Tuesday, April 20, 2010

Writing up thoughts on an entrepreneurial failure

Applying for the VC analyst job has been bringing back memories of my own little startup: FSBOAustinTX.com. USV's operating thesis is that the internet is a paradigm changer in every facet of life. I agree. Take real estate, for example: as Americans -- and then the world -- fully migrate to the internet, real estate agents will probably become an endangered species. That was my epiphany in 2004 while doing real estate school online. So I decided to try and make it happen.

My landlord in Austin was from Madison, WI, and he suggested that we start an Austin version of fsbomadison.com. And so was born FSBOAustinTX.com. fsboMadison is a site that had started to sell the owner's house for sale by owner (FSBO) and then just kept growing organically because it had traffic, cost $200 and could save buyers and sellers thousands. It made six figures on Madison alone with a website that even its developer was ashamed of (he couldn't convince the owners to upgrade). The owners liked the idea of people using their idea, and we figured Austin was as similar to Madison as possible: state capitol, university town, crunchy liberals, high internet usage, etc. Downside was that some nationwide fsbo sites had started and there was already an attempt at a local FSBO site.

After we did a little bit of research, we weren't too concerned about the competition. The nationwide sites were non-starters who had no local presence and would never get the search engine traffic. The local site was non impressive.

We knew we'd have a problem with unrealistic prices. For reasons I do not understand, most FSBO "sellers" set the price way above market value. Here's an example of madeup but realistic numbers: the Smiths have a $200,000 house, which they try to sell FSBO at $220,000. After a couple months where it doesn't sell (duh!), they put it on the market for $205,000, and it eventually sells for $200,000. The owner and seller in this case each pay $6000 (3%) to real estate agents, making the effective house price 206,000, but the seller receives only 194,000. So they tried to sell FSBO at $26k more than they would actually take, when they might have been able to sell the house for $205k, made 11k more while the buyer would have paid $1k less. Argh...so irrational!

That is the background, here is my quick list of successes and failures.

- Search engine rankings in competitive keywords. I taught myself search engine optimization and got the site into the first page of 'austin real estate' within 6 months, while ranking 1st in the niche keyword terms in the meantime. The frustrating part of this is that my business partner did not value this at all. I, on the other hand, view search engine rankings as absolutely vital for a business like this.

- The signs. One of the big reasons people still pay money to real estate agents is a recognizable sign. We made big, recognizable awesome signs and put them in yards. Getting brand recognition through your signs is almost as important as being able to get good search engine traffic.

-Launching. Hey, we dealt with our developer and got the sucker built. We pulled off some press (if I recall correctly), and at one point had a pretty full site in terms of listings. If it worked, it was very scaleable and we kept costs very low.

- A real estate agent tried to get us shut down for violating the law. He lost. My research kept us on 100% safe legal ground.


- Not waiting enough time. In truth, this happened largely for personal reasons, because I broke up with the girl I was dating and wanted to get out of Austin. However, the next two reasons are why I didn't it wasn't just the girl...

- Team. Who you work with really matters. Executing poorly will ruin even the best strategy. As indicated above, my biz partner and I just didn't agree on some things that I never would have foreseen. Some of this was a fault of planning, certainly. But the team has to really be in synch in order to make it worth it.

- Never solved the overpricing problem. I still haven't really: I think the answer is just to establish the brand name, let the signs establish themselves, get some press and eventually let word of mouth take over. It is tough though, because in the beginning if homebuyers come to the site and see overpriced (and thus eventually stale) listings, they won't return. This is why it is so important to be hyper-local, with someone managing the signs and making sure the listings don't become stale in the local metro area.

- We were overanxious. To get listings on the site, we found existing FSBO sellers and let them list for free. But we still kept the list price of $200 on the site. That was a big mistake: we should have gambled on getting more users and more listings by offering it for free. Now, we got some advice from folks doing similar sites that this was the way to go, but I think we should have established the site as a big community resource and then started charging. After all, if we could have proved the site worked in getting a house in front of prospective buyers, $200 would be totally reasonable.


Today FSBOAustinTX.com has zero listings. The site isn't even as good as when I was doing it, and it is 5 or 6 years later. If I go to Texas for an MBA (my planned safe school), I might try and buy the site and give it another go.

Would I ever do this idea again? As I indicated above, yes. I still think the basic idea has merit. While in some ways Craigslist fills this niche, it isn't local enough. I still think that someone could do this nationwide, but with hyper-local sites in each market. And in fact, I think that eventually something like this will happen, but it might be a long time.

It was fun trying to launch and build something. I'll probably do it again in my life, although temperamentally I am probably more built to be a venture capitalist: I would much prefer to take small company risks in a bunch of different companies (although the Dot Com bubble burst shows that there is always systemic risk) than take one big massive company risk that is a startup.

No comments: