E-Commerce in Higher Education: Can We Afford to Do Nothing?

With e-commerce emerging as the driving force behind commercial websites, its impact is beginning to be felt in the non-commercial sector. Those of us working in HE, in particular, need to prepare carefully if we are to reap the rewards and avoid the pitfalls by establishing appropriate ‘business’ structures as well as refining technical and managerial approaches.

These were conclusions of the e-commerce parallel session at the Web Management Conference at Bath University in September. A survey of existing and planned e-commerce systems in HE, illuminated by a HE case study and a detailed presentation from online payment and security specialists Trustmarque plc were the main features of the session(1).

I have tried to draw these threads together to look at the state of e-commerce in HE and to see if “Essential. It’s happening!” (a comment from the survey) really sums up the present situation, and likely future developments.

Online e-commerce survey(2)

Of the 76 HE organisations who responded over 80% claim to have some kind of e-commerce system in place now or plan to by the middle of 2001. (E-commerce here is defined as having an online system for ordering and paying for goods or services, although the method of payment may not necessarily by credit/debit card.) In fact, 70% of the institutions surveyed will continue to use paper-based systems (such as invoice generation, follow-up cheque) rather than online transfer of funds. More secure or false economy? Surprising conclusions reached at the parallel session are discussed later.

Selling publications, payment of fees and alumni goods make up the largest categories of transactions (74%). Other uses included supplying advice and information (5%), sales of concert/theatre tickets and printing/photocopying charges.

So who runs these systems? In many cases it was web staff (60%), although academic departments themselves show significant involvement (13%). Perhaps surprisingly, Finance Departments appeared to be involved in very few cases (6%). Only one institution surveyed had a project officer dedicated solely to e-commerce development.

On a technical note, almost half the respondents have a digital security certificate (45%) (regarded as essential for e-commerce as it assures your clients of the security of your system and authenticity of the institution), although the majority have secure servers (SSL) in operation.

Security and satisfaction

The survey also looked briefly at the online shopping habits of the individual respondents, with the aim of seeing how much our consumer experience might affect our approach as providers. Almost all (98%) of those who had bought goods or services online were satisfied, and most (92%) regarded this as a safe method of making payments. Incidentally, the most poplar sites used were Amazon (62%), Easyjet (11%) and Go-fly, Expedia and Tesco equal in third place (4%).

The survey contrasts sharply with figures given at the Parallel Session by John Williams, CEO of Trustmarque plc(3): Taking a world view, 25% of e-commerce transactions are never fulfilled, and 53% of online customers are concerned about security and privacy. Furthermore ‘conversion rates’ of visitors to buyers currently run at about 1.8%. Evidently the experience of e-commerce in HEI is more positive, and this may be an important starting point when planning or running such a system.

Credit and credibility

The ‘brand’ of UK HEIs already has much global credibility – no one suspects that a British University is going to steal your identity or defraud your credit card, and they already have a reputation for providing ‘quality’. This may help to reassure users in the future who may be paying substantial course fees online, or signing up to an expensive distance-learning MBA.

Apart from this, there is much that can be done to re-assure users of the reliability and safety of what you have to offer:

• Exemplify privacy, disclosure and quality control policies and : how will you deal with a complaint if something goes wrong? Do you sell people’s e-mail addresses etc?
• Make it clear what the customer is buying and who they are buying it from. Who can they contact for help?
• Consider seals of approval – logos from VISA, etc., your security certificate provider (such as Verisign or Thawte) and the institution can build confidence.
• Make the design of the e-commerce site as professional, easy to use and as fast as possible. Quality institutions produce quality web sites!
• Decide on the level of security needed and deploy appropriate technology
• Consider outsourcing - it may turn out cheaper in the long run. Online credit card payment and transfer of funds would have to be outsourced in any case.

It’s a business

Doing e-commerce has far more to do with running a business than designing websites. It’s better to think of opening a shop; the stock, the staff, the systems, the pricing and relationships with banks, etc. all have to be in place before you can open the door to customers. Many of the spectacular e-commerce failures of the last few months have involved ‘fulfilment’ problems, whereas few have failed because their website design or choice of technology was wrong.

Three likely business models are:

1. · Physical fulfilment model
   The web site allows customers to order and pay for items that will be delivered later. Examples include books, recordings, alumni t-shirts, etc. Stock control is needed, preferably linked to the online ordering system, and the cost of delivery needs to be taken into account. Policies on returned items, non-delivery, breakages etc, need to be decided upon and made clear to the customer.
2. · Online fulfilment model
   In this model the item purchased is fulfilled on line, either at the time of purchase or later. Examples of this are software, electronic journals, digital-art, etc. This model may be useful for the supply of learning materials for distance learning courses. The onus of correct fulfilment falls on the technology employed – having paid for a piece of software online the customer needs quick and robust access to the data. Telephone/e-mail support might be necessary.
3. · ‘Ticketless travel’ model
   At least one ‘budget’ airline now sells 75% of its flights through its website(4) – there is no actual fulfilment, apart from the supply of a number or code to be produced at the time of travel. This model could be used for selling concert and theatre admissions, conference registration, payment of course fees, printing and photocopying charges, etc. As in each of these cases supply will be limited (e.g. by the capacity of the theatre) the system needs to be carefully designed. What happens if the customer’s choice is fully-booked and thus not available?

