Welcome/Artist Search Top 30 Most Popular Artists Monthly Features Art Museum Sites Worldwide Fine Art Links More About Us The Most Efficient Ad Targeting in the World Popup Glossary of Art-Related Terms
Artcyclopedia


Art Valley
 

Art Valley: The Marketplaces of Art

Florence, Italy
November 17-18, 2000

Digital Art: Methods and Business Models
Remarks by John Malyon


My website is called Artcyclopedia. It's kind of a "museum of museums". Artcyclopedia brings together all of the important art museum web sites and image databases from around the world, and organizes the artworks by artist.

The result is a database of 25,000 links to artworks by 7,500 great artists. The website is used by 15-20,000 people every day.

What I'd like to do is give you an overview of how it works, and then explain the underlying business model. In the process I'll describe a couple of concepts that, in my opinion, are key elements of any Internet business model.

Let me show it in action:

[New Window: Title screen]

This is the front page of the site. There are articles and a few other features available, but the principal way of using the website is through the database, which is accessed through the search panel in the center.

Now, I want to make an aside here: earlier I heard one of the speakers suggest keeping the number of clicks required to get to useful information to five or less. I completely agree with the sentiment, but would suggest that even five is too many. If at all possible, the number of clicks should be at most two or three.

There's an important principle in writing: with every extra word, you lose a reader. That's why the most important part of the writing process is editing, editing, editing. There's a very similar process at work on the Internet: with every extra clickthrough, you lose half your audience. That means that if you hide information three levels down in your website, only one-eighth of your visitors will ever find it.

I've seen this principle at work again and again, and that's why I'm careful to keep the number of steps required to get to useful information to an absolute minimum.

For example, if you type an artist name into the search box, you can often get to the page for that artist in a single click. Let's type in "Michelangelo", for example...

[New Window: "Michelangelo"]

Now, one of the advantages of a subject-specific search engine is that we can build some intelligence into the search process. In this particular case, "Michelangelo" is not a unique identifier, so the site needs more information. It suggests that the artist being sought is probably Michelangelo Buonarroti, but there are a few Michelangelos in the database (for example, Michelangelo Merisi, better known as Caravaggio). Now we can click on "Michelangelo Buonarroti" and go straight to the page for that artist.

The search engine can also recognize different versions of an artist's name (for example, Hieronymus Bosch is known as Jérôme Bosch in French and El Bosco in Spanish), and even some common misspellings.

So it took two clicks to get to the Michelangelo page, and on it we find some brief biographical information, as well as forty or so links broken down into categories: "Museums and Art Galleries", "Image Archives", "Articles", and "Other Web Sites". The links are organized by quality as well, so that particularly useful or important websites are listed first, followed by less important sites.

If you compare Artcyclopedia to a search engine like Alta Vista or a general Internet guide like Yahoo!, I think you'll agree this interface is much more appropriate for the subject matter and much more useful.

Now, one important reason for collecting information into a database is that a database is flexible. Data can be viewed in different ways and accessed through different paths.

There are seven different ways a visitor can access the Artcyclopedia database. Let's see if I can remember what they are...

1. By artist name, as we have already seen.

2. By artistic Movement.

[New Window: Movement]

If we click on this link, we are presented with a chronological list of major movements in art history.

[New Window: Baroque Art]

We can click on "Baroque", for example, we can a brief description of the movement and a list of dozens of artists working during the Baroque era.

3. By Medium.

[New Window: Medium]

If we click on this link, we are shown a list of fine art media.

[New Window: Sculptors]

From here you can click on "Sculptors", for example, and see a chronological listing of all of the sculptors in the Artcyclopedia database.

4. By Subject.

[New Window: Subject]

By the same token, you can browse lists of Landscape artists, Wildlife artists, or Portrait artists.

[New Window: Landscape Artists]

5. By Nationality.

[New Window: Nationality]

This is not so useful for countries with a huge number of artists in the database, such as Italy or the United States. But it is a good way to get a quick overview of Polish art, for example, or Canadian art.

