News and links
Developing an Indexing Product: www.winediva.com.au
Paper delivered at the 3rd International Conference of the Australian Society of Indexers. Indexing the World of Information. Sydney, September 2003.
The seed for Wine Diva was planted at the Robertson Conference Indexing in the Electronic Age in 1996.
At this conference publishers were invited to give their perspectives on the future of publishing in the electronic age. They were also invited to give some indication about what role they saw indexers playing in the development of electronic publications. Trouble was they were grappling with the issues themselves and made only oblique attempts to figure indexers into the electronic equation.
It was a very good conference because it did what a good conference should do and that is to spark people's imagination and their resolve to examine their current circumstances in a different light. When I left Robertson, I could see that as an indexer I had to determine what my future was going to be and not leave it in the hands of the publishing houses.
I also knew at this time that indexing was a very difficult business to be in. The nature of the work meant that there were stretches of idleness followed by frantic flurries of activity that invariably coincided with social commitments and long weekends. My business was also reliant on a lot of smaller publishers which I predicted would go the way of many businesses as the economy 'rationalised' industries into smaller numbers of players. The trend had in fact set in during the eighties so it was well under way when I started to consider a shift in direction.
The structure of the indexing industry itself was also of concern. The word free in freelancer was unnerving. Pushing fees up while remaining competitive was difficult and with the introduction of the GST, impossible. I also started to get a hankering for the trappings of business, an office away from home, and staff so the business did not grind to a halt every time I had day off.
What I had observed in successful businesses was a shift from the provision of services to products, which was odd given the pundits were telling us we were living in a service economy. However my local petrol station was now bent on moving product rather than filling my tank. Time was becoming both an expensive input and an exploited resource.
I decided that I too needed to develop a product so I could release my business from the constrictions of time and thereby through my utopian coloured glass came my primary hope that time would be reallocated to living more fully rather than to work alone.
Question - What kind of product should I develop? I concluded that an Internet directory would be feasible. Nearly every time I searched the Internet I ended up grumbling about the length of time it took and musing on how I could to it better. Good directories where no were on the horizon as the vein of thought that presumed natural language searching would deliver results was still coursing through the IT industry. Of course it did produce results, hundreds of thousands of results every time I put a search term in. Human intervention was needed. It would also be financially feasible as I did not require a large amount of capital to get a pilot up ... only time and the expertise of others was needed.
So Wine Diva was spawned in a pond filled with skills and experience, preparedness to take up a challenge, a perceived need for better search tools on the net, a gap in market for Internet directories that worked, economic necessity and the support of family, friends and colleagues.
Sketches were made in 1996. These, however, were early days for the Internet and the technology to translate the ideas into a working site was still to come.
Why the Australian wine industry?
My initial interest was in the horticultural industry/gardening. However, it is a multifaceted industry with slow uptake of the web so I looked to the wine industry as a horticultural industry which did have sufficient numbers of organisations on the web. I also needed to develop a pilot directory to minimise fallout if the model proved uneconomic. The wine industry is a wealthy horticultural industry with a large number of individual businesses that are isolated from their main markets, the cities, and rely heavily on export markets. In 2002/03 there were 1,625 wineries producing 910 million litres of wine with export earnings of $2,386,000,000 (1). This adds up to an industry that must place a huge emphasis on marketing if they are to successfully export, and draw wine tourists from the cities to their cellar doors. The Internet as a marketing medium is cost effective enough to allow small and large players alike to market their wine.
It is also a good industry to target as there are large numbers of peripheral wine businesses and organisations that can be included in the directory such as wine merchants, education course providers, tourism service providers in wine regions and suppliers to the industry itself such as cork and barrel manufacturers.
Information Model: the principles
Principle 1. When people go in search for information they will try and go to the source if they can.
The first decision was to have a directory of links to connect the wine lover and wine professional to wine and wine related organisations in Australia. Let the wine sites themselves satisfy the user's need for information.
As an indexer I would be building a pathway to the information not re-constituting it.
Principle 2. Provide information at the level people request it
Putting users in the driver's seat when it came to information searching meant building a database and introducing a classification system into which the contents of wine web sites could be classified. We commissioned Master of Wine, Toni Paterson to provide information about grape varieties, listed the wine products, which we retrieved from web sites and classified products under wine type, i.e. red, fortifieds, dessert, etc. We also classified content using categories such as organic, sacramental and non-alcoholic and then preceded to include such user centred categories as cellar door, restaurant and accommodation. Classification has also had the advantage of bringing like things together so the user can for example search for all the wineries in the Hunter Valley which produce Semillon and have a cellar door and a restaurant or all the organic wine producers in Western Australia or vineyards who grow the Barbara grape variety in the Mudgee region.
This was about putting database power into an Internet directory so the data could be crunched the way the user wanted it to be. It was the decision to build a database that differentiated Wine Diva from other wine sites. It also meant we had to wait for the technology i.e. MS SQL Server, Ultra Dev and ASP all of which were not available in 1996.
Principle 3. A directory which fails the comprehensiveness test; fails
The decision to give each organisation a free basic listing was critical to the success of the project. By the time Wine Diva was launched in 2001 many a directory had been and gone on the Internet because their designers had hoped to make people pay for a listing. With only a limited number of listings these directories were unable to provide a comprehensive information service to users.
