2007-11-11

E-Commerce Solutions Roundup

I've been playing around with basic e-commerce packages lately, trying to set up an online store. My requirements are basically a reasonably core set of features, like catalog management, a shopping cart, and order management. I also require it to be a CSS-driven layout, to make design changes easier. I would also like bulk import/export of products as a "nice-to-have". I am not affiliated with any of the solutions being described here, although I have worked with some of them.

The landscape of different solutions out there for setting up an online store is daunting, to put it mildly. There are services where you sign up for a complete package, including hosting (Like Shopify, MonsterCommerce, and Yahoo Small Business), free software to be downloaded and installed on your own server, (like Zen Cart and osCommerce), and various commercial software (like Miva Merchant).

I'm leaving off heavier stuff like Microsoft Commerce Server, which is more of a set of libraries for developers to build their own e-commerce site. It's definitely not an "out of the box" package, and is comparable to almost building a site from scratch (at least version 2000 was, maybe 2007 is more suited for my needs). For semi-completeness, there's also The Apache Foundation's ofbiz ("Open For Business"), which is a complete enterprise business solution, which never met an acronym it didn't like:
The Apache Open For Business Project is an open source enterprise automation software project licensed under the Apache License Version 2.0. By open source enterprise automation we mean: Open Source ERP, Open Source CRM, Open Source E-Business / E-Commerce, Open Source SCM, Open Source MRP, Open Source CMMS/EAM, and so on.

I know some of those TLA's, sort of, but yeesh... English, please!

I'm also leaving out ecommerce modules that sit on top of Content Management Systems like Joomla. Mainly because there's a double learning curve - Joomla, which brings with it a hefty amount of overhead, on top of the module itself. If I already had a serious Joomla site going, I would consider adding a module like that to enable ecommerce functionality for the existing site, but doing it from scratch with a CMS just seems like overkill. The complexity and time factor basically disqualifies this approach from the get-go.

I'm sure I'm missing about 99% of the solutions out there, but these are the ones I've actually heard of and have checked out (if briefly, to be honest).

What's interesting about many of these offerings is the "tiered" approach to selling the products. For example, MonsterCommerce lets you set up a website for $50/month, but if you want to change what your website looks like in any significant way, you need to upgrade to $100/month. Yikes. Miva seems to be one version for everyone ($1000 one time payment for a "domain license", but you can find lots of hosts who offer it for $15/month or so). ZenCart is free but support isn't. There are no "getting started" links. You need to buy the book for that...

Staying with ZenCart for a moment - I liked their feature list, and their marketing copy:
Some "solutions" seem to be complicated programming exercises instead of responding to users' needs, Zen Cart™ puts the merchants and shoppers requirements first.

Sounds good to me. A little too good. And in fact, the out-of-the-box installation I did looks like this:



Ack! I'm not sure what that's supposed to be, but it doesn't look like a respectable store. I could tell immediately that I would have to spend my time just cleaning out the banner ads, polls, latest news, and all the other stuff I just don't need (telling customers their IP address in the footer is not something I want as part of my store). I didn't find the admin tool to be intuitive enough to figure out how to do all these things quickly and painlessly, and so far I'm not inclined to buy the book to learn this.

My next stop is Shopify. So far this is my favorite overall solution, at least for my purposes. It's the easiest interface I've seen by far, although they seem fixated on the T-shirt business model. I'm not being facetious - T-shirts have specific properties that don't apply to my site, like sizes. Every t-shirt is typically available in a variety of sizes (S, M, L, etc). Since it doesn't make sense to have a separate product for each size, the standard practice is to use "variants". But, again, since variants don't really apply to my site, it mostly just gets in the way for what I'm doing (Shopify requires a variant for every product, and it takes some hacking to get around that).

One of the cooler aspects of Shopify is the "collections" feature - it works largely like a playlist. You can have regular collections (just add products to the collection one by one) and "smart" collections (give the collection a bunch of rules, like between $500 and $1000). Then you can use the collections throughout the site.

It's also the easiest that I've seen to customize as far as layout goes. It comes with a bunch of pre-designed layouts that "just work", and look good. The template system is based on Ruby, which is a great choice. It's definitely the fastest and easiest way to get a small site up and running, like a boutique shop with a small inventory. It gets a little pricey as they charge both for transaction fees (up to 3%! And not even including 3rd party checkout transaction fees for credit card processing) and hosting fees, however. And it's hosted, which could be a dealbreaker for some (I personally prefer using my own hosting solutions).

This article is getting a bit long-winded, so I'll just wrap up with Miva and MonsterCommerce. I have extensive experience with Miva (version 4 - they're now up to 5), most of it negative. Miva has a large feature set, but I've had baaad experiences with them. Miva 4's out-of-the-box store was the opposite of ZenCart's - instead of a huge mess, it was practically empty (and the colors and fonts that were there were downright ugly). The only default navigation is a clumsy category tree on the left. The layout was table based, and very hard to customize. The only way to really work with Miva 4 was to buy 3rd party modules, each with their own quirks and problems. Also, Miva uses DBF files for data storage, as opposed to to a SQL solution, like MySQL. It makes it easier to install, but it also means that it's really really hard to work with the underlying data directly (if you're writing custom software, which I have). Which means the path of least resistance is to use MivaScript, Miva's hideous custom scripting language. I'm not getting started on that. But Miva is actually very easy to get started with, since many hosts offer Miva as an option with plans as low as $15/month. So it does have some appeal. But it's definitely off the menu for me.

To be fair, Miva 5 might address many of these issues. I spent about 5 minutes with the Miva 5 admin interface (also ugly, clunky, and confusing), and I saw very little that seemed different than 4. And while I'm forgiving, I would really have needed Miva to be rewritten from the ground up for me to even consider it at this point.

Finally, MonsterCommerce. I haven't really played with it yet, mainly because it's a bit hard to. There's no real demo (just a single annotated screenshot that you inexplicably get when you click on "demo"), and no trial to sign up and play with it. My main attraction to Monster is that the live sites they link to who are using MonsterCommerce look really good, and pretty close to what I want. But since the lowest price of admission is $100 ($50 "setup fee" + $50/month), which is a bit of a commitment, I'm not biting without a lot more information (preferably hands-on).

So there you have it. There may be the perfect e-commerce package (for me) out there, just waiting to be discovered, but I'm a bit skeptical at this point. There's always going to be considerable tradeoffs.

Hopefully this has been helpful for you, if, like me, you're evaluating e-commerce solutions.

UPDATE: T was nice enough to offer some suggestions in the comments: Magento, which is free open source software you install yourself, and Volusion, a hosted solution. A rep from Volusion also left a comment. Magento is still in beta, and under heavy development, but looks very promising. Volusion looks great but is pricey. See the comments for more info.

--YY

p.s. I didn't get to osCommerce and Yahoo. Yahoo is really expensive ($50 setup fee + $40/month), but seems featureful. No demo sites to log into and play with, though. Plentiful screenshots, but screenshots are not enough if I'm plunking down hundreds of dollars... osCommerce has serious marketing problems. Their screenshots are tiny even when you click on them. Even so, I've played with it a while ago - maybe it's time I gave it another shot...