Gary has an analogy that he’s started using when people ask how we’re different from other local online plays (past present and future) with national ambitions.That’s a key question for us, because not (m)any of those have really been a hit. The answer goes something like this:
It’s like in the game Risk. There are two basic strategies that people play by: Some people spread their armies all over the map. Others set all their pieces in Australia, control it, and spread out from there. We’re Australia people.
Parker Brothers’ own strategy guide says:
Concentrate your force to attack. The best way to attack is to stack as
many pieces as possible into a single space and then roll, roll, roll
onward. Take a territory, move into it with everything you can, and
start a new invasion from that space. Keep doing that until you’ve
achieved your objective or run out of steam.
I also found an interesting ancillary bit of strategy:
Because of that, it’s not worth taking a continent if you can’t hold
it. You’ll only invite vicious counterattacks. It’s better to build up
for a while and then grab the last few territories with enough strength
to hold them.
As a rule, people who try to "scale local" must have been horrible Risk players as children. Yeah, it doesn’t cost much — only one army — to plant a flag in twenty, a hundred, a thousand markets. And that low-Risk strategy wastes an army (but only one!) when a better-heeled competitor drops a scant four armies in your market.
There are a lot of great localweb ideas that have underperformed because of this strategy. Digital Cities. ServiceMagic. Angie’s List. CitySearch.
Now, comes citizens journalism and local search. And from what I see, the primary strategy is still to create an engine and put it everywhere from Greenland to Kamchatka.
Why doesn’t the flag-planting strategy work?
It’s about the technology and aggregation. So much, particularly in the latest iteration of web technolgies, is achieved by a computer aggregating things that are already there. See Topix. See Bloglines. See Google. Aggregation doesn’t work so well on local though. First, because it is constrained by what there is to aggregate. And local news as we know it comes from relatively few sources. Increasingly, hyperlocal news just isn’t there. And local search? Not yet. So there’s a relevance problem. That could be ameliorated by widening the net, but unfortunately most web sites don’t tag with a Zip+4.
So this is what aggregation gets you in local: Above is a screen capture of the breaking news box from Citysearch Dallas this morning.
For the record, I’m a big fan of Topix, and we think that we’ll have a lot of Topix links on our site, for instance to get our users to aggregated national information on local companies. But aggregating local is tough because…- It should be about original unique content. Local news and info is at the farthest reaches of The Long Tail. It is self-limiting because of the realities of geography. There may be enough fans of Jandek to create a vibrant online community. But there are probably not enough fans of Jandek in his native Houston to do the same. That means that professional resources have to be deployed to generate the content that brings the audience in reliably.
- It leaves a lot of "incompletes." To use a site for shopping reviews, I’ve got to know that every result that comes up will have a review. Otherwise, it’s just hit-or-miss and I might as well go to Google.
- It doesn’t put feet on the street, at least not in a meaningful way. At most, the flag planters put a couple people in a city that isn’t their original stronghold. It takes multiple armies to solve problem one and even more to reap the rewards of that effort (by selling ads).
- When the niche is local, all other sub-niches must be seamless to the user. I’m sure that somewhere in a file drawer at IAC is a business plan that tells a story in which a guy decides to throw a party celebrating a home remodel planned at ServiceMagic; creates invitations at Evite; then researches potential locations on CitySearch; making the final call based on a coupon in The Entertainment Book; and finding a date for the shindig on Match.com. It may have recently been edited to show the whole chain of events starting with an AskJeeves search.
The local world doesn’t work that way. You don’t pick a national brand to help you with your local problem. (Or if you do, it’s because, like a Home Depot, it’s physically located near you.) The biggest mistake that newspapers have made in the past ten years is that they haven’t each become their market’s own ServiceMagic. Because they have the advantage in the area of…
- You can’t be static. Simply shuffling lead homepage reviews or user posts around or being a search-only mechanism guarantees that people will only visit you when they’re actively looking for something, as opposed to as a daily (or hourly) habit.
There’s a final point that (so far) only relates to the big corporate sites: There’s a certain sameness and sterility built in (probably to enable "scaling) that screams to the user: "This is the coolest slickest system we could deliver you without actually taking the trouble to put people in your community to understand what it’s all about and the difference between the cool stuff and the not-cool stuff." It’s like the beverage machine in The Hitchhikers Guide to the Galaxy:
When the Drink button was pressed it made an instant but highly detailed examination of the subject’s taste buds, a spectroscopic analysis of the subject’s metabolism and then sent tiny experimental signals down the neural pathways to the taste centers of the subject’s brain to see what was likely to go down well… [I]t invariably delivered a cupful of liquid that was almost, but not quite, entirely unlike tea.
And let’s not forget that the makers of said machine, the Sirius Cybernetics Corporation, were "a bunch of mindless jerks who were the first against the wall when the revolution came."
Re-reading this overcaffeinated Sunday-morning rant, it strikes me that it’s another one of those posts whose sense can be lost in the fact that we don’t yet have a product out that answers these criticisms. In a way, though, it’s the explanation of why we don’t. We believe that we have to take Australia before we fan out o’er the world, and that throwing one army in isn’t enough. We’re looking for "armies" to deploy in building such a product. While we certainly will keep scale in mind while we deploy those armies, their on-the-ground mandate will be to help our users kick ass. But first, we need armies. And while they ain’t cheap, they cost a lot less than the alternatives.
