Interpreting The Numbers: How Well Is The VIP Art Fair Actually Performing?

by Paddy Johnson on January 28, 2011 · 10 comments Art Fair

Salon 94 at The VIP Fair

“As of Thursday, artworks in [VIPs] virtual booths have logged 5.3 million pageviews from online browsers since the site went live” ArtINFO reports. This sounds like a really impressive number, but I wonder how much it tells us. I did a few calculations, and came up with roughly 2,160 images displayed on the site (the total number of booths and respective image allotments + 5 percent for swapped images and private showings). This would mean that each image received 491 views per day (assuming they are counting image views as page views, which I assume they are since each image has its own url). That seems like a lot to me, but also not particularly relevant. As a dealer I’d want to know how many people have used the site, how much TIME they are spending on my images, and how often they are returning. I wrote VIP requesting those stats, but haven’t heard back. In lieu of this, a few reflections and estimates from a friend in the development business:

You can’t really apply a standard e-commerce pageviews/visitor here, because the site is so heavily browse-driven. I assume the stat they’re surfacing for the press counts a pageview for every artwork viewed, including the clicks that scroll from work to work in that AJAX “gallery wall”. As you point out, VIP Art Fair are keeping the more interesting stats for themselves; pageviews aren’t really interesting–time on each image, average and breakdown of number of galleries visited, and total unique visitors are the numbers you want, and extrapolating is tricky.

That said:

We know they had a huge concurrency problem when they went live, which suggests they actually did have a ton of people visiting. In online magazines and newspapers, you’d expect something on the order of 20-30 pages per visit, and this model should be higher than that if they’re counting each artwork view as a pageview. I’d figure something like 6-10 “galleries” visited, with an average of 10 images per gallery [Editor’s note: This sounds about right to me]; let’s use 60-100 pv/v.

Somewhere between 53K and 88K visitors since launch, with 60-100 artworks viewed on average? Doesn’t seem off by an order of magnitude. Some of those visitors will count twice for visiting on different days, so the actual number of human beings visiting the page will be a bit less than that.

As for time on site, you have to figure that if they’re using one pageview for every artwork shown, time-on-page has to be far lower than average: probably < 10 seconds, probably around 5-10 seconds. That works out to anywhere between 5 and 16 minutes on site–I don’t trust that number so much, but we’re not talking 1 hour on site, and we’re not talking one minute either.


Forrest January 28, 2011 at 11:24 pm

500 views per image actually strikes me as incredibly low. Even if those were all absolute unique visitors, that would mean each artwork is getting seen by an average of 500 people. How many go through each booth at a normal fair? And how much bigger is the percentage of non-collectors here than at a normal fair? The barrier to entry is so much lower (you don’t need to travel to a different place, and it’s not intimidating at all), it seems like many more casual visitors would be going through with no intention of buying.

It will be interesting if we hear about sales, which presumably matters much more than people looking at the pictures.

Anonymous January 28, 2011 at 11:30 pm

uhhh. That’s because it is low. I meant to write that that number was per day. I’ve updated the post.

Chris Rusak January 29, 2011 at 12:06 am

I think the near-future of this type of art fair depends on only one metric in the eyes of the gallerists: return on monetary investment, a figure I believe outside the scope of whatever analytics are embedded in the site.

Carolina A. Miranda January 29, 2011 at 11:48 pm

@forrest the numbers could be low, but could this be a more motivated audience than one found at a fair? seems to me that lots of folks at fairs are there to show off their outfits/arm candy, not look at or buy art. i’d assume that someone browsing an online art fair would be pretty darn motivated because it’s absent of the social experience a fair offers. but that’s pure conjecture on my part.

Luuk Christiaens January 31, 2011 at 12:21 am

500 views per image actually strikes me as incredibly low…. How many go through each booth at a normal fair?…

I looked up my stats for some major art fairs (Art Basel, FIAC, Frieze) and the number of visitors is even (much) lower than 500 based on the results communicated by organizers.

Forrest January 31, 2011 at 5:49 pm

@Carolina Actually, I think way fewer people (by percentage) are at art fairs just to be there. There’s some percentage of locals who just come, and some percentage of people who are there for other professional reasons (journalists, artists, etc.) But the internet has far more “locals,” and all of the people who come to Miami for example but don’t spend have probably visited VIP.

The barrier to entry is just so much lower: you don’t have to pay to get in, you don’t have to stay long, you don’t have to travel (either to another city or even within your own city). I’m not buying anything and I’m not covering VIP, but I still logged in for 10 or 20 minutes just because it didn’t require much (beyond clicking the “resend activation e-mail” link 4 or 5 times over the course of 2 days).

For everyone who likes to be seen at art fairs, there is another person who doesn’t like crowds but likes to look at art, right?

@Luuk A little less than 50,000 people went to Miami Basel this year, even if the average visitor only attended 20% of the booths (which seems like a comically low estimate) that would be an average of 10,000 people per booth, right? I would be interested to know more about what statistics your referring to and analogizing to the image views.

Luuk Christiaens January 31, 2011 at 9:16 pm

@Forrest Two remarks.
10,000 visitors per booth would result in 2,000 visitors/day or a constant of 250 visitors/hr during the fair… (time per contact for the gallery owner would be 14,4 sec.)

20% of the booths = 50 booths (ABMB announced 250 participants). 50,000 visitors visiting 20% of the booths = total of 1,000 visitors/booth over the 5 days.

Forrest February 1, 2011 at 2:14 am

Right, so even with absurdly low assumptions we would be getting around double the figures, and this is people going to a physical place and paying money in many cases to get in. I would think to have a comparable impact to a fair like Miami Basel, VIP would need to see many many times that result.

You are right though, as I and many others said I think the real metrics that matter are related to sales. Galleries can put their own jpegs online, the question is does this fair expand their ability to move inventory.

Luuk Christiaens January 31, 2011 at 10:51 pm

@Forrest My reply reads a bit harsh. Unintentionally. Sorry. I don’t think we have to focus on ‘numbers of visitors’ solely to measure success. It’s but one parameter (an important one I agree) but the quality of visitors is much more decisive. “After a few participations at a fair, you know exactly who to invite at the VIP evening. It’s your best guarantee to make your participation a success” a gallery owner once told me. We all know that the VIP evening is crucial and all gallery owners measure their success on the results of that one evening. The four remaining days… well. Maybe it’s one of the reasons James Cohan stressed so much the 1/5 of the price of a normal art fair when presenting ViP.

Anonymous February 1, 2011 at 5:17 am

This is one of the reasons I find it so annoying that there’s no ability to purchase works through the site. You really want to know how many people visiting are actual buyers and I’m not sure I understand why there’s a need to remove that kind of functionality from the site. Maybe they couldn’t develop it fast enough?

Anyway I’m sure at the very least the length of time people stayed on the site and looked at individual images correspondes with purchasing so I wish they would release those numbers.

When I read that only a dozen galleries had sold work on last Tuesday (January 25th), that sounded far lower than the 500 people per booth stat.

Comments on this entry are closed.

Previous post:

Next post: