CoE Lead at Asana.
Problem statement
When exporting the data from Anaplan module, Anaplan’s default chunk size is 10MB. The common challenge is that the chunks produced may have the rows split not just into the separate files but also split the row in the middle of the field. For example, with the data below, we may end up having a portion of a record in Chunk 1 (“12345678,1/5/25,Ea”) and second half of the same record in Chunk 2 (“st,CustomerAA 91011121,2/3/25,West,CustomerBB”).
Such default chunking behavior makes merging the chunks into single file less intuitive (unless Anaplan Connect or Python are used). The alternative solution, described here, includes breaking the data down into logical segments before exporting.
Setup
In our case, when module’s data manually exported the file is >100MB (17 columns x 700k rows). We use Snaplogic to run export action in Anaplan and then load the exported data to Salesforce. To avoid default 10MB chunks with the rows split in the middle of the field, we chose to create several saved views of the exported module, each view resulting in a file size <= 10MB.
To create a column with chunk’s sequential number, the following line items were added to the module:
The reference system/lookup module was created to populate the formula for “Chunk” line item (to avoid any hardcoding):
Next, the “Chunk” line item formula is created:
And the last step, is to create the series of saved views filtered based on the “Chunk” line item value:
Once the export definitions for each saved view are created, Snaplogic runs the process that results in several .csv files that are fixed number of rows and under 10MB in size, without the issue with the rows split in the middle which allows for easier subsequent merging in Snaplogic in further processing and upload to Salesforce.
Questions? Leave a comment!
