As a new user, the storefront is really frustrating to use. An out-of-the-box solution like Shopify is probably worth the investment

[jtbg]

I know that legacy code has a way of freezing a product in time, but it's likely true that the lower rate of cart abandonment *alone* is enough to offset the cost of a few contractors for a few months to build out a storefront on Shopify or another platform. Hell, I'd volunteer my time to contribute :)

[Brian Bauman]

Thanks for the note, jtbg!

Updates and long-needed improvements (including a redesign) to the store are certainly on our roadmap, but unfortunately it's not quite as simple as "a few contractors for a few months".

Paizo's business model was built originally out of magazine subscriptions rather than web retail, and so many facets of how we do business and operate as a company tie back to that origin. When we added a more traditional web store, many of the design choices were made to integrate well with Paizo's primarily subscription-based business. Things like the sidecart are a good example of that - where you can mark products to be charged and shipped with your next subscription order to save you money on shipping. And even today, our primary business isn't flipping third-party gaming accessories, but rather publishing Pathfinder and Starfinder products to regular customers, many of whom hold one or more subscriptions.

In many cases, our business rules aren't trivially reproduceable in a lot of out-of-the-box solutions. In some cases, aspects of our business model are very customer-friendly and/or are important to Paizo's identity, so these would need to be preserved. So while an out-of-the-box solution does provide some very useful baseline features "for free", there's a lot of "special sauce" that we would have to re-implement in a brand-new environment.

There's also an issue of data. We have orders dating back almost 20 years which would have to be integrated with a new solution. Sales and subscription data are used all over the website, so we would have to write a compatibility layer so that both the old and new platforms would be able to see and update all the same data. If we were using fully-hosted solutions like Shopify, we would also have to share sensitive customer data with a third party. Many of of our customers might not be ok with that.

And, of course, all of this additional implementation would have to be strenuously tested, especially since it would handle PII, PCI, and financial data. An eventual cutover would have to be fairly swift - we would need to have the new platform almost complete before we could begin using it, so our visible productivity would drop for the duration of the initial development. During that time, we'd have a greatly reduced capacity to make other improvements to the website or internal Paizo infrastructure. As you said, legacy code has a way of freezing a product in time - and that's precisely because working with legacy software requires considerations far exceeding what is required to set up a new store or website from scratch.

But I'm happy to report that we are working our way through the site making serious improvements, both visible and invisible. We'll get to the store (especially the checkout process), but there are other components we have to get in place first.

[Steve Geddes]

Thanks, Brian. I've heard some of that before, but the behind the scenes perspective is always useful.