Principle 4. Index the wine industry in Australia for links as you would a text
This meant following indexing process, this being:
In an online index of links (directory) this translated to:
Structure - finding out how the wine industry was organised and how the industry marketed its product. As a consequence of this Wine Diva structured its listings according to Australia's wine regions system referred to as Geographic Indications (GIs). The Australian Wine and Brandy Corporation are responsible for the GI system, which was introduced to Australia on the insistence of the Europeans who would not import Australian wine unless the labelling of the wine was strictly controlled on a regional basis. The aim - if a bottle of Cabernet said Coonawarra it legally had to come from the Coonawarra. Australia has 55 official wine regions.
This has proven to be one of the keys to the success of Wine Diva. It has enabled us to join the marketing stream the wine industry adopted to market and promote wine as well as enabling us to provide a comprehensive wine tourism information service based on the regions. This has also increased the number of listings and therefore extends our potential revenue base.
Content - firstly we researched a large number of wine web sites to see what kind of content they were offering, how the content was structured and what terminology was used. From there we constructed our categories for the database. Once this was done we then started in earnest to index wine sites. For our category Wineries & vineyards we collected data for each winery and vineyard on: state, wine region, wine type, wine products, grapes grown, special wine, (organic, kosher etc) cellar door, accommodation, restaurant, facilities. This process was repeated for all our categories.
Style - style on the Internet is a very different thing to style in a printed index; however, some of the same principles apply such as readability and accessibility.
Principle 5. Never build a website without doing usability
Indexers are experts when it comes to categorisation. We can build sites using apposite categories to reflect the content. The web, however, requires an additional approach that is commonly referred to as the "top down approach". Alinta Thornton speaking at 11.00 am will go into this in greater detail. It is about finding out how users respond to categories, navigation, layout and content and is absolutely essential if you want your website to be used effectively by a wide range of people.
In designing Wine Diva we drew our initial categories from the web sites we were indexing and then we tested them on users to come up with the categories used in the menu today. We also tested for content and asked, "What do wine consumers want in a wine site" as well as asking users about navigation. The answers were revealing and makes you realise that how you search for information as an indexer is not necessarily how other users search.
Principle 6: You can only build a site like this with the right people
From the beginning Wine Diva utilised the talents of exceptional people in a range of fields; Anne Melano who programmed the site, Alinta Thornton who advised on and conducted usability; Trevor Colton who designed the Wine Diva logo and other original artwork, Toni Paterson, Australia's first female Master of Wine who writes articles for Wine Diva and Kate Lyons-Dawson who maintains the integrity of the database and is involved in marketing and creative development of the site.
The Wine Diva site will not be a success unless it generates revenue.
In developing Wine Diva we needed to make critical decisions that would either make or break the financial viability of the site. The first thing that was done was to choose a long-term view. The Internet is not the place to make a fast buck. The Internet does have its success stories, such as Hot Mail and e-Bay but these sites developed from wonderful ideas and a solid foundation of development. In other words they have done the groundwork. The principle to slogging it out applies whether you are big or small on the Internet.
Principle 1. Labour costs should be minimised
This was reflected in the decision around criteria for inclusion in the directory. It was an Internet directory so organisations had to have an Internet presence. We surfed the Internet and established the directory with an initial listing of 650 wineries.
The directory would not be economic to build if we had to mail out to organisations for their details. People are notorious for not filling in 'yet another form' for a directory so we needed to do this work ourselves. We had to get organisations onto the site fast without the costs of a mail out. We procured the necessary information for listings from the websites themselves. We also placed a registration form on the site to allow organisations with just email to register.
Today we have 987 wineries and vineyards listed, most of who have websites.
Principle 2. Don't compromise the success of the information model for the sake of short-term financial gain.
We knew that the directory had to be comprehensive therefore a basic listing had to be free. Free information services are common on the web. In Internet marketing speak it is called mainstreaming were the purpose is to provide a free service or product in order to gain the lead position in the market. In directory terms that meant providing a free basic entry coupled with revenue raising advertising for people who want more visibility and better positioning on the site. This model is not new since a phone book uses that same model. The directory's revenue model is referred to as 'ad-supported content' because content is made freely available over a network. This is in stark contrast to print publishing, which represents 'user pays content', for example book purchases or in the case of magazines a mix of user pays and ad-subsidised product. Internet commentator Clay Shirky believes that the Internet is very similar to TV in that it attracts attention, therefore ad based revenues are feasible...."not only can ad-supported content work on the Internet, I believe it can't not work. It's success is guaranteed by the net's very makeup - the net is simply too good at gathering communities of interest, too good at freely distributing content, and too lousy at keeping anything locked inside subscription networks, for it too fail." (2)
Principle 3: Provide a range of products so small, medium and large businesses can access advertising on the site as well as other services
Wine Diva's primary revenue base is drawn from advertising. We sell:
Other products include:
Today: Wine Diva records over 8,000 visitor sessions per week.
Sales of all products have been made albeit movement has been slow.
Just as in 1996 we had to wait until the technology caught up in 2003 we have to wait until business catches up. All our marketing so far has included an education component so we raise consciousness about the value of advertising on the Internet and at the same sell advertising.
We have to overcome the bad press around the Internet with the dot com crash which is well remembered in the wine industry due to the failure of big players such as Wine Planet. However, if you go onto Wine Diva today and check out the e-merchants and e-wine auctions you quickly realise that the online wine business is in full swing. Unfortunately a bad reputation takes a long time to dispel.
The attempt to create a product using the skills and know-how of indexing has I believe proven to be very successful.
It is also good to know that it is possible to cross the species divide so an indexer can become a publisher and bookshop owner.
Postscript: Wine Diva closed down in 2014. The information model was popular, but the financial model was not.
Designer: Alinta Thornton