[arcturus17]

> If you really go through the links you can see the quality of the code and ease of use

Disagree on this completely; I've studied this space before and using something like Spree and Saleor seems like a world of trouble unless you are willing to commit significant, permanent, and highly specialized development resources to it.

Also, the ecosystems for many of the solutions you describe are quite fragmented and niche; I looked into Spree / Solidus / Saleor devs in my country and couldn't find one, whereas I could find thousands for other more popular (PHP-based) solutions. In my case this would mean that if you wanted to hire local Python / Rails devs, you would have to train every single one of them in the use of the framework.

Finally, I don't think these solutions are as fully featured as some of their competitors. In my comparison I saw that Spree had a plugin for bundled products but the code repo looked desolate, full of issues and left to rot for years, whereas many others had much better support baked in.

Due to all of the above, many medium-sized companies with typical use cases are going to be better served by a more out-of-the-box solution, rather than taking charge of a Rails or Python codebase that's going to grow wild with custom code.

As for Shopify, it's definitely not for mom and pops anymore - they are serving massive customers throught their Plus offering, and their move to an open and comprehensive API (with both REST and GraphQL) allows anyone to build highly customized headless apps around it.

[dragonsh]

There are many more option not just spree or saleor. Having not just studied but by building add-ons for shopify working with Shopify API including the new graphQL now I am sure it’s designed for mom-pop store trying to fit itself by custom coding into medium and large size. A medium or large size customer is better served with a custom solution built with open source base which provides all the necessary API and data models to start quickly including UI design see the storefrontUI and vue-storefront+magento.

Shopify and bigcommerce are good for small or mom-pop kind of stores which wants to spend $30-$1000/month based on sales, here shopify can take some trouble out of woocommerce and opencart kind of platform. If it’s more than that open source is still better with custom additions.

[arcturus17]

Shopify is serving multiple billion-dollar brands, and that’s a fact [1]

My point is that even in the large enterprise space there are businesses that have fairly straightforward eCommerce needs that can be better served by ready-made solutions.

If your requirements don’t fit that, then sure, go with a framework and build around it. My secondary point was that I believe in this scenario it’d probably be less risky to go with one of the long-time PHP incumbents like Magento, which you now mention, than something like Spree or Saleor.

[1] https://www.shopify.com/plus/customers

[dragonsh]

>Shopify is serving multiple billion-dollar brands

Your are right except an exception Shopify might be used in some very small parts of big brands with few SKU's. For moderately large SKU most use flexible systems and shopify or BigCommerce is not one of them. Indeed like many companies even if one small division in one country of an MNC or brand is using BigCommerce or Shopify they will add their logo on their site to show they have this big brand as their customer.

Having worked for many multi-national brand can say they use Magento, Hybris, Demandware (now salesforce commerce) more than Shopify or BigCommerce.

Indeed even all those brands can be served better by open source multi-channel commerce system whose source code they control, otherwise they might have same issue when Oracle and IBM stop investing in ATG and WebSphere commerce platform. Indeed many of our old customer moved to in-house system as ATG, WebSphere commerce and even hybris cannot keep pace with the fast paced changes required in multi-channel commerce. Shopify and BigCommerce will need to primarily serve their SaaS customers and being limited by single code-base for all customers, cannot serve the needs of individual brands or multi-nationals that effectively. Even if they come with a program to provide on-premise or deployment option on a cloud of customers choice, given domain expertise they still will not be able to provide that flexibility offered by customized solution.

So for those brands its a better choice if they retain control of their core multi-channel commerce platform than leave to the whims of Shopify or BigCommerce which are anyways not that flexible and all the additional functionality needs to be developed in-house with partner using plugins or extensions.

[jasode]

>Having not just studied but by building add-ons for shopify working with Shopify API including the new graphQL now I am sure it’s designed for mom-pop store

I don't know much about Shopify but if the following article is accurate, it's not just "mom & pop" stores:

https://wemakewebsites.com/blog/37-of-the-biggest-brands-on-...

Examples of big brand names with more than a few SKUs include Hasbro toys, Penguin books, etc.

[gavinballard]

Yes! I don’t really get the jump from using the GraphQL API to “this can only be used by small players” when it’s empirically not.

That article is absolutely bang on, We Make Websites is one of the best in the business when it comes to Shopify.

Disclaimer: my company has worked with WMW on a number of projects, including Hasbro (which last time I checked weren’t particularly “mom and pop”).