[New Window: Canadian Artists]

6. By Gender. You can browse through a list of about 750 female artists, from Renaissance painters such as Sofonisba Anguissola and Artemisia Gentileschi, right up to the modern day.

[New Window: All Women Artists]

In principle, we could also provide a list of male artists, but nobody has ever asked for one.

7. Finally, you can browse artworks by title. For example we can start with a simple example, Leonardo da Vinci's Mona Lisa.

[New Window: The computer operator types in "Monna Lisa"]

"No matches found". I'd forgotten that the title is spelled that way in Italian. In English, "Mona Lisa" is spelt with one "n".

[New Window: "Mona Lisa"]

There we go. There are three versions of the Mona Lisa available, including the Louvre Museum of course, as well as links to where you can buy a poster of this painting.

This brings up an important point: the multilingual nature of the Internet. The Mona Lisa is probably the most famous painting in the world, but it’s still not that easy to find. The problem is that we’re using an English-language search engine to find an Italian painting residing in a French museum.

I can't offer the perfect solution to this problem, but I think it's clear that every museum website should be looking at putting it's content online in several different languages.



Now we come to the business model.

The reason I love doing what I do is that I create value for the visitors to my site. I create something out of nothing, like an alchemist.

I make the world a better place. In a very small way.

And that capability is the reason why the Internet is so important.

A business that relies on advertising, as mine does, must create value for its advertisers as well. It must make the world a better place for its advertisers.

In other words, you must create an environment in which your advertisers can make a profit. If they are not making a profit, your business is not sustainable. That's an important point that is often forgotten in discussions about doing business on the Internet.

What I do is allow advertisers to target their message with great accuracy, by placing a customized ad on one single page of the website.

For example a gallery wishing to sell a painting or reproduction by Pierre-Auguste Renoir can put their message on our Renoir page and reach an audience composed exclusively of people researching that artist.

An auction house promoting a sale of Inpressionist works could place a notice on the pages for each of the Impressionist artists in our database, and it would only be seen by people with an interest in Impressionism.

To use one more example, the Bayly Museum of Art in Virginia recently mounted the first-ever retrospective of the American Abstract Expressionist Adja Yunkers. They put an ad on the Adja Yunkers page of the Artcyclopedia. Now, Adja Yunkers is not that well known, and placing the ad was very inexpensive because the museum only pays for the actual audience that it reaches. But that audience is likely to be very interested in knowing of this temporary show.

In fact, the average response rate (i.e. clickthrough rate) for ads on our website is over 20%. Compare that to conventional banner ads, which get an Internet-wide response rate of less than 0.5%.

I believe this is a good business model: my advertisers are very happy and the Artcyclopedia is profitable right now. But this is just one possible solution. The other possibilities are only limited by the human imagination.



Rules for developing a success Internet business model:

(These rules are of course not sufficient for success - they don't guarantee a profit. But I believe they are necessary for success.)

Number 1. Make the world a better place. I've discussed one possible approach to doing so.

Number 2. Be the only one to do what you do. The only one in the world. Why not? On the Internet this is completely possible.

The reason for this is, if you cannot offer unique value, then your product will likely turn into a commodity. A generic product. There's so much competition on the Internet, and it's so easy to turn to a different supplier, that this is inevitable.

That's the difference between Artcyclopedia's ads and conventional banner ads. Artcyclopedia's ads are unique: there's nowhere else you can create this kind of customized ad or reach such a precisely targeted audience. Banner ads are generic: there are literally hundreds of thousands of websites trying to sell banner ad space, and although each tries to differentiate itself by claiming to have the wealthiest or most attractive audience, the reality is that they are primarily competing on price. The result is a huge glut of banner ad space, which I believe is irreversible.

If you are hoping to rely on banner ads for income, I hate to deliver bad news, but I strongly suggest that you look elsewhere. Look for that one great thing which you, and only you, can offer.

It's there!

And that's your business.