Five Tips on E-commerce Data Migration

Over the years, AISN team members have designed, built, and optimized multiple e-commerce websites using waterfall and agile development methodologies and processes as required. From brochure websites that tell a story to data analytics sites with millions of data points to e-commerce sites handling a healthy sales volume, you could say we’ve been around the application development block more than a few times. Throughout our years building digital storefronts (that’s to say, either developing a brand new site or re-platforming/refreshing an old site), we’ve learned quite a bit about successful e-commerce implementations. Here are five tips on e-commerce data migration for successful digital storefronts:

1. Know Your Data and Keep it Clean

An e-commerce platform migration is a big deal. However, E-commerce data migrations can be an even bigger deal because the former’s success hinges on the latter. Suppose the client already has a digital storefront. In that case, they may want to redesign it for aesthetic reasons or change platforms because the old site’s platform cannot scale with the business’ growth or deliver a first-rate user experience. A platform migration is about taking your client’s entire online store off one platform and transitioning it to another, including moving design features and integrated apps.

On the other hand, you have the corresponding data migration, which is a transfer of product and customer data. As with many government sites that we’ve re-platformed, this typically involves moving data from legacy systems to the new site — and a location that better facilitates data management. That can get complex, very complex. E-commerce sites handle a lot of data, so getting the migration plan and implementation right the first time is critical.

As part of developing the data conversion/migration plan and process, we strongly recommend that the team familiarize themselves with the product and customer data by searching for anomalies. Finding and correcting data in the source system before the migration can have immense benefits and help ensure the project is completed on time and within budget.

We recommend looking for the following:

• Field Inconsistencies
  • Example: Are all the states two letters?
  • Example: Are all the zip codes in the same format, or are some five digits while others are nine? Do some have hyphens where others do not?
• Duplicate Data
  • Are there duplicate customers? Use the ‘Find Duplicates’ conditional formatting function in Excel to help in the search.

2. Prioritize the Data That You Need

Project scope creep is always a danger. One of the best ways to mitigate scope creep is to prioritize features and functions realistically into categories such as ‘must have’ and ‘nice to have.’ Data migration should likewise be prioritized. This is a golden opportunity to be strategic about what you need to bring over versus just bringing over a pile of legacy data for the sake of bringing it over. Reserve the ‘must have’ category for those items truly needed to launch the platform.

For example, most online stores’ product lists are the only ‘must-have’ data set. The new platform could technically launch without historical sales data and existing customer lists. It is not the ideal situation, and the risk is low in this case, but the root of the lesson we have seen and learned holds. Also, launching without the ‘nice to haves’ does not mean it never gets done. It simply means that it potentially happens after the launch.

3. Use This Opportunity to Rethink Your Data Structure

This is yet another golden opportunity to think about structuring your product data. How did you use it in the past? Do you like the legacy structure? Is it strategically capable of serving you in the future?  Consider the difference between tags and meta fields in a platform such as Shopify. How does that information get used on the site’s front end? How will you be updating it later? If you think you will continue to use the same standard import tools for creating your products, have you considered that you may want to choose something that can be more readily updated via those tools?

4. Test and Retest

Being thorough helps ensure that no data gets lost but ends up exactly where it should be. Check it for accuracy after the first pass at a transfer to get the client data over to the new location. Perform front-end data alignment testing and then resync it to ensure your store is updated. Before launching a new site, we perform front-end alignment testing once more to ensure that the data have been transferred faithfully.

5. Once Is Enough, So Be Strategic

Plan to get it right so you only make this move once (you hope). Remember, you’re shifting the entire foundation of your business — right underneath your customer base. Even if AISN does its part seamlessly, we don’t want you to migrate everything in, launch, roll along, and discover that you’re missing a bunch of critical data you needed. — Or, on the flip side, discover that you have too much unnecessary data dragging the site down and impacting site performance negatively. Then, you are stuck in the same rut you were in with the old platform. So, be strategic. Once is enough to move your entire foundation.

Are you ready to transform your e-commerce site? We have deep experience in this area and would love to help you build it out. Is it worth a conversation? Please email us at [email protected] for a free consultation.