home built shopping carts - worth of the effort?

[Habtom]

I was thinking of building a shopping cart of all the things I wanted on my website.

Is it worth of building it . . . or better to pick one from the market?

Habtom

[deltakid]

so far it seems it makes sense to use zen cart instead of oscommerce. i have to write some new features and use a very different design for the shop.
i really like self written stuff and enjoy my own cms way more than mambo that i used before im just not sure how much effort it is compared to the existing solutions.

[ckarg]

Off-the-shelf = up-and-running faster (usually), less flexibility in the long term
Custom = requires development know-how, potentially buggy, does everything you want long-term

Basic shopping carts aren't complicated. My last one took about a week to build from scratch (mod_perl). The most tricky aspect of a self-build shopping cart is getting the design and functional requirements right. Its easy to skip sound design, resulting in a poor visitor experience later on.

[imstillatwork]

I fyou really want to use something like oscommerce, try zencart, its based on osc, but slimmed down (code-wise, not feature-wise)

otherwise, I say, build yer own. It's FUN!

[anchordesk]

Small mom/pop operations are almost always better off using existing. As your business grows, you'll want your own coded shopping cart system tailored to your specific list of needs. If you are at the starting line and don't know what to do ... you might benefit most from an existing reputatable shopping cart system to handle your needs.

[bcc1234]

I rolled my own for two reasons: testing and tracking.
It's nice to be able to identify just the people who, let's say, added more than $50 to the cart, or people who added a particular product/combination of products, or people who qualified for free shipping.

[sharbel]

Well, I may be partial because I am a programmer but I really believe that a 'roll your own' cart can be a huge benefit to an ecommerce site. Don't get me wrong, I know that it's totally not feasible for some but for those who can do it themselves, or afford to hire someone to do it for them (assuming the company they get to do it creates it to allow for future expansion by whoever is taking over the project in the future), it can be a huge advantage.

I develop for and own a few e-com sites and the ability to work in custom features such as custom coupons, tying in with my fulfilment system's back-end, tying in with my custom administration applications, you simply can't beat it.

Of course re-inventing the wheel is silly, I have base-classes that I developed a few years ago that all my e-commerce projects inherit from (yay for OOP) but for custom business logic that is unique for every store, it's a huge advantage to develop code to specifically address these needs.

Now, one will say "Just use open-source and customize it", which is a valid point. I, however, have always been of the mindset that open-source projects are fantastic *learning tools*. I just won't use them for production, non-trivial applications. Not at all slamming anyone who does either, please do not misinterpret.

When all is said and done, the most important thing is to provide a easy to navigate, secure, user experience. How ever you accomplish this will widely vary from e-com site to e-com site...

[bono]

Just finished rolling out my own shopping cart :-) Its just so much easier to customise anything you want. plus once youve done it once you canuse your code again and again for other projects..

[afnid]

Shopping carts are easy, it would take me longer to build a toaster that someone else would want to use.

I never found a toaster I needed to customize, but I have seen a lot of cookie-cutter shopping carts that I could never live with.

I have never used a cookie-cutter e-commerce system/cart. I tried to setup a Yahoo store as an alternate channel to see what additional traffic might come my way.

I wound up replacing my entire yahoo store web-site with a single 50 line php script that proxies all requests from one of my other sites and bypasses their cart, just to get it up and running, over a year ago..

[spikey]

Nice to hear lots of others built their own for the same reasons I did.
One suggestion, take a look at the architecture of a bunch of open source carts and see what you like and don't like. I took ideas from a lot of different places.
The thing I like most about building my own is that I know every line of code like the back of my hand. If something needs a change or goes wrong I know exactly where to find it in seconds.
That being said, if your business grows to a certain level you're going to have to start interfacing with enterprise level business applications at some point. That's never pretty.

[john_k]

How soon do you want to be selling? If you take an off-the-shelf solution (buy or free), you can literally be up and running in a few hours. (provided merchant accounts, hosting, etc. has been set up) If you roll your own, well, it depends on how well you roll and how much you stuff in there. But even if you have no other business or life concerns, and you can code for 15 hours a day, it would probably be a couple of weeks.

I am a programmer and I also operate an ecommerce site that is not related to programming. In 2000 I headed a team of 5 programmers to develop a large, custom ecommerce site. The owner dropped around $200,000 on the development. (If they did the site today, it would be less expensive, but they would still need a custom solution)

When I started my own ecommerce site last year, I was on my own. No admin staff, no developers to help out, no buyers or bookkeepers, etc. So I spent about $50 for CandyPress (mod delete name if that's a no-no). I was up and running in a few days. It would have been foolish for me to waste my time doing a custom solution and thereby delay the start-up date.

The point is, if you have the resources to develop the custom solution while you attend to other business matters, then it will pay off in the end. But if you're a one-person or small shop, then it probably makes sense to start out with an off-the-shelf solution, even if it doesn't quite fit.

[Habtom]

Thanks John!

I would definitely consider yours and the other mentioned comments.

Habtom