We are lucky to already have a web developer on board for this project (me ), but if we didn’t have this experience in house, then we would need to look elsewhere. Choosing a web developer will depend quite heavily on your budget, expectations and the scale of your operation.

Once you know what your budget is, what you expect/what from the web developer and you know just how big the site will be, and then you will need to go out and find a web developer. This could be anyone from a freelancer, to a small web company, to a much larger web company, and this decision mostly will depend on your budget. Most importantly you need to be sure that the developer you choose is capable of handling the work and will be able to help you grow.

Here are our guidelines for how we feel a company or individual should go about looking for someone to develop their new eCommerce site.

1. Work out a realistic budget for what you would like to spend on the development work
2. Work out what your expectations/requirements for the web developer are. (This could be anything from custom stock integration to a specific payment provider).
3. Determine how big your site will be (in terms of products and visitors), as this will determine the right software and systems to use for your site.

After this, you should have a good idea in your mind of what sort of web developer you require. When choosing between a freelancer, a small company or a large company, it is necessary to keep in mind your requirements now, but also your future requirements:

A freelancer would be an ideal low cost solution initially, however as your company grows and the workload increases, it may lead to them having to pass the work onto a small company in order to better facilitate your requirements. By this I mean that they might not have the current capacity to handle your increasing workload and they may not want to grow larger than being just freelance.

A large company would be a very expensive solution initially, but they would be able to handle the growth and would possibly have more experience in dealing with large projects and be able to help you grow your business quickly in a well thought out manner.

A small company would not be the cheapest solution, but they would also bring to the table the capacity to grow alongside your new ecommerce business.

Once you have these sorted out, you need to then choose a web developer, it is best to approach a number of different companies/people and get quotes from all three.

• Ask to see examples of their previous work.
• Ask if you can speak to some of their current clients to see what they think.
• Ask them what makes them stand out above all the rest and what sort of technology will they be using to run your site.
• Ask them how they would intend to host your site, as you want to make sure they are using a secure solution that is backed up every day with the backups being kept for at least a week.
• Make sure you tell them as much detail about your project as possible and ensure they have an exact specification to work to when calculating a quote or writing a proposal for you.
• When you get a proposal back from the company, read it carefully and watch out for hidden terms and conditions and ensure that everything you want is in the proposal. As anything not specifically written in the proposal document is 99% of the time, not included in the price.

Golden Rule: It is important to remember that the cheapest option is not always the most sensible option (“pay peanuts, get monkeys“). However, it is still important to remember that big companies will charge more because they have more overheads involved.

It is also prudent to make sure you are not paying for a ‘big brand’ company to work on your site, when they might not necessarily do anything different to a smaller brand company.

I’m not saying here that you should throw away your money, but you must make sure that the decision you make is not solely based on cost, but is a compromise of quality and keeping costs within your budget.

If you are building your own eCommerce site, then very early on, you will make a decision on the technology that will make the basis of your site.

Your first decision is to determine if you wish to develop your own system from scratch, or do you want to use readily available open source software and adapt it to meet your requirements. This usually comes down more to your situation, budget and time frame. If you have a short time frame and a low budget, then you really don’t have time to be building your own software, so you need to use open source.

In our case, we will be developing our own system from scratch (or rather rebuilding our old system from scratch), this is because we have time to work on the development of the system and it gives us an opportunity to blog about the system as we go, explaining how and why we did things as we go.

Open source software is a quick and cheap alternative to get your eCommerce system up and running, and some of the software currently available (magento, etc) is phenomenal. One of the major downsides that we find with open source eCommerce systems, is that generally they don’t quite fit with how you want to do things, and if you don’t know them inside out, it makes it much harder to change things to be more suited to your needs.

We also chose not to use open source technology as we would need to invest a significant amount of time familiarising ourselves with the system and how it works and how the coding works, so that if we decide to integrate it with a stock control system, we would be able to do so. This all takes up valuable time, which by the time we’re confident enough to be writing useful additions to the open source software, we could have just written a system ourselves that is tailored to our exact needs, but if we do need to modify it, it’s easy because we built it, so we can change it 10x quicker than anyone else could.

