Orderly E-mails + Collection deletion "quick-fix"
Were you impaced by the incorrect collection deletion incident on the 31st of March? Here you might have your quickfix that might help you contain some of the damage this app might have caused to your store. Get our RELOFLEX - Smart Collection Automation app that helps you automatically (re)create all your collections based on your product data and metafields. We write "quick-fix" because while it might not recreate what you had, it may save you significant time and potentially thousands of dollars, depending how impacted your store is.
If you’ve ever worked with a large catalog, you already know this isn’t just a visual problem. Collections are structure. Navigation breaks. SEO pages disappear. Internal linking gets messy. And rebuilding everything manually is not something anyone wants to do under pressure.
This post is not about the incident itself. It’s about a practical way to rebuild and stabilize your collection structure quickly using product data.
Collections are just rules
Most collections follow simple patterns:
- Product type = Shirt
- Color = Black
- Brand = Nike
Even if the collections are gone, that logic still exists in your product data.
Instead of rebuilding collections one by one, you can rebuild them from:
- Product type
- Vendor / Brand
- Tags
- Metafields
Step 1: Clean your product data
Before doing anything, make sure your data is consistent.
Look for inconsistencies like:
- Different naming (Shirt vs shirt)
- Different tags (Black, black, blk)
- Missing or incomplete metafields
If your data is messy, your collections will be messy.
Step 2: Decide your structure
Pick the dimensions that actually matter for navigation and SEO.
Simple structure example:
- Product type → Shirts, Shoes, Hats
Two-level structure example:
- Product type + Color → Black Shirts, White Shirts
Brand structure example:
- Vendor + Product type → Nike Shoes, Adidas Hoodies
Custom structure example:
- Metafields → Slim Fit Shirts, Waterproof Jackets
Keep it simple at first.
Step 3: Generate collections from data
Define your structure based on product data.
Example setup:
- Product type + Color
Your catalog might look like:
- Shirt / Black
- Shirt / White
- Hoodie / Black
This results in:
- Black Shirts
- White Shirts
- Black Hoodies
Combinations that don’t exist won’t be created.
Step 4: Avoid useless collections
Large catalogs create too many combinations.
Example:
- 40 product types
- 10 colors
→ 400 possible collections
Most of these are not useful.
Only create collections when there are enough products behind them.
Example:
Only create collections with at least 5 products
This removes:
- Thin collections
- Low-value pages
- Navigation clutter
Step 5: Let collections reflect reality
These collections are based on product data.
That means they follow your catalog.
If you add:
- Green Shirts
→ A matching collection becomes relevant
If you remove:
- All Red Shoes
→ That collection is no longer relevant
Your structure follows your actual inventory.
Step 6: Refine where needed
This approach rebuilds structure quickly.
From there, you can:
- Adjust naming
- Create curated collections
- Control what is featured
This gives you a solid baseline to work from.
Summary
If collections disappear:
- Clean your data
- Define your structure
- Generate collections from product combinations
- Filter out weak collections
- Refine manually
This lets you rebuild your structure without starting from scratch.