Brainstorm: The Perfect eCommerce Software Solution: Wish List

Malcolm Cooper

Free Member
Sep 23, 2006
224
2
Cheltenham
Straight off the bat let's all please avoid the word "Free". You get what you pay for and if you get a great return on investment that's fine.

Given that statement what would you want from the ideal eCommerce software package? Let's really brainstorm this and sort out what the ideal is. We all know about OSCommerce, Zen Cart, xCart, Mr. Site etc. but why has none yet emerged as industry standard?

We all need a simple package that will allow us to concentrate on product, sales and customer service. eTailers need not be programmers.

Let's get real. What defines the perfect eCommerce system? How much are we prepared to pay for it? What does it need to do?
 

darren atkinson

Free Member
Sep 21, 2005
812
174
Why has non emereged as industy standard?:

I think one problem lies with the fact that different ecommerce sites have different requirements. I think it would be near impossible to have a standard 'off-the-shelf' solution which would fit everybodies needs.

This generally means that an existing solution will need customizing somewhat to provide the required functionality.

Along a similar vein, if an eTailer doesn't want their store to look like everybody else's then they are going to have to modify it themselves or seek out somebody to do this for them. Many companies who provide this service will have their personal favoured ecommerce solution, often depending on their skill base and experience. This is a big factor in deciding which systems an eTailer will use.

Take the following example:

A client calls me and asks for me to set them up an online shop. They give me the product details and logos etc...

When I deliver the finished product they don't know if it is a custom built solution for them, or OsCommerce, and they probably don't care either.

I think this is a major factor for no industry standards occuring.

Darren
 
Upvote 0
C

Coding Monkey

I twas reading in a web designer magazine this morning that.....

"45% of companies built their own eCommerce sites. Of the remainder, 85% used a web designer."

I find that pretty surprising, and I can't find out where they got the stats from.
 
Upvote 0
Having checked out 10million ecommerce packages (slight exageration) , I would have to chuck in my 10 cents worth.

For me, the following are essentials -:

1) Ability to take ALL payment processor options.
2) Numbers of products should be at least 5000, can't see the point in the sites that offer less than 100.
3) Customisation of site colours and to an extenet the layout of menus.
4) Ability to upload your own custom banner, yes some sites still do not have this freedom.
5) Automatic resizing of product images for thumbnails, some sites you still have to upload 2 seperate image sizes!!!

These are just a few of the issues I have had with various packages. The one I use at present does all these and alot more so I am happy at the moment :)
 
Upvote 0

ukanalyst

Free Member
Jul 15, 2006
106
0
Northumberland
Our company provides full end to end options to predominantly mail order/retail replenishment clients and we have obviously come across a wide variety of websites/eCommerce systems across many different platforms.

Where a client has an existing website, be they cosmetic or transactional and wants to maintain it or extend it, we have to negotiate the varying levels of code on the sites.

It's obviously easier if we build the site or bring in a third party as we can determine the coding and access, but in essence, to get round the multitude of code variations, we built a multi functional interface framework, so that we could integrate to the websites and thereby run the back end of the process through our Mail Order and Warehousing Systems. In addition, this framework is used to communicate with the client's legacy systems, like DC operations.

If a techie could take the time to build an all encompassing eCommerce package, at what point would they stop the development, and would they also weigh up the need for functionality against cosmetic appeal.

Could you feasibly build a generic package that would not be out of date tomorrow. The "in-thing" drives the requirement

In terms of cost, i would be looking for a walkaway package, which only needed updating with new products, but no reprogramming, and i don't think 5-10% of turnover is unrealistic for this vital piece of your machinery.

But, by developing a generic package, i would suggest you would dampen the innovation that the techies can experience by trying to create the bigger and better package.
 
Upvote 0
If this is indeed a brainstorm and any ideas can be thrown onto the flip chart, then I'd like to stir up another zone of the grey matter - for a second pull attention away from the debate on technical pure design, build and configuration issues ...and onto psychological events.

