I have watched founders get this decision backwards more than any other on a registry build. They fixate on building their own shop, because owning the checkout feels like owning the business, and they treat the "buy it on Amazon" option as an afterthought to bolt on later. Then reconciliation bites: someone buys the crib on Target, it does not get marked off, and the parent receives two cribs and a very awkward group chat. The checkout model is not a choice between your shop and the retailer. It is a two-rail system, and the rails are the easy part. Keeping one honest list state across both is the actual product.
Why two rails exist at all
Babylist made the universal registry normal: add items from tens of thousands of retailers including Amazon and Target, plus cash funds and favor requests, all on one list, and let buyers purchase either through the Babylist Shop or at the original retailer, with 9M+ annual shoppers moving across both. That dual checkout is not an accident of scale, it is the only honest way to be universal. You cannot stock tens of thousands of retailers' catalogues. So you sell what you can profitably sell yourself, and you let everything else be bought where it already lives. The buyer wins either way, and you capture margin where it is available without ever blocking a sale.
The two-rail checkout
Here is the model I sketch for every founder weighing this. Two rails, one list state. The rails differ in economics and operations; the list state is shared and sacred.
The economics, honestly
The own-shop rail keeps the retail margin, which is far more than the affiliate cut that the original-retailer rail earns, often low single digits. That makes the shop look obviously better until you count what it costs. The shop rail carries inventory, working capital, fulfilment, returns and customer service. The affiliate rail carries none of that. So the real comparison is contribution margin after operating cost, not headline margin, and for a long tail of low-volume items the affiliate rail is genuinely more profitable per unit of your effort. The discipline is to own a narrow, deliberate range where volume and margin justify the operational weight, and let the affiliate rail carry everything else. Our breakdown of how Babylist makes money puts these two revenue lines side by side, and how fulfilment works covers the operational cost the shop rail actually carries.
Tell us what you have in mind. We turn AI prototypes and fresh ideas into shipped, scalable products, from India, for the US and UK.
What everyone gets wrong: building the shop first
The instinct is to build your own shop first because it feels like the real business, and to treat the affiliate rail as a fallback. This is backwards for a new platform, and it is a lesson I have learned watching it go wrong. Launching your own shop means inventory, working capital, fulfilment and returns before you have a single validated best-seller. The affiliate rail lets you launch a genuinely universal registry with no inventory at all, learn which items actually sell in volume, and only then stand up a shop for that narrow, proven range. Start with the rail that has no working-capital risk, add the rail that has margin once you have earned the right to. This is the same lean-launch discipline we apply everywhere, and it is why adding items from any store is the natural starting surface, covered in add items from any store.
The other thing teams get wrong is the temptation to steer buyers to the higher-margin rail by hiding the cheaper or faster option. On a gift, buyer trust is the entire asset. A gift buyer who feels manipulated does not complain, they simply never use your registry again, and they were often a future parent themselves. Show both rails honestly, present your shop where it genuinely wins on price, speed or a bundle, and present the original retailer where its trust or price wins. Honesty on the checkout is not a constraint, it is how you earn the second visit.
The reconciliation spine, where it lives or dies
Everything above is straightforward next to the real challenge: keeping one list state true across two rails. When a crib can be bought in your shop, on Amazon, or on Target, and a well-meaning grandparent can also buy it in a physical store and tick "I bought this elsewhere", you have four ways for one item to be claimed and only one correct outcome. The system has to reconcile your own orders, affiliate purchase confirmations, and manual marks into a single source of truth per item, fast enough that a second buyer sees it is taken before they pay. Get this wrong in the cheerful direction and parents get duplicates; get it wrong in the other direction and a gift shows as bought that never arrives, and a thank-you note goes out for nothing. This reconciliation is unglamorous, it is the least demo-friendly part of the build, and it is exactly the part a cheap quote quietly drops. Do not let it.
What we build first
- The shared list state and reconciliation spine, so one item equals one purchase across every rail, before either checkout ships.
- The original-retailer rail, because it lets you launch universal with zero inventory risk.
