The story about Golden Record and how you get total control over your Content Provided data.

Golden Record logo

A couple of years ago we had a case where a customer needed help with their Specifications. The loading times for the specifications where way to long and it made it very time consuming and frustrating for them to work with.

Golden Record

So what had went wrong? It turned out that they were building their specifications with a custom built integration to CNET. This integration let CNET become master of their specifications. So basically it created new “SpecificaionFieldType’s” whenever an attribute existed in CNET but was missing on the Specification Template in PIM, and it was always created as a CVL field. This lead to huge Specification Templates connected to 1000 of CVLs that could have 1000 of keys and values in it. A lot of the values were also duplicated since CNET could have different keys for the value “Yes”. There were even some cases where fields got created with the same “DisplayName” but with different ID’s from CNET. CNET is an awesome Content Provider with an incredible amount of specifications, but their data structure does not fit a PIM system.

Golen Record

So after some analysis of our Customers specifications and how data was sent and structured by CNET we presented a solution for our customer. The “Golden Record” was born.

In the first version of Golden Record the focus was to decrease the amount of fields and to decrease the amount of CVL values and keys in the system. We stopped the integration from creating more “SpecificationFieldType’s”. We implemented functionality in Golden Record so that our customer could merge “SpecificationFieldType’s”, merge CVL values and keys, transform Specification data types, map CNET attribute ID’s to already existing fields and set “Update Behavior” on how to handle new information from CNET on existing specification field data. Version 1 of Golden Record had a quite basic UI where most of the mapping had to be handled by text boxes and drop down options. The project was a success and loading times for specifications went down for the PIM users. But there were still things with Golden Record that could be improved. For example it was quite difficult for the users to learn how to work with it and there were wishes from the Customer on getting even more control over their specifications and content provided data in Golden Record. So it was time for version 2!

Golden Record

This time we decided that Golden Record should be the main administration system for Specification Templates in PIM and we focused a lot on usability. The goal was to make it easy and fun to work with Specification Templates and together with CNET data we could get most of the structure “for free” and also improve performance in PIM. To improve performance we decided to completely get rid of CVL’s so we cleared all our Customers Specification Templates and let them start from scratch giving them total control over their Content Provided data and Specification structure.

So we created a couple of Apps in a web application and implemented drag and drop functionality to administrate Specification Templates. With the drag and drop functionality the users could create  “SpecificationFieldType’s” and do mapping of CNET data at the same time!

Golden Record

The result of Golden Record version 2 was amazing and I can stick my neck out and say that our Customer today have the best looking and most relevant specifications on their products in their market. Our customer is now working with their specifications every day in Golden Record.

If you want to learn more about Golden Record you can have a look at our technical demo:

 

You can also find Golden Record here at inRiver's Market Place over here. Golden Record works for both on premise versions of inRiver PIM and in iPMC.

 

Daniel Jansson

Daniel Jansson
inRiver PIM Developer
daniel.jansson@sigma.se

2018-09-07

PIM
Unified Commerce