With three main entities in every online story - the project's author, the web technician(s), and its Market - I think the definition of Perfect is above all the ability for the e-commerce solution to enable psychological stickiness through the point of sale (and perhaps beyond).

It is easy to say that ultimately the site's author and the technical people (if distinct), are trying to integrate their work to get the system to sell stuff... However - whatever the context of the particular market - I want to throw in what this means in the less obvious psychological dimension. There is a growing science building of good practice to get sales conversions. Thus perhaps, the author and technician should be working towards perfecting the following:
  1. Drawing their already trusting visitors to product pages,
  2. Making the focal point of a product page simple - the product, the picture (if possible), and its price,
  3. Having a layer of more detailed info one click away along with product variants,
  4. Judiciously offering expert views and reviews,
  5. Offering access to detail on availability, deliveries, guarantees and warranties, but
  6. Avoid frustrating the buyer losing them in navigational complexity and the use of back buttons on the way to the shopping cart,
  7. Using a clearly visible call to action or 'Buy' button
  8. Making the payment transaction clear.
This is in most part, an extremely brief summary of an article I received in a newsletter, by James Maguire. It has much more detail information on a 'well-built product page[FONT=&quot]'[/FONT].

I hope it contributes to the debate.

P.
 
Upvote 0

Malcolm Cooper

Free Member
Sep 23, 2006
224
2
Cheltenham
Very interesting, thanks for the input.

When Ampalian Jewellery was started over 3 years ago it was impossible to truly predict the level of success (or otherwise) that we'd experience so it was based on free OScommerce software. It's been hacked around an awful lot and the code is a real mess. Nevertheless Ampalian is page 1 on Google for most jewellery related terms and business is very good indeed.

Over a year ago the decision was made to re-develop Ampalian from the ground up with bespoke written software, coded by a highly talented programmer, backed by 3 years live trading experience in this fiercely competitive market. Ampalian is certainly a top 5 online jeweller and we are now just a few weeks away from having cleanly coded, intuitive software that we intend to make commercially available to other eTailers (but not competitors, sorry :). We are not ecommerce theorists who are creating software that we think will work. We are eTailers creating software from live and ongoing ecommerce experience that we know works.

We will be continuously developing and improving as technology allows and there will be some key principles maintained as standard:
  • Code will validate to all standards
  • SEO built in from the ground up
  • Admin sales statistics as standard
  • Ability to operate with all payment gateways
  • Constant user exposure to product with buy now always available
  • Simple navigation with blistering speed for both browse and search
  • Fast product filters to drill down to what you're looking for at astonishing speed
I'll keep this thread up to date as we approach a launch date. I'm sure it'll be interesting.
 
Upvote 0
I don't think there will ever be the perfect solution that meets everybody's needs. In a competitive market, we can only strive for perfection, and continually improve what we have. I have yet to see the perfect online shop, in fact I have yet to see any website which I would describe as perfect or sublime.

I'd like to pick up on a couple of things with Amazon. Now Amazon are in total control of their e-commerce solution, which they built from scratch, and they have been around for quite a while now, so you would expect they would be getting closer to the ideal e-commerce website by now. Unfortunately, IMO, they still have obvious problems with their shopping experience:

1. Most of their books, DVDs, and CDs just have the one image of the front, e.g. front cover of book:
www.amazon.co.uk/o/ASIN/0553175211/
whereas if you are in a real shop customers like to see and read both the front and the back. They are beginning to put this functionality in place for a few products, e.g.:
www.amazon.co.uk/Good-Bad-Queen/dp/B000IAZ3E0/
but they don't seem to be doing it as standard across all products.

2. Music samples for the UK site still require the realone plug in, e.g.:
www.amazon.co.uk/Good-Bad-Queen/dp/B000IAZ3E0/
why couldn't they use Flash?

