Building a Shopify app: the decisions before the code
Most briefs that arrive as “we need an app” are a Flow, a metafield or an extension. If you do need one, four choices made in week one decide what it costs for years.
BLOG
Specific, opinionated writing on Shopify Plus architecture, migrations and selling internationally, often including the argument for not buying the more expensive option.
Most briefs that arrive as “we need an app” are a Flow, a metafield or an extension. If you do need one, four choices made in week one decide what it costs for years.
A bulk mutation is a machine for making ten thousand changes quickly, which is also what it is when the change is wrong. The safety is in the rehearsal, not the API.
Put performance in the redesign as a constraint, not as a workstream. A speed project bolted onto a rebuild gets cut, and a speed project run just before one gets thrown away.
It recovers less than the report claims, it only reaches people who already gave you their address, and it is usually a plaster over a checkout problem you could fix.
Every case study you have read about this measured the wrong number on somebody else’s store. This is what the choice turns on.
Export the catalog, count what is broken, and decide what you are not moving. Every hour spent here is an hour you do not spend debugging a cutover.
A CDP does not fix your customer data. It gives the mess a new address, and charges you per profile for the privilege.
Extensions let you put almost anything into checkout. Almost everything people put there costs more conversion than it earns.
Shopify’s built-in returns do more than most stores realize. Three gaps justify buying a platform, and here is the arithmetic that tells you when.
Dragging tiles works until it does not, and it fails silently. The grid keeps rendering long after anyone stopped maintaining the order.
Product pages chase terms nobody searches. Collections match how people actually look for things, and most stores ship them as a bare grid.
Your 3PL already runs a warehouse management system. Buying a second one rarely fixes the problem you have, which is almost always that nobody agreed who owns the stock number.
NEXT STEP
A senior Shopify engineer reviews your storefront, theme performance and checkout, then sends a prioritized list of fixes.