If your company sells online, you know keeping product data accurate, complete, consistent, and up to date is critical. This is even more challenging when using a third-party data source like AD, Syndigo, 1WorldSync, or DDS. See our Product Data Quality article for additional tips and strategies. But what does that process look like for your organization, and where are the biggest opportunities for cost reduction and accuracy improvement?
If you are manually importing data today, have you considered how to auto-import data into your PIM?
Factors to Evaluate Before PIM Data Sync
- Do you have a lengthy manual process to obtain, normalize, and standardize data before uploading into your PIM?
- Do the products you sell have complicated attribute models that require mapping and validation?
- Are you displaying all the fields you want from the third-party feed on your eCommerce site, or are useful fields being dropped?
- Are feed formats and ranges easily ingested by your PIM data model? For example, the feed provides 32 degrees but your model expects 32°.
- When a product is out of stock, can you immediately show availability, substitutes, or replacements based on ERP status?
Each of these factors poses technical challenges you should solve before implementing an auto-import mechanism. With a deep understanding of product data and PIM architectures, Layer One has repeatedly addressed these problems.
Client Example: DDS Feed to inriver
One wire and cable client struggled to import a DDS feed into their PIM, inriver. Five channel managers were manually mapping multiple fields for every SKU at import time. Their previous partner had said this was the only way.
Auto Data Sync Solutions
We addressed two core issues:
- How to map feed fields to an inriver Specification template.
- How to convert the feed’s values into the PIM’s expected format.
Their data model used separate fields for minimum and maximum temperatures, while manufacturers provided ranges and mixed scales. Our import tool identified, transformed, and placed values into the correct fields. We also surfaced additional feed fields the client wanted but had never captured, then mapped them. Finally, by connecting ERP status to the PIM, we detected out-of-stock reorders, phaseouts, and discontinues to auto-suggest substitutable products. The result was a fully automated pipeline that saved dozens of hours per import and increased accuracy.
Technical Example
At a high level, products in the incoming feed are looked up in the PIM, their Specification template is inferred, and a configuration file is used to connect feed fields to Specification fields.
In the example above, the inriver Specification “Heat Shrink Tubing” maps the feed field “S Ratio” to the Specification field “Shrink Ratio.”
Sometimes, a field’s data needs to be converted to a different format. The Layer One tool applies transformation rules during import using configuration entries like this:
In the transformation above, any temperature string containing the word “degrees” is converted to the degree symbol.
Results and Next Steps
Automating product data ingestion reduces manual effort, improves data quality, and speeds time to market. It also unlocks downstream wins in search relevance, merchandising, and customer experience.
Layer One tackles business issues like this every day. If you would like to know more about automating data feeds into your PIM or solving other product and eCommerce hurdles, reach out to us.
Product Data Quality is the third of Layer One's 
See the other 13 Focus Points for Manufacturers and Distributors here!