If online shops are to compete with retail shops, then they should at least try where possible to provide customers with good product information, and allow the customer to have a good look at the product, like they can in the retail shop. It isn't just amazon at fault here, there are plenty of electrical online shops that don't show enough product images, e.g. with a DVD recorder it would be nice to have a photo of the front, a photo of the front with any lid open showing front sockets available, and a photo of the back showing what sockets are available there too, and these photos should expand to be a big enough size for the customer to view what is important. I'm sure the same can be said for other products, e.g have multiple images to show off the full design of the clothes, or use other mediums to show off the product (video, etc).
 
Upvote 0

darren atkinson

Free Member
Sep 21, 2005
812
174
I agree that the ability to display various product photos over just one is quite a major one.

It is pretty well know in technical circles that Amazon break many usability rules on their site, however as they are the market leader they can get away with it. So much so that in some cases it is better for smaller sites to mirror Amazon functionility in some respects as customers could well be used to using the 'Amazon Way'.

Darren
 
Upvote 0
rather tham mention the word "free" I want to hit on the most important aspect of system like os commerce - that is they are open source.

You can make as many changes to the systems as needed legally against any patents or copyright infringements. This is a major thumbs up from me as to a base system that can be customised into any functionality needed.

However you must weigh up the time it will take to add your custom functionality against any product that has it from standard.
 
Upvote 0

Malcolm Cooper

Free Member
Sep 23, 2006
224
2
Cheltenham
I take all of this on board fully. I think I should replace "perfect" with "best". When we make Amcart available it will have multiple and flash image capability as standard along with all of the attributes already mentioned.

I take the point too about open source. Whilst I have total respect for the likes of OSc and ZenCart they do have problems. OSc put me in business in the first place but the code is heavily flawed. A Camel is a Horse designed by a committee.

Our objective is the cleanest and fastest shopping cart solution that it's possible to create using the the very latest technology with no "bloat". The priorities are to get found and to sell products. The look and feel of the sites it creates will of course be adaptable but with any shopping site the products need to take centre stage and be highly "buyable" with the minimum of clicks and hoops to jump through.

A live demonstration will be available early in April when I hope we can continue this thread and brainstorm the best shopping cart, bar none. Your input is appreciated and special deals will be available to UKBF members.
 
Upvote 0
I take the point too about open source. Whilst I have total respect for the likes of OSc and ZenCart they do have problems. OSc put me in business in the first place but the code is heavily flawed.
That is quite a sweeping statement, can you please qualify why you think the code is heavily flawed. Are you saying this from your jewellery industry background or from some other software development industry background that we don't know about? After all you did say "eTailers need not be programmers".

Or are you just repeating what your IT guy/company has told you. The reason I ask this is because 'put down statements' of existing systems is a common way for IT people/companies to convince clients that it is worth starting from scratch, reinventing the wheel, and spending a lot more money on a system. Another trick IT people/companies sometimes use to sell a custom build project to a client, is to say that the system will be so good the client can sell the system to other companies too. If the IT person is an employee, then there are other ulterior motives to consider too, e.g. some software developers simply prefer creating their own software with greenfield projects rather than improving other people's software (the 'not invented here' syndrome), and some software developers would prefer to be working for a software development company rather than a jewellery company. Question is do you want to be a software development company rather than a jewellery company?

As I have said before, there is no perfect sublime e-commerce system, or website, or large sofware development project. They all have their flaws. With large software development projects, the advantage of open source systems over closed source systems is that you can see where the flaws are, they are not hidden. For example, osCommerce - unless heavily modified (which is possibile with open source) - is not really suitable for shops with thousands of products, hundreds perhaps, thousands no. With open source you can also see what existing flaws are being addressed in the new version (e.g. osCommerce v3 alpha).

Our objective is the cleanest and fastest shopping cart solution that it's possible to create using the the very latest technology with no "bloat".
Just bear in mind that with generic solutions, what one shop owner considers as "bloat", the next shop owner might consider as a "crucial feature", and vice versa.