Once you have made a choice between open source and developing your own system, the next choice will only really relate to those that chose to build their own system, as choosing an open source software package will have already dictated to you what language will be used.

If you are building your own system however, you will need to decide what DB to use and what programming language you want to use in order to the develop the system. One of the most common combinations is a MySQL database with PHP, however Ruby on Rails is proving to be a big contender for PHP and is certainly worth a look in.

Because we have more experience in working with PHP, I chose to continue down the route of using PHP in conjunction with a MySQL database as it will enable me to quickly work through the development without having to learn new things by using ROR. At the end of the day, I think that whatever technology you choose to use, will not make much of a difference to the end result, it is just about what you are more familiar with and how confident you are that you can develop such a complex system using that technology.

It is a good idea to plan out the development of the software, so that you can keep on top of things and keep track of where you should be with the development, so that you can deliver the software in time to launch your new online store. With this in mind, I would suggest creating an implementation schedule, where you list all of the tasks and items that need to be built/developed in order to make the system, and then prioritise these, setting targets based on how long you think it will take to make. An important thing to remember is that you should prioritise features that are considered to be CORE functionality features, and then once these features have been completed, then you can start to look at features that are not a necessity to launch, but may well make your admin life easier.

Today we received a new comment on the blog with regards to our “Setting a Budget” post. I felt it was so well written that it was worth responding to in full. The comment was from Oliver Ker, a freelance designer from Huddersfield, UK.

Thanks to Oliver for taking the time to write such an interesting response. This has actually made us realise that we hadn’t addressed this issue in full and has prompted another post.

You can find a full response to your comment here.

I don’t have a great deal of experience of budgets when setting up, is this budget what you would expect your customers to pay to you as a start up looking for your services?

As a designer I am worried about the design/development being “swept under the carpet”. Are you noting down the hours spent developing this project to calculate the cost later?

Also I would be worried about £100 as a marketing budget? What does this cover? Marketing can make or break your company. What do you class as marketing? Writing these blog posts and your twittering about the project is classed as marketing and Im sure you have spent more than £100 of your time.

Personally to start a business on a shoe string, I would have used a platform that already exists and anyone could set up (I have done this, and (http://mlbadges.com/schools) is now quite successful, and would not like to add up the hours spent developing, designing and marketing it). BUT does this not contradict with why you started this experiment, as your client left to seek a bigger platform to handle a bigger turnover.

As you can tell I am following this project and interested to see where it goes. Good Luck

I think it is worth clarifying that the budget we’ve set out in our original post for Eliot Turner only includes our start-up costs. It does not include our staff time for the design, development or creation of the marketing strategy.

You can see from the post “How Strong is Your Team” every member of carrotmedia Ltd is becoming a director in the business and taking share options. With this structure in place we have all committed to carry out the necessary work setting the site up outside office hours. We are in a unique position to have all the skills and software at our finger tips to launch such a site with little cost to ourselves. Before you ask, yes there was a cost originally for developing our eCommerce system. However, these developments are required for our existing eCommerce clients and if we add any new features these will be released back to our clients as upgrades to the existing software. We therefore felt it would be inappropriate to build this cost into the startup budget. If a client required our services to setup an online shop we would be charging for a full design and build so the costs we’ve set out in our initial budget will seem very low.

An eCommerce website is not a part time business which can be managed outside office hours once launched. We fully understand once launched it will require work during office hours. At this stage we will keep a track of our time and raise a management charge from carrotmedia Ltd to Eliot Turner Ltd for time spent on the site. The management charge will be based on a cost price for our members of staff and not the retail rate being billed to our clients.

So how are we calculating the management charge?

• Salary + Tax + Employers National Insurance
• Divide this by 235 days to allow for holidays & weekends
• We work to 5.5 billable hours in a day so we divide the amount by this number.
• We then add 10% to the figure to account for overheads to give us our hourly rate.
• For example, on a salary of £20,000 we would be looking at around £20 per hour.

