For years, I have thought about building a shopping cart with one goal: maximum conversion rates.

Your article should have been titled:
If you have loads of cash to burn, years of conversion rate experience, lots of time to waste, then a hosted cart may not be for you.

It’s funny, but can you tell me how many successful online stores use hosted carts vs. self built systems? Well, I can give you some insight. I know of several successful companies, and most of them have hosted, or prepackaged systems to some degree, even if they are enterprise level. I even know a couple of stores that process 1 – 3 billion dollars in sales, and yes, their systems use many pre-packaged ecommerce pieces. I can’t tell you how many homemade ecommerce systems I have seen flop, because while they may be good in theory, their execution has so many opportunities to fail. In fact, so much of what goes on behind the scenes, and few people realize the power of automation, and large volume order and product processing, reporting, marketing add-ons, etc… One single marketing feature in a particular cart I use accounts for 10% of my sales, and it’s free. I would have never wanted to, or thought of, building such a tool.

Just because you know how to build a cart, doesn’t mean you know a thing about successful conversion rates. In fact a single split A/B test can easily cost 6 figures (e.g. $5 per paid click x 1.5% conversion rate x 500 conversions).

I build lead generation sites (i know they’re not ecommerce) that convert at about 80% using pay-per-click, and the same principles apply across all of them. Conversion rate optimization is not as varied in my opinion.

Sometimes, using an ecommerce framework is more ideal as they provide a basic starting point.

I will also say that there are very few hosted ecommerce providers that I think are really good.

Anyways, I could write for days on this, but I think your article is way to biased.