The advantages of open source systems like osCommerce, ZenCart (Zencart is a derivative of osCommerce), VirtueMart, etc is that they haven't been designed with one shop owner in mind, but with many different types of shops in mind, and they have a proven track record in this respect (we know they can be used for different types of shops because people have already created these kind of shops with these systems).

Luckily a system originally developed for selling jewellery should have some advanced features due to the advanced requirements of an online jewellery shop (e.g. product options that affect price, stock control at the product option level, good product viewing capability, etc), so such a system should have more than a lot of shop owners need, and it is better to have more than less. But will a system developed originally for selling jewellery have other features that are totally unnecessary for jewellery shops but crucial for other shops, e.g. digital product downloads?
 
Upvote 0

darren atkinson

Free Member
Sep 21, 2005
812
174
While I agree in parts with what you are saying Paul I do have a different opinion in some respects.

I took over the running of a clients site a while back which somebody else had produced using OsCommerce, the previous developer had made some changes to the default code base by utilizing mod's, and I can tell you the code was a real mess.

I have spent a lot of time sorting out the mess and it is vastly improved, however I am recommending that my client opt's for a different route.

My argument is that there is just too much work needed on their existing site to continue on with it for the future.

Like Malcom stated the OsCommerce got them up and running and was fine for their needs, however to really develop the business further I don't think it is up to their needs. They already have nearly 2000 distinct products with related product attributes, this makes certain areas of the admin area run really slow.

From another point of view, I want a store for them which is entirely CSS based to cut down on the bloated HTML of table design layouts, reworking the OsCommerce files to a CSS only fix is something I looked into, I redesigned just the header and menu area and cut out something like 8000 characters of extra html code, however this was not a quick job and it was mutually decided that we would not progress with this.

This site is doing really well in the search engines with around 5000 visitors per week and is making a profit, yet the conversion ratio is pretty poor. I have highlighted many areas of the site which need to be improved, even a small increase in conversions would mean a massive difference in profits.

To make all these changes in OsCommerce would be a major task yet again, and I'm not certain some of the stuff I want to do is really available without some heavy duty custom programming, which I don't want to get into as I have already stated, parts of the PHP code base are really messy.

One other thing which makes me laugh a little (This is genuinely not aimed at any one person) is when looking for eCommerce packages you come across companies offering their own solution which turns out to be a modified version of OsCommerce. Whilst some of these may offer extra custom built modules, many just have installed a few of the more popular freely available ones.

Again, I know OsCommerce has many benefits for a startup business there is no question of that, I am of the opinion that for online retailers looking for the next step a better solution is required.

Regards

Darren
 
Upvote 0
I agree with most of what you are saying Darren. osCommerce cannot currently handle thousands of products. osCommerce installations with various contributions installed can lead to really messy code if not done carefully, and certainly something that is not fun for someone else to maintain. The problem is that the base osCommerce system (and osCommerce templates) do not have enough as standard to make it a good system for an online shop and so you really do need some contributions added.

Contributions can interfere with one another too, and since most contributions have install processes which are just instructions on how to change certain code assuming you are installing on a default installation, there does come a limit to how many contributions you can install since you are not installing on a default installation anymore. All this should also be weighed up against future compatibility with the osCommerce base code too, since the more contributions you add and the more changes you make, the more difficult it will be to upgrade to osCommerce v3 in the future.

This is why I try to let people know what they are letting themselves in for when they decide to go the free DIY route of trying to set up an osCommerce system (or even an osCommerce template, which isn't much better) themselves, and taking on the financial responsibility of owning, maintaining an advanced e-commerce software system, and I often ask them the same question: are you in the retail business or in the software development business.

