<img height="1" width="1" style="display:none;" alt="" src="https://analytics.twitter.com/i/adsct?txn_id=nups8&amp;p_id=Twitter&amp;tw_sale_amount=0&amp;tw_order_quantity=0"> <img height="1" width="1" style="display:none;" alt="" src="//t.co/i/adsct?txn_id=nups8&amp;p_id=Twitter&amp;tw_sale_amount=0&amp;tw_order_quantity=0">

Customer Retention. Automated

New Year, New Data Structure: How to Clean Up and Validate Your Data

Posted by Polly Flinch on Jan 10, 2017 8:00:00 AM
Find me on:

New Year's New DataAs we all know, data is the new currency in the eCommerce world. It’s used in many different platforms for many different initiatives. So what happens if your data is inconsistent or low quality? Data quality is imperative to creating successful campaigns no matter what platform you are trying to integrate with. I sat down with our Integrations Team to pick their brains on the ins and outs of data validity and gain a better understanding of how retailers can, rather easily, increase the quality and consistency of their eCommerce data.

Q: Explain the role of your team at Windsor Circle?

Lucas & Jonathan: Big picture, we coordinate with a client’s technical team to gain access to and integrate with the client’s technology stack. More granularly, we troubleshoot plugin issues, ensure data integrity is up to par, and manage the data flow from eCommerce to ESP, while appending data analytics.

Jenn: I head up the project management aspect of the integration, coordinating timelines with the client and with our teams internally. I provide updates to the client and relay the information from Lucas and Jonathan.  

Q: What Can Retailers Do to Mitigate Data Issues?

A: At a high-level, there are two things that retailers can do:

    1. Ensure you have adequate process around product catalog information and how it’s kept updated. Bottom line, make sure it’s consistent and adhered to.
    2. Make sure you have a central location where all of your product and order data lives. Switching eCommerce platforms is something that happens, but when it does, make sure that your SKUs stay the same from platform to platform. Likewise, be mindful of either preserving customer IDs across platforms or setting them up in such a way that there is no confusion or overlap with duplicate IDs.

Q: If I asked for a list of things retailers could do to increase data quality and consistency, could you provide one?

A: Absolutely, we look at data on a daily basis and there are some relatively easy fixes that retailers can implement to ensure a quicker integration with higher quality and more consistent data.


Missing product categories mean holes in your data, do a walkthrough of your product setup and make sure that all of your product categories are accounted for.


Retailers will often remove a product from the website and forget to mark the product “inactive” in the backend. While this is not an issue on your site per say, if you enable onsite personalization or send emails with dynamic product recommendations, you may end up showcasing a product that is defunct since the system has the product marked as “active.” This can cause some friction with the user experience, since the product URL will be invalid, and the customer will receive a 404 error. To fix this, do an inventory of your existing product catalog to ensure all your products are marked correctly and set up a process for removing product pages that incorporates marking the product inactive.


If an image link is invalid, then the image will appear broken, if the web link is invalid, then the user will not be able to access the product page. Both lead to a sub-par user experience. Similar to the issue with an inactive product page, you should do an inventory of your existing product categories and double check that your links and image URLs map correctly.


There are two standard methods for organizing your product SKUS; a nested product setup where you have parents and children or a flat catalog where you have one SKU  for every discrete product. If you use a SKU setup that varies from either of these two, wires can get crossed and your data could end up being inconsistent.While this is a larger project, it is extremely important. Decide whether you want a nested product setup or a flat catalog and get to work re-organizing.  A nested SKU setup allows for more sophisticated product recommendations within Windsor Circle, but it is not required for product recommendations to be used in our emails. Each eCommerce platform differs in how to go about updating this information, so make sure to check with your eCommerce provider for their best practices.


Every once in awhile we’ll run into a situation where the eCommerce subscriber data is reliable; however, 90% of their customers have opted out. No matter how valid your data is, if the data is low quality you’ll get a low quality output. The first step to fixing a problem of this sort is to rethink how you’re marketing to your existing customers (email may not be your best option at the moment), and create a new plan of attack for net new customers who are making purchases. Pinpoint where your unsubscribes are coming from by assessing your marketing strategy. Do you see an increase in unsubscribes after a certain email? After a certain number of emails? After a certain number of days? Use this information to create a more informed marketing strategy, this will strengthen your data set and bolster your retention efforts.


Take the following example a retailer hosts product images on one eCommerce platform and their data is in another eCommerce platform. They now have to integrate with two different platforms and marry the data together. While it’s not impossible it can create inconsistencies, and longer integration times. The easiest way to fix this problem is to host all of your data in one place, images for products and customer data should live in the same place. This will ensure that your data is clean and consistent across the board.


Two common mistakes we run into, retailers haven’t enabled eCommerce tracking and transaction IDs that are stored in Google Analytics are not stored in their eCommerce platform.  Learn how to enable eCommerce tracking here. Explore how to ensure transaction IDs are stored in both Google Analytics and your eCommerce platform here.


When pulling product images dynamically, inconsistent image ratios can cause problems. Bottom line, pick a ratio and stick with it for your high-res and web optimized versions. If you only have time to resize one, your web optimized images are your top priority. Consistent image ratios make the integration process exponentially easier, not to mention it creates a more enjoyable customer experience.


One of the biggest barriers retailers run into when doing any type of data validation is getting buy-in from the right teams across departments. Depending on the complexity of the integration and your team’s technical ability, we recommend bringing in development or IT technical resources. To ease the process, make sure someone on your integration team has intimate knowledge of your site and data infrastructure.


To recap, while updating and validating your data may not sound like the most enjoyable project or even the easiest, the better your data quality, the better served you will be when using this data in marketing campaigns. Learn more about integrating your software with Windsor Circle here.

Topics: Best Practices