It is unlikely that one e-commerce web site could service more than one of these models – they need different management, resources and technologies. In the case of models 2 and 3, the products to be sold need to be designed with e-commerce in mind.

Recent experience at Goldsmiths College with online conference bookings showed that having numerous accommodation options all with slightly different prices was confusing to the user and increased the complexity of system design. An ‘electronic version’ of a paper-printed journal that consists of a few PDF files is unlikely to be as good quality a product as one initially designed for the web. In the case of the Physical Fulfilment Model, the price of a small item (such as a pack of Christmas Cards) may turn out to exceeded by the cost of stocking, packing and delivering it. As with any business venture it is vital that all direct costs, overheads and other expenses are taken into account when setting up an e-commerce system.

Getting the money

As well as different business models there are also different ways in which money can be obtained from the customer. In choosing the most appropriate method security, costs, speed and convenience need to be considered. Inevitably most e-commerce revolves around the use of credit/debit cards, but in the HE sector (especially for ‘business to business’ transactions) this may not yet be the most appropriate means. The survey indicated that 52% of the respondent organisations had purchased items and services online. On the other hand, it emerged at the Parallel Session that not all HEIs allow payment to be made by credit card in any case. Therefore, choosing the right payment system may be crucial to the success of an e-commerce venture:

· Full online card payment and fund transfer

The customer submits card details which are passed electronically to an ‘online payment service provider’, such as Barclays Merchant Services, Trustmarque, etc. the funds are transferred immediately and fulfilment can commence. Costs: normally a charge is made per transaction ( about 25-50p, depending on turnover) plus the normal percentage taken by the card company.

· Off-line card processing

The customer submits their card details, the number of which can be checked using a modulus analyser for validity, which are then send in an encrypted e-mail to a finance officer who processes the transaction using a ‘PDQ’ machine, or similar. The success (or otherwise) of the transaction needs to be fed back into the e-commerce system before fulfilment can begin. Costs: the time of the finance officer and the card company’s percentage.

· Payment by Standing Order, Direct Debit or Cheque in post

As all these instructions require a signature, the only feasible way to do this is to generate a document which is printed off and returned by post, signed by the customer or with the cheque. The order needs to be ‘held’ until this is received and processed or cleared. Costs: manual processing time.

· Generation of invoice for subsequent settlement

Perhaps only suitable when the customer is already known to the seller. The goods or services are provided ‘on trust’ on the assumption that the invoice will be paid in accordance with the seller’s terms and conditions. Costs: processing time, risk of default.

Certain business models imply one or other of these payment methods: for example, immediate online fulfilment requires immediate online payment processing, whereas physical fulfilment may allow a margin of time for off-line processing or for a cheque to be received. As in the traditional non-electronic business-to-business trading model, the supply of goods pending payment of invoices requires knowledge and trust, and an assessment of risk needs to made which may be reflected in the price.

Although the full online card payment and fund transfer method appears to be the most expensive option, practical experience at Goldsmiths has shown that this may not in fact be the case. Manual processing of card details has many pitfalls, such as the details being submitted incorrectly, insufficient funds available, wrong matching of payments to customers, etc. Good system design and adequate training could overcome many of these potential problems, but bearing in mind the cost of this and time taken to sort out mistakes (let alone international telephone calls!) a full online payment system may be better value, especially if the value of the transaction is relatively high, as in the case of a conference booking, for example.

What do you need?

The technology for e-commerce does not need to be hugely powerful, complex or expensive, especially as realistically the numbers of ‘hits’ on an HE site are not going to be enormous, especially at first. Online Conference Booking systems at Goldsmiths have successfully used a variety of systems, ranging from Mac-based Filemaker Pro databases to a more robust solution using PHP on a secure Unix server. It’s unlikely that much online trading could be achieved without using at least one database (to record the transaction details), and in reality most of the content of an e-commerce site will probably be database-driven. CGI scripts, PHP, Active Server Pages or Cold Fusion are commonly used methods of linking databases to HTML pages, and this needs to be done by someone with a thorough understanding of the chosen technique.

Once your system is set up on an SSL (Secure Sockets Layer) server you will need an digital security certificate(5) to prove it. These can be obtained from a number of companies of which Verisign(6) is probably the best known. Expect to pay £200-400 per annum for this depending on how many servers you operate.