I wouldn't describe these problems as flaws since osCommerce is supposed to be merely a base on which a developer can develop a final system. Design and code flaws are usually problems that cannot be overcome either now or in the future. The typical osCommerce problems a DIY shop owner can encounter, e.g. lack of basic functionality (e.g. autothumbnailing, SEO, product option stock control, etc), having to become or pay a PHP/MySQL programmer to change the system, maintenance, support, improvements, future proofing, etc. can easily be overcome if a shop owner decides to go for a hosted managed and supported service offered by a company that has their own custom version of osCommerce and already offers an online shop with the features the shop owner requires.

A similar thing happens with the linux market too. People can download unsupported linux for free, or the wise ones will buy a value added custom build of linux that is supported by the linux supplier company suitable for the support and features they need, e.g. redhat, suse, mandrake.

The thing to be careful about is how much value add and support the supplier is giving. There are hosting companies that provide base osCommerce installed, which is pretty useless, and they do not support it. There are e-commerce companies that provide cheap e-commerce hosted managed solutions with pretty much the base osCommerce installation again but they support it. And there are companies that offer a lot more value add and offer a better custom version of osCommerce as standard. I'd like to think that we fall in the latter category, and with our sitebuilder solution, our custom version of osCommerce is one small component of our solution, it sits as an online shop section of a larger website created with our online sitebuilder.
 
Upvote 0
...The reason I ask this is because 'put down statements' of existing systems is a common way for IT people/companies to convince clients that it is worth starting from scratch, reinventing the wheel, and spending a lot more money on a system...
Very nicely put Paul; you only realise you have bought the 're-packaged' wheel, when you have lost so much time taking off the wrapper (or find you can't).

Darren, if two of the core features of this new design is that it's optimised for the latest CSS standards and control of the declarations therein, and also focussed on enhanced SE referencing activity, I should be pleased to read more about that part of the work.

Regards,

Paul.
 
Upvote 0

Malcolm Cooper

Free Member
Sep 23, 2006
224
2
Cheltenham
I wasn't "putting down" osCommerce, just seeing it for what it is. Even as a base platform the code isn't clean. I don't rate myself as a programmer but I've spent many hours in the code adding contributions and making mods and I know enough to know that the foundations are shaky before you start down that road. Having said that if it didn't exist someone would have to invent it. I deeply appreciate its existence but there comes a time to move on and get serious.

The proof of the pudding will be in the eating. I hope there'll be one or two beta testers here when Amcart is ready to roll.
 
Upvote 0

Malcolm Cooper

Free Member
Sep 23, 2006
224
2
Cheltenham
It will be used to replace ampalian.com nect week. Look and feel will remain similar although it will be 100% width. The code however will be totally different. One of the benefits will be site searches down from 3/4 seconds to less than half a second.

I'll should when it goes live, I expect next Wednesday.
 
Upvote 0
I wasn't "putting down" osCommerce, just seeing it for what it is. Even as a base platform the code isn't clean. I don't rate myself as a programmer but I've spent many hours in the code adding contributions and making mods and I know enough to know that the foundations are shaky before you start down that road. Having said that if it didn't exist someone would have to invent it. I deeply appreciate its existence but there comes a time to move on and get serious.

I agree that osCommerce has its problems, if it didn't we wouldn't have needed to implement our own custom version of it for our sitebuilder service for our clients, and we wouldn't feel the need to advise business people to avoid going the DIY osCommerce base install or osCommerce template install route.

However it is a very complex e-commerce system - multi-lingual, multi-currency etc - and the more complex a system gets, the more difficult (but not impossible) it is to keep the code clean. For a system developed via the open source community with a number of previous versions behind it, it is not bad, for a large system it is split up to a fair extent code-wise and database table-wise, (as always, it could be better, which is why v3 alpha is currently being worked on, and so on).

osCommerce has been doing something right though, since it has met so many shop owner's diverse needs and it has so many (possibly the largest number of any open source e-commerce system) contributions from third party developers, and even brave non-programmers have taken it on and built shops (although I don't recommend this).

