Automating Magento 2 Migration - our experience

If you are familiar with Razoyo, you know that we started as (and still are) merchants whose business grew into a thriving development resource for hundreds of other merchants. So, like most merchants, we experienced both excitement about Magento 2 migration and hesitance to celebrate – knowing the potential effort required to migrate between platforms.

We are no  longer developing new stores on Magento 1.9.x and have switched to creating new stores on Magento 2. However, Magento 2 migration is a different rodeo altogether, as we say here in Texas.

Why Magento 2 Migration?

So, before I delve into our recent experience, let me just give the top reasons a merchant might want to consider migrating to Magento 2:

• The admin panel is much more workable. Sure, the first time you log in you will feel like you have moved into a new house. Some things will require more effort to find and habits will need adjustment. However, this admin panel is 10x more intuitive than the old Magento admin panel.
• The admin interface is more modern. In the product profile alone, categories are added like tags, you choose your base image by simply clicking on it, and images are rearranged by dragging. It’s more like working in Wordpress - it has more of a content management system feel than a developer’s tool feel to it.
• Magento simplified the product creation process. Merchants who merchandise their products as configurables or bundles will love the new look and feel for creating products. We used to often create (or upload) the simple products first. Now, we’re creating the configurable product and letting Magento 2 do the heavy lifting.
• Product videos. You can add YouTube or other videos to the product image manager EASILY. Love it!
• Upload spreadsheets are more manageable. Many merchants have plunked down the $1300 for uRapidflow simply because of the ‘blank lines’ issue in Magento 1’s upload/download spreadsheet format. The boys in Austin (yep, that is where much of the Magento 2 planning was done) mercifully changed this to allow for multiple values within individual cells. As we move merchants over, we’re evaluating if they need to continue uRapidflow or not (there are still reasons to use it and it is available for Magento 2).
• Customizing the base theme is more efficient. Responsive design, fewer design assumptions and open layouts. Luma, we love you.
• Security. The upgrades to the security features are significant.
• Efficiency. Yep, it works faster.

There are other reasons from a developer’s standpoint, like being able to use more updated design tools like LESS, better versions of PHP, more reliance on PHP classes and less on Magento’s magic classes.

When using the migration scripts to move the data over, I have to admit, those scripts worked surprisingly well. Products and orders moved over seamlessly. Attributes and attributes sets were created (though attribute groups were renamed with a ‘Migration_’ in front of them - we actually saw this as a plus). Products even had the same product id numbers and category assignments.

Price rules, content blocks and most configurations came over with no problem.

So, what were the problems?

• You may have to reconfigure your payment method. We use Braintree and its settings did not migrate though our Paypal Express settings did.
• Theming is completely different. If you’ve been thinking of a major redesign for your UI, that’s a good time for Magento 2 migration. After all. you’ll pretty much have to redo your theme, anyway. I would note that we didn’t expect the design to come over.
• Your extension data probably won’t migrate, especially if you are using a complex extension or a lot of them. Some of the extension developers have created individual migration scripts, which, though we haven’t tested them, probably range from bulletproof to worthless.
• Some bugs persist. I wouldn’t say there are MORE bugs than Magento, but, there is a different set of BUGS. Overall, though, there are fewer conflicts as far as I can tell (my devs may flame me for saying that… check the comments!).

Overall, we’re impressed. The folks in the Austin-Culver City-Kiev cabal that got Magento 2 out of the door (finally), really took migration to heart and did a great job with it. Most merchants will still need to turn to a developer for help. However, at least we developers don’t have to charge you an arm and a leg for writing custom migration scripts. Rather, we can both concentrate on building great stores and not wrangling data.