I established our corporate Web site with an off-the-shelf e-commerce suite, but am now having doubts about whether or not I did the right thing. With the amount of traffic we are dealing with increasing by the day, our solution is looking ever creakier. Now I'm wondering whether I should scrap our package and opt for a custom-built approach, or whether I should leave well alone now that I've set a course? Which strategy is potentially the most damaging to our e-presence?
Find out what your customers want
Many organisations react to increased traffic by simply increasing infrastructure, and shoring it up. This is important, but not nearly so much as understanding what is happening on your site. In other words, tracking true end-to-end customer and potential customer activities and thinking. With this information you will be able to focus your Web site developments and support around what your customers want. For example, the system may fall over at a critical moment, or be delayed long enough to cause a potential customer to give up and go elsewhere. Wouldn't it be wonderful to know when they gave up, why and where they then went? Technology to truly answer these issues is now becoming available, and should be the first course of action for anyone who wants to transform their e-presence and Web sites.
Chris Hilder, head of information risk management
Plan your upgrade now
You should count this as a success. You do need a good Web site. You need to develop something that can offer the right product, at the right price, with the right design and at the right speed. The differentiating features are limited on the Web so if your site is too slow or doesn't look as good as your competitors you might loose custom.
As you know, developing your own site from scratch will require good staff as well as substantial implementation time and money, but this still doesn't guarantee success. An upgrade to a faster and better off-the-shelf solution would probably be a more effective approach to take in developing your e-business strategy and certainly less risky. I suggest that you start to plan the upgrade now, while you still have time. You will need to anticipate when your current system will become unusable and to do this you will need to know the trends of use and the maximum numbers of users who could possibly visit the site. This information will enable you to plan your timeframe for the replacement.
Cranfield School of Management
Dr Elizabeth Daniel, senior research fellow in Information Systems, in conversation with Kevin Scott of Core Innovation and Dr Chris Hemingway, researcher
Consider your options
Leaving alone is certainly not an option! Customers in the physical world are increasingly disloyal to companies and it is expected that this absence of loyalty will be even more apparent in the online world. Customers using e-commerce have very high expectations - they expect to be able to find accurate information that matches their needs quickly, every time. If firms fail to provide this to customers, their competitors are only a few clicks away. A well-known rule in marketing is worth bearing in mind - it is more profitable to keep an existing customer than to acquire a new one.
So, what should you do? You have a number of options. You could analyse the traffic you are generating to your site. Is there something about your service that you could alter to help with this traffic problem? Do you have time-critical offers or services that cause peaks in demand? Is there some way you could encourage customers to spread their demand more evenly? What are customers doing when they are at your site? Could you offer customers routes through your site that would reduce the demand on your system?
However, if the increase in traffic is due to growth that is likely to be sustained, you will most probably have to consider changes that will increase your capacity. First, it is necessary to forecast your likely traffic growth over the next year or so. Given this estimate of future demand, you could see if there is sufficient expansion in your current solution either through additional or upgraded hardware or by simultaneously operating mirror sites. If that is not possible or will not provide sufficient capacity for you, you could look around for a bigger off-the-shelf package. It is important to understand if such a package is scaleable to cover your forecast growth. Finally, the third option is to specify a custom-built solution. Once again, ensure any solution built is fully scaleable so you can meet your forecast growth - and with a generous amount of extra capacity to grow into if it transpires your forecasts are too modest!
Computer Weekly Professional Series, Butterworth Heinemann
IT is a labour of Sisyphus
Every IT manager should know about the ancient Greek myth of Sisyphus. This Greek hero displeased the gods so much that he was punished by being given the task of pushing a huge rock up a steep hill. As Sisyphus managed to get the rock to the summit it slowly rolled back down to the base of the hill and he was required to push the huge rock back up to the top again. Sisyphus was condemned to do this forever. Does that sound just like an IT manager's job? No sooner has the project been completed than the users have a whole new wave of requirements and it's back to the drawing boards again, or at least back to preparing a new release. Simply, the IT job is never finished.
So why should this be different with an e-commerce suite? It doesn't matter a jot whether the system was an off-the-shelf package or custom-built. It will need enhancing and upgrading before long and maybe even throwing away and starting again.
The simple rule is go and find out what the business really needs today and deliver solutions to current business challenges. Sometimes you have to admit that you haven't got an optimal solution in place and, when this happens, the best way forward is to replace it as soon as you can.