I think this success is partly due to its pragmatic, get the job done, approach. The original developers obviously weren't designing the system to win any awards for object-oriented architectural design from the academics and purists out there, if they were they would probably have chosen a different language to PHP 4 in the first place. Their prime concern was results for the shop owner and open source freedom, and perhaps in some ways the people behind osCommerce are pragmatists like you, since you say: "We are not ecommerce theorists who are creating software that we think will work. We are eTailers creating software from live and ongoing ecommerce experience that we know works".

If the osCommerce team had cleaned the code up too much, adding extra layers of abstraction and loose coupling to the design, perhaps its original uptake wouldn't have been so good since only talented developers who understand the more advanced design concepts behind large systems would have worked out where in the code (or otherwise) a change was required, who knows? ZenCart is a derivative of osCommerce which has focused on cleaning the code up more, but it is not so widespread, not so well supported within the community, and does not have so many contributions as osCommerce, ZenCart has not taken over from osCommerce, it still follows in osCommerce's shadow.

The fact remains that the osCommerce source code is open, you can see it for what it is, warts and all, and if you don't like it you are free to change it. I don't know whether the code behind Actinic is clean, dirty, buggy or flawed since it is a closed source system and I cannot see the code. On the other hand, I also cannot see the code behind the Microsoft's Windows operating system to comment on whether it is clean or dirty, but I do know its design is heavily flawed, otherwise there wouldn't be viruses, trojans, spyware, firewalls, etc. and I do get the impression that some of Microsoft's software is created by 'theorists' rather than pragmatists.

The proof of the pudding will be in the eating. I hope there'll be one or two beta testers here when Amcart is ready to roll.
I wish you the best of luck in your project. It can be difficult enough designing a system which meets the needs for one client, let alone 2 or 3 or many clients, and designing a truly generic system can be an even more challenging (and sometimes thankless) task.

I also wish you luck with your ampalian website, and I hope the coming migration from your osCommerce system to your brand new AmCart system doesn't throw up too many problems, doesn't upset the old customers who like and prefer the old system!, and doesn't affect your google ranking too much.
 
Upvote 0
At a very quick first glance, 3Dcart is a US managed hosted service, which means that your websites will be US hosted (not so good for UK SEO). Since it is US hosted you will probably find a lot of american terms which you might not be able to change (e.g. add to cart, zip state in address?). It also looks like the system has a problem with autothumbnailing, looking at demo site 1 some of the thumbnail images look very poor quality due to low quality resizing (obvious pixel loss, stepped slanted edges).
 
Upvote 0

JamieM

Free Member
Mar 22, 2006
2,318
351
Just a couple of points with regards to features I have struggled to find.

Real Time Accounting Integration - Kashflow API looks promising but no carts using it yet that I am aware of.

Wholesale Logins - I think this is possible with Actinic but a little restrictive and I'm not that keen on Actinic in general.
 
Upvote 0
S

Short Time Stationery

Web designer magazine, are giving away EROL4 with the latest issue, is it any good?

Having visited their site and looked at some working demos from sites that are using the software, it seems okay. But then again, I'm not a web designer.
 
Upvote 0

ken_uk

Free Member
Jul 27, 2007
2,213
240
56
Not got the time to do any integration stuff at present, but I did take a quick look at the cash flow api pages, they seem very sparse on info which could be putting people off developing anything.

The soap pages show what needs to be passed, but are pretty generic, ie it just lists the field name and its type string, integer etc.

It would be helpful to include such things as the exact format for the strings, max and min lengths etc. Same for numbers, the min and maximum values, also what validation is performed at your end, and what needs to be done client side.

Info on how it handles failed updates/transactions etc would be good too.

A database schema would be handy, so people can see what fields will need populating at a glance, and can then check the api docs for detailed info, along with an overview of dependencies (ie, if you update this you must also update this etc)
 
Upvote 0

Latest Articles