A lot of you are going to disagree with what follows. Especially if you use one of these services or, worse yet, have spent a lot of your time, energy, and money building one of them. To you folks who will dismiss the analysis and predictions below, I have this to say: I hope I’m wrong. I really do. But unfortunately as I see it today, the recent rash of community-based tasting note web sites will not succeed.
ONCE UPON A TIME
When I started this blog two and a half years ago, one of the first folks who contacted me asking for a reciprocal link were the guys running a service called WARPA. I never really knew what that acronym stood for, unless they were suggesting that they were a wine version of DARPA (the Defense Advanced Research Projects Agency — the guys who invented the Internet). In any case, these guys had a nifty little service that they pitched as the answer to the hegemony of the giants of wine criticism. As they saw it, they were going to inject a little democracy into the wine reviewing process. The idea was simple: have lots of people write tasting notes and scores for lots of wines and with enough reviews in the system, it would gradually, through the power of statistics, become a much more authoritative source on which wines the majority of wine drinkers actually liked, and what they might like if they tried them. Think “Amazon.Com customer reviews for wine” or some such notion.
In principle this isn’t a bad idea. But as Yogi Berra once said, “In theory there’s no difference between theory and practice. In practice there is.” But I am getting a little ahead of myself.
THE NEXT WAVE
For a couple of years these services operated (or should I say, languished) in relative isolation. None of them really picked up any steam, for reasons that I will come to in a moment.
Then, Web 2.0 happened. About 18 months ago, AJAX started popping up on the Web. MySpace was taking off, LinkedIn had built critical mass, Google was wowing the world with maps and the power of search, and new services like Flickr and Del.icio.us were showing that there was a future (not to mention a lot of cash) in user-generated content.
It was only a matter of time before someone applied these principles to the world of wine, and sure enough, starting about six months ago, I began getting e-mails about every three weeks from new websites that promised to bring the power of Web 2.0 to wine. Corkd, LogABottle, Winelog.Net, TastyDrop, OpenBottles. The list is still growing, as I’m sure there are a couple of others still in development.
While they may have slightly slicker designs than clunky old WARPA, they are all based on the same model, and founded with the same principles — amassing a large community base of tasting notes as a viable alternative to the established oligarchy of wine criticism that you might call The Park-Tator.
Yet, just like WARPA, all of these services suffer from the same four fundamental flaws.
THE FUNDAMENTAL PROBLEMS
#1: In order to be really useful, you gotta have a hell of a lot of wines in the system
Imagine going to the library searching for a book on American rattlesnakes and once there, finding that not only does the library not have the book on American rattlesnakes you wanted, but they only have one book on snakes, and it’s about the benign garden snakes of southern France. The librarian tells you, “well, you could write the book yourself, then we would have the book you need!” but what you really wanted was to take a look at the book and see if you wanted to go out and buy it.
Most of these sites have exactly the same problem. Because all the content is user generated, it takes an incredible amount of time to get enough wines into the system to make it useful to anyone. Add to that the fact that the majority of people only drink a very small minority of the wines available on the market, and you’ve got a huge problem getting anything more than the few most common wines into your system with any certainty.
Don’t believe me? Try searching any of these services for one of your favorite wines. Nothing totally obscure, but something reasonable like, say, a 2003 Spencer Roloson “La Herradura Vineyard” Syrah. If you’re like me, you come up with no results 18 out of 20 times. If I’m looking for tasting notes or recommendations on when to drink a wine, or what to eat with a particular bottle, none of these services actually help.
#2: We users are stupid and we don’t know how to write.
User generated content is always dependent on the user. Unfortunately one of the biggest problems with users is that they are not dependable. They make mistakes. Or worse (and much more common) they don’t know enough about wine to actually enter the wine names into the site correctly. My last search one one of these sites was for a Chateauneuf-du-Pape, a notoriously hard to spell wine. Imagine my surprise when the system actually came up with a result. Here’s what it provided me:
Tasting note: Loved it !
I don’t think I could have invented a more perfect illustration of my point. While this user had, through sheer luck, or by copying the text right off the bottle, managed to spell Chateauneuf-du-Pape correctly, the lack of a producer name (not to mention country designation) and the extremely detailed tasting note make for a deadly combination.
Even when users do manage to get the full name of the wine right, figure out the appellation if it’s not part of the name of the wine, and list the right country of origin correct, very few people can actually write a tasting note that means something to someone else. Of course, we could get into a discussion here (and a perfectly valid one) about the value of tasting notes in general, but hopefully you’ll humor me and admit that if written well they can be a lot more helpful in evaluating a wine than nothing at all. But sadly, very few users of these systems can actually write. I’m not even going to get into whether their palates are reliable.
One of the ways of getting around this problem is to seed the site with a relatively “complete” wine database, which you force users to select their wine from before they can enter a tasting note (along with a protocol for adding wines to the database if the wine they are trying to make notes about is actually not yet in there).
But even a good wine database can’t fix the biggest problem.
#3: There is not enough incentive or reason to use the system regularly
With any online community site that relies on user generated content. the usefulness of the system is directly correlated to how much content it has in it and how regularly it gets used. The more people who are using it (that is, both reading and contributing content) on a frequent basis, the more valuable it becomes.
The problem with these wine tasting note sites is that people just won’t contribute enough. Like all community sites that provide people an opportunity to build an identity for themselves online, there will be a few users at each site who are the core of the community. These people have their egos (and I don’t mean that in a pejorative way) wrapped up in their use of the site. They get pleasure out of being the resident experts, and will outperform all other users by a factor of ten. Some of these folks may enter hundreds of tasting notes for their wines, and actively add comments to new wines that other users put into the system.