Typically an HE e-commerce site will involve personnel from many departments, possibly ranging from the Alumni Office to Finance and the Computer services. It is vital that all those involved fully understand how the system works and that there is effective communication between them, just as on a shop floor. Real team work will make the venture a success, so avoid having to rely on sceptics or doom-and-gloom merchants! As with any kind of team project there needs to be definite leader or manager empowered to make decisions quickly and unilaterally.

Perhaps above all, you need to advertise the existence of your e-commerce offering, so a marketing strategy is vital. Simple things like making sure the URL appears on printed material, and that there are plentiful links from the ‘main’ website must not be overlooked. A direct e-mail campaign may the most effective way of building potential customers. Probably as much effort needs to be put into marketing as goes into all other aspects of building and running an e-commerce web site.

What do we all need?

Many recent national initiatives have implications for the way HE uses the web to market itself, such as HERO(7), UCAS entry profiles(8) and the e-university proposals(9). As yet there does not seem to have been any parallel attempt to support the growth and management of e-commerce in HE. There are a number of issues that can only meaningfully be tackled at a national (or even international) level:

Authentication standards

The use of ‘digital signatures’ or other means of authenticating a potential student or customer is still in its infancy and not standardised(10). This is seen as a major hurdle in accepting applications online, organising video-conference interviews (how do you know who you are really interviewing?), course-work submissions, etc. Until progress is made in this area the use of online payment systems will be limited for many types of admissions transactions.

Kite-marking

A nationally-agreed quality assurance scheme for HE would re-assure potential online purchasers that the site was reliable, secure and run by the institution named. This would be exemplified though a logo and a link to the agreed terms and conditions relevant to the transaction.

Site design standards

There is nothing more worrying to a user than, having gone through several screens of menus and forms, finding that the site hangs or dies. It may then even be difficult to know if the order has been sent or if the payment has been deducted. Many of these pitfalls could be avoided through the use of standard technologies (especially HTML) and careful browser-compatibility testing. A central online ‘e-commerce validator’(human, maybe) would be useful, as it can be difficult to validate dynamically generated HTML with existing tools.

National portals

Portals covering key relevant areas of HE e-commerce (such as online journals, alumni goods, ticket sales) would provide a show-case for developers themselves where ideas, approaches and technologies could be compared as well as an easy interface for potential customers. A single point of access to all UK HE publications available online would be an example of a useful portal.

‘CHEST style’ deals

Nationally-negotiated deals with online payment processing services and off-the-shelf packages would be financially beneficial and ensure that chosen suppliers are credible and reliable.

Training

Bringing together training that covered basic business analysis, stock control, marketing etc. in conjunction with relevant IT skills would enable more institutions to benefit from e-commerce. Online course support and documentation would perhaps be the most appropriate means of delivery – learning by example.

Conclusion – do nothing?

In drawing a conclusion I wonder what would be the effect of doing nothing. In the short term probably not very great, but institutions who don’t start now may well find it harder later; comments from the e-commerce survey indicate that there is a steep learning curve for both the person tasked with developing an online payment system and the institution of which they are part. One of the challenges facing web site managers has been bringing together disparate people and parts of a college or university to convey a useful and attractive image on the web. With e-commerce it seems we may have to repeat the process of forging key alliances, choosing the right technology and devising a business plan, but with the even more ‘mission critical’ objective of making money. Already e-commerce has revolutionised banking, travelling, and book selling. Other financial services, recorded music and food shopping seem likely to be next in line. It’s happening in Higher Education among those institutions who see it as essential to their future prospects. I wonder how many will be doing nothing in a years’ time.

References

(1) A full report of the session is available at http://www.gold.ac.uk/webmanager/e-commerce/index.html

(2) The aim of the survey was to discover the scale of usage (and planned usage) of e-commerce systems within the HE sector. The web-based questionnaire was available from 17th August to 1st September 2000.

Full analysis: http://www.gold.ac.uk/webmanager/e-commerce/ecommerce-survey.htm

(3) Trustmarque plc is a provider of ‘trusted e-commerce systems’: http://www.trustmarque.com

(4) EasyJet, reported on http://www.easyjet.com/en/about/

(5) Also referred to as digital passports, X.509 certificates, or public-key certificates.

(6) Verisign: http://www.verisign.com

(7) HERO: http://www.hero.ac.uk/

(8) UCAS entry profiles: http://www.ucas.ac.uk/profiles/index.html

(9) E-university information from HEFCE: http://www.hefce.ac.uk/Partners/euniv/

(10) See Overview of Watermarks, Fingerprints, and Digital Signatures (a JTAP report) at http://www.jtap.ac.uk/reports/htm/jtap-034.html

Author Details