- A deliberate own-shop range, stood up only for proven, high-volume items where the margin justifies the operations.
- Defer broad shop catalogues, marketplace sellers and complex bundling until the two-rail reconciliation is rock solid.
This is how our MVP and product development team sequences a hybrid commerce build, and when the own-shop rail grows into real storefront work it becomes an ecommerce development engagement, built from India for US, UK and EU founders with the commerce accounts in your name. The rails are commodities. The reconciliation spine is the moat. Build the spine first and the checkout model quietly becomes the most reliable thing on your registry instead of the most embarrassing.
Frequently asked questions
What is the baby registry checkout model?
It is the decision of where a gift buyer actually pays. Babylist runs two rails at once: buyers can purchase through the Babylist Shop, or buy the same item at the original retailer such as Amazon or Target and have it marked off the list. One is your own commerce with your own margin and fulfilment; the other is an affiliate handoff. A universal registry that lets you add from tens of thousands of retailers almost has to support both.
Why run both an own shop and original-retailer checkout?
Because they solve different problems. Your own shop gives you margin, data and control of the experience, but it means carrying or dropshipping inventory. The original-retailer rail gives buyers the trust, price and speed of Amazon or Target with none of your inventory risk, but you earn only an affiliate cut. Running both lets the buyer choose and lets you capture margin where you can without blocking a sale where you cannot.
Which rail has better margins?
Your own shop, almost always, because you keep the retail margin rather than an affiliate percentage that is often low single digits. But margin is not free: the shop rail carries inventory, fulfilment, returns and customer service costs the affiliate rail does not. The honest comparison is contribution margin after those costs, not headline margin, and for many items the affiliate rail is more profitable per unit of effort.
Do I need to hold inventory for the own-shop rail?
Not necessarily. You can carry stock on high-volume, high-margin items and dropship or use distributor fulfilment for the long tail. The point of the two-rail model is that you do not have to stock everything: anything you do not sell yourself still gets bought through the original retailer and marked off the list. Start by owning a narrow, deliberate range and let the affiliate rail cover the rest.
How does reconciliation work across two checkouts?
This is the hard part. When a gift can be bought on either rail, you must make sure an item bought on Amazon is marked off so nobody buys a duplicate, and that a shop purchase updates the same list state. That means reconciling your own orders, affiliate confirmations and manual "I bought this elsewhere" marks into one source of truth per registry item. Get it wrong and parents get two of everything, or a thank-you note for a gift that never came.
What are the trade-offs of a hybrid checkout?
Complexity for flexibility. Two rails means two order systems, two reconciliation paths, and two support experiences to keep consistent, which is real engineering. In return you never lose a sale because you did not stock an item, and you capture margin where it makes sense. For most registries the trade is worth it, but only if you build the reconciliation spine properly rather than bolting the second rail on later.
Should a new registry start with both rails?
Usually start with the original-retailer rail and add your own shop deliberately. The affiliate rail lets you launch a universal registry without inventory, fulfilment or working capital, which is exactly the lean MVP a new platform needs. Once you know which items sell in volume, stand up your own shop for that narrow range where the margin justifies the operational weight. Build the list state to support both from day one even if you switch the shop on later.
How do buyers know which rail to use?
You guide them. Present the own-shop option where it is genuinely better on price, speed or bundle, and present the original retailer where the buyer's trust or a lower price wins. What you must not do is hide the cheaper or faster option to force your own margin: on a gift, buyer trust is the whole asset, and a buyer who feels steered will not return. Show both honestly and let the choice build loyalty.
Can appico build a hybrid registry checkout?
Yes. We build both rails and, more importantly, the reconciliation spine that keeps one list state true across your own orders and affiliate purchases, from India for US, UK and EU founders at a fraction of onshore cost, with the code and commerce accounts in your name. We scope the reconciliation first, because the duplicate-gift problem is what a hybrid checkout lives or dies on.
“Disciplined, committed, over-delivers. Three years in, I would re-hire any day.”
“A factory of ideas.”
“A fantastic-looking and performing website.”
Talk to the team, we reply within 24 hours, and the first consultation is free.
Start a conversation →