The problem is, there just aren’t enough of these users. For these core super users, the building of their identity is enough of an excuse to come every day, or a couple of times per week to add new tasting notes, but for the majority of the population, they just can’t be bothered to do it that often, no matter how much they agree with the principles behind the site, and no matter how much they understand that their contribution increases the quality of the experience for everyone.
Amazon.Com book reviews are a case in point. While there are those users who clearly spend a lot of time going to the site for the express purpose of entering reviews on books they have read (again, to build their own identity online) the vast majority of us submit reviews only in the process of shopping for books.
Let me restate that. We most often enter reviews only because we happened to have gone to the site for a completely different (and I might add, more important) reason: to buy something. Luckily for Amazon, a heck of a lot of people buy books from them frequently, so there are lots of occasions for people to go on and write reviews of books they have read.
None of these wine sites really provide any other incentive to visit the site other than to read or write tasting notes, and frankly there just aren’t enough people with the desire or the discipline to do that regularly to ever amass the database of tasting notes and scores that could come anywhere close to achieving the admirable goals that inspired these sites to begin with.
#4: There just aren’t enough wine lovers to go around
It kills me to write this, but it’s true. There just aren’t enough wine lovers out there that are competent and interested enough to make a system like this work. Nearly everyone reads books, and so Amazon can have a pretty decent amount of coverage of customer reviews for their vast inventory of books, but there is nothing like that population in the wine world. If you take all the English speaking (as these sites are only in English) wine lovers in the world; then subtract out the ones who wouldn’t know a tasting note if it bit them in the ass (the ones who are responsible for Beringer and Sutter Home White Zinfandels being the most consumed restaurant wine in the US); then you subtract out the ones who don’t have any interest in doing “wine related” things online; and then subtract out the ones who already have their hands full participating in one of the two or three major wine forums out there, reading their wine magazines, maybe even (gasp) reading wine blogs, and then finding time to actually drink wine, you are left with an unfortunately and pitifully small group of users. This group grows larger all the time, no doubt, but remains small nonetheless.
WHAT ABOUT CELLAR MANAGEMENT?
In that list I just outlined of things I that wine lovers are doing online, I conspicuously left out one thing. This will finally mollify those of you who have been shouting at your computer for the last few paragraphs, “yeah, but what about CellarTracker!?”
Now that is an entirely different story.
For those of you who don’t know, CellarTracker is a free online web application that is designed to help users manager their wine cellars. It provides basic inventory management, cost evaluation, integration with sources of official reviews (so that you can look at one of your wines and see what the Parker-Tator said about it), and drink date recommendations.
It also happens to have one of the most authoritative databases of community created tasting notes in the world. There are a few other free cellar management / community sites out there, like BottleCount but they don’t have the depth of content that cellar tracker
So how does Cellar Tracker manage to get around the problems I outlined above? Well it doesn’t entirely, but it has managed to not fall prey to enough of them to minimize the effects of the ones it can’t control.
First, cellar tracker has precisely the sort of authoritative wine database I talked about. You can’t enter a wine into your collection, let alone write a testing note for it without they system first checking to make sure that the wine doesn’t already exist in the extensive database that runs the site. There is a mechanism for users to correct bad listings, as well as a mechanism to get new wines, new wine regions, and new varietals added to the system, all of which go a long way towards preventing the 1995 Chateauneuf-du-Pape problem. Also because this is a commercial wine database, CellarTracker isn’t dependent upon users entering content for wines to be in their system. Their hit rate for wines that I randomly type into their search field is nearly the inverse of the community tasting note sites: only one or two out of twenty come up empty handed.
Secondly, cellar tracker provides an excuse for people to return to the site frequently. After all, folks are using this site to manage their cellars. The main users of the site are moderate to rabid (isn’t that the spectrum?) wine collectors, who are, with reasonably frequency, adding to and subtracting bottles from their cellar. And this means that they’re coming back to the site to change the inventory of their cellar with regularity. The more they drink and the more they buy, the more they’re online adding and subtracting wines from their inventory, which, just like buying a book at Amazon, provides them the perfect opportunity to take a couple of minutes to enter their tasting notes and suggested drink dates for their wines. Heck, I know one wine lover who has a computer in his kitchen and most of the times I’ve gone over for dinner there’s a web browser sitting open on the desktop with his cellar inventory in it.
Finally, I don’t know how many users cellar tracker has, but it’s been around for some time (6 years?) and has grown slowly to the point where it seems to have enough users, and active ones at that, to actually provide a useful set of data about wine that might have a prayer of being an alternative to (or at least a useful augmentation of) the Parker-Tator’s web sites.
As a purely editorial side note, CellarTracker would be a lot MORE successful if they let someone redesign what must be one of the most godawful web application interfaces I have ever seen. The thing simply stinks, and for this reason only I do not use the service. Did I mention that the UI sucks?
Great. Glad to have gotten that off my chest. Now back to the essay.
Well, that’s a good question. I didn’t just spend the last two hours of this flight I’m on to Chicago to be a wet blanket on the idealistic wine lovers who have spent the effort to build these applications. I’m sure they’re good folks with good intentions and I would love to see them succeed, but I just don’t think they will the way they are going.
So here’s what they need to do: solve three of these four problems.
There’s nothing we can do about the last one, short of sabotaging other online wine destinations and forcing users to relocate their online wine activities to somewhere new. Heck, online wine communities (actually ALL online communities) have a way of imploding one their own anyway, so it may just be a matter of time. But as for the other three problems, the first two are imminently solvable and the third, well, the third may require some changes to the business model.
If my traffic numbers, and in particular the number of referrals I get from Google searches on specific individual wines are any indication there’s definitely a population of folks out there who would love to get more information about the wine they’re drinking online, it’s just a matter of figuring out how to harness their desire to build something that all of us can benefit from. The answer, however, is NOT “just like del.icio.us, but for wine!”