I agree with your assessment of the marketing budget of only £100 being small. This £100 will cover the first month of trading and we will be modifying the budget throughout the year which will be based on the margin achieved on turnover.

There is an important business issue which is partly why the marketing budget is so low. We are looking to test the market in the first month of trading and don’t want to commit to a large amount of stock which may or may not sell. If we commit a large marketing budget we may sell more stock than we have in hand. We never want to be in a position where customer service is anything less that 100% and overselling could kill us before we’ve even begun. I’ve made no secret that buying jewellery is not my primary occupation so being cautious and accepting slow organic growth in the first few months is advisable. Once we’ve gained the confidence and developed a strong marketing strategy we can build up our stock levels. It takes some time to take photographs and add items into the eCommerce system so until we have someone dedicated to the task we have to limit marketing spend to control sales.

The comment about our client leaving to be supported by a larger platform isn’t correct. The client left due to them feeling we lacked the marketing experience to take their turnover into the millions and they were approached by a dedicated eCommerce design agency. Unfortunately we didn’t get the opportunity to demonstrate that we did have this experience. In reality this is why eCommerce experiment exists as we’re looking to show we can not only demonstrate skills in web design but also business acumen.

Content management within an eCommerce site is often quite difficult to get right. The majority of eCommerce systems either skip over it entirely, or bolt on a poor / cut down CMS as more of an after thought than really planning how to make it work right. Another reason may well be that the developer simply doesn’t have the time to deal with functionality that isn’t always classed as a ‘CORE’ requirement.

A common way of dealing with this is to install a purpose built content management system along side an eCommerce system, but this makes things ultimately more complicated as the systems may differ significantly and no one (client or developer) wants to have to log into two different systems every time they want to change things.

Currently, we have our own content management system and we have our own eCommerce system. When it comes to eCommerce sites we have tried running a CMS alongside an eCommerce system, which didn’t work at all, we’ve tried putting a small ‘bolt-on’ CMS into the eCommerce system, which also didn’t work very well and required a developer to add new pages etc, the client only had a simple content box.

When we redevelop the new system, I intend to integrate a simple, yet effective and reasonably powerful CMS within the system. It will come with features along the lines of the following:

• Ability to add/remove pages and folders
• Ability to add content items to specified pages
• Ability to upload multiple images/files etc and link to them from the content.
• Will have set xhtml templates based on the design of the ecommerce site, that will be set by the developer and then the client can use them.
• Set up sections within the templates that can be edited by the user.
• Will use the same CSS as the site, but can be edited by the admin.

By working to integrate this from the start, it should make things a lot simpler.

We have always used our own software for our ecommerce sites from the beginning. Since then the system has been added to and modified as and when required to add or improve functions on a very ad hoc basis. We had been planning various upgrades to the system in order to improve it’s functionality and capabilities, but I quickly came to the conclusion that perhaps there was a better way to go.

After spending some time looking around for open source alternatives to our system (which there are some extremely smart packages out there i.e. magento, etc), I felt that these solutions would become more of a problem than a solution. We would need to invest a significant amount of time into familiarizing ourselves with the code and how the system works and the processes involved. Without this it would be very difficult for us to make changes to the underlying system if we needed to add to the functionality to meet a clients needs.

Deciding that our current system wasn’t up to the job and the open source alternatives weren’t going to be suitable for what we needed, I took the decision to completely redevelop our software from scratch. This would enable us to eradicate any problems from the old system, implement new functionality and most importantly, improve the usability of the admin area to make it as simple to use as possible.
I will be building the new system alongside and around the ecommerce experiment site, this will enable us all to personally use the system day to day as part of the experiment, so we will quickly be able to iron out any bugs or usability issues.

The plan for the development is to build up a basic framework for the admin area and the database, ensuring it is functional and robust enough for our needs. I will then work with Ryan to integrate the new design for the site with the system, whilst completely rewriting the code that deals with the checkout process, making the flow from basket -> checkout -> order confirmation much more simpler for the customers to use.Technorati Profile