Key Considerations For Importing Data To Your eCommerce Platform

Data is a huge part of good eCommerce. Detailed and accurate data gives you the ability to create a better, more streamlined user experience. It can allow you to personalize a customer’s information and recommendations, which can increase their sense of trust and even generate more revenue. However, when you make a change to your eCommerce platform, it can be complicated, and therefore expensive, to import all your data.

Click here to download these key considerations in PDF format.

Data is a huge part of good eCommerce. Detailed and accurate data gives you the ability to create a better, more streamlined user experience. It can allow you to personalize a customer’s information and recommendations, which can increase their sense of trust and even generate more revenue.

However, when you make a change to your eCommerce platform, it can be complicated, and therefore expensive, to import all your data. 

The reason is that, typically, things don’t line up perfectly from one system to the next. Oftentimes there are holes or historic data needs to be made up, since data in one system does not have an equivalent in another system.  Values need to be transformed to align to new meanings.  Sometimes it is not clear what the data even means. Other times, it’s an opportunity to put your information in new categories for business reasons.

If your data is like food in a grocery store, one system might lump all the produce together, while a new system might have separate categories for fruits and vegetables. Where does the tomato go? How about the avocado? Should you be sub-dividing the Gala apple vs. the Honeycrisp apple? 

One system had a unique internal identifier for every item in the store another just used the UPC code.  You historically carried greeting cards but no longer do, should you bring those sales and products into the new system? There is a lot of analysis involved in trying to sort through data and make sure it ends up in the right place and with the intended consequences.

Some issues are easier to solve than others, but depending on the quantity of data involved, importing it can quickly become complicated and time-consuming. 

In eCommerce migrations we have facilitated, when customers want to keep all of their data, we end up writing scripts to analyze tens of thousands of pieces of data (or more) and convert it to fit the requirements of the new system. It involves hours of testing and combing the systems to make sure the end result is accurate and prepared for the go-live date, since much of the sales data has to be converted up until the old system is shut-off and the new system takes over.

Before you invest the resources in converting your data, here are a few key things you should consider. Click the image below to download these pointers in PDF format.

Three Types of Data 

There are three primary types of data we deal with on an eCommerce site:

  • Product Data: This includes product titles, descriptions, images, prices, warehousing, technical literature, etc. Data changes here can usually be frozen and thus kept static until after go-live.
  • Customer Data: Contact information such as name, address, phone, email, company details, contacts within the company, username, password, etc. Sometimes payment information is also included here (although special considerations are required in that case).  This data cannot be completely pulled until the moment of go-live as you will be taking orders up until the old system is no longer available.
  • Order History: Order history combines customer data with product data along with order dates, shipping and delivery data, and other information.  This data also must be transitioned at go-live so all orders from the old system are accounted for.

Each type of data plays a different role in the functionality of the site. Every company has a unique situation depending on the depth of your product catalog or the length of time you’ve been taking online orders. It can also be impacted by your sales channels and other systems that may be connected to your customer service or fulfillment processes.

It’s possible to import certain types of data and not others to save money and create the best transition and end result. By looking carefully at your company’s current situation and future priorities, you can determine the right course of action.

To Import or Not to Import?

While you can choose not to import either your customer data or products, you can’t generate the order history without the other data. So the first thing to determine is, “Is it worth having all of the history or not?” 


It’s easy to jump to conclusions and assume “of course we want all the data we had in our old system!” However, upon closer look, sometimes starting fresh can give you a chance to improve your site in useful ways. Consider these questions:

  • How good was your data in the first place? Is it scrubbed and up-to-date? Are there old products or customers that no longer need to be in there?
  • Are your product listings similar, or are you adding a lot of new info? Is this a chance to revisit your descriptions, images, and other product details that aren’t on your current site?
  • How would it impact the user experience if you didn’t have the order history or contact information? How important is it really? And how could you communicate it in a way that is positive for the customer? 

Timing Is Key

If you decide to import your data from an active system to a new system during a migration, the timing is critical. Because you are still taking orders on your old system, making the move without losing data can be tricky. 

When we manage a large migration for our clients, here are a few things we do to ensure a smooth transition.

First, we write an automated script or program to do the data importing automatically so it can go quickly. That way the site will only be down for a couple of hours, rather than weeks! 

Second, we test the scripts during a “mock go-live” where we emulate the migration process in a testing environment to uncover any bugs and fix them before the full implementation.  

Finally, it is important that business users review the data.  Don’t let your team be lured by the thought that the program ran without errors, therefore, it must be correct.  There are so many ways to successfully run an import that has bad results.  Also, tie your data inputs to your data outputs.  If there were 7,456 transactions in the old system, then there should be 7,456 in the new system or at least a detailed explanation accounting for where all 7,456 ended up.

Is your company considering migrating your eCommerce site to a new system? Having a plan for importing your data is a critical part of that process (among other things!). Schedule a free Migration Consultation, and we will be happy to talk to you about your specific situation.

David McDonald

Director of Development

About

David McDonald

David serves as the Development Director at Brilliance Business Solutions and is an Episerver Certified Developer.  He has over 25 years of software development and database knowledge from companies such as NASA and Rockwell Automation that equips him to provide architecture and expert services to Brilliance clients.

Related Articles