Our new location is blog.oodle.com.
It’s been a few months since my last post. Apologies to everyone that reads this blog. I'm back in the saddle and resume posting on a more regular basis.
Much has happened in the last few months. Probably the biggest news is the launch of Oodle UK. Why the UK? It’s probably one of the most dynamic markets for online classifieds. The major media companies all compete aggressively online with vertical classified sites. These companies are also investing in sites targeting niche verticals (probably the fasting growing segment in the US). The free classifieds model has also been around for sometime and is definitely not a one-man-show (Loot, ExchangeAndMart, Gumtree, etc.).
We've also been busy with with various partners. We just launched our first UK partnership with The Sun, the most popular daily newspaper in the UK. We’re powering the search for their classifieds marketplace. About a month ago, we launched a similar deal with the Washington Post's free daily newspapers Express.
Partners hav also been rolling out more implementations of the Oodle API such as The Examiner. As always, anyone interested in looking at the API or classifieds search for their site should contact me.
We’ve been very happy to see the prominence of search in the classifieds market grow so quickly. We believe that search is a vital element to making online classifieds a great medium for both buyers as well as classified advertisers. Another search engine for classifieds launched yesterday. We welcome Vast to the crusade.
I have received a bunch of questions in the last 24 hours. Thought I'd post some of my replies here:
How many sites does Oodle reference in it’s index?
We index both ends of the long tail and point to over 50k sites.
How many listings does Oodle index?
In our current coverage area (just over 100 metros), we have 10M active listings. This only includes listings that are fresh and relevant: we keep track of all the listings we’ve seen and auto-expire old ones that are still online and exclude things that look like listings but aren't (reviews, spam, etc.). This number also doesn’t include listings currently in our index but outside of our coverage area. In next month, we’ll be rolling out millions more listings as we expand our footprint.
How old is your index?
Our index is refreshed every few minutes. From the time we first see a listing online, clean/tag/rank it, and push it out to our index is usually under 5 minutes (and we’re working hard to reduce that number).
It’s essential for a classifieds search engine index to be fresh. Unlike auctions, where you want to be the last person to respond, with classifieds you want to be the first. That’s also why we offer alerts, so people can be notified in minutes when something they are looking for becomes available.
How does your crawling technology work?
We use a hybrid approach. Some crawlers look for sites to crawl. We also have a farm of "micro-spiders" that target collections of similar sites. We also get data feeds from a large number of partners.
When will you offer an API?
This is mentioned in my previous post. We have an API that we've deployed with a few partners. It will be publicly available in the near future. If you're interested in using it in the meantime, email us at firstname.lastname@example.org.
We also offer RSS which has always been available.
Using the API benefits a marketplace by helping their users find what they're looking for -- both from listings posted directly on their site or from listings available in other local marketplaces. And by more broadly promoting Oodle search through this API, we continue to help the partners in our index who gain more brand exposure and more traffic. As always, Oodle search results link directly to the source and are attributed as such.
We plan to make the API publicly available but we're currently only working with interested partners. We still need to shake out the implementation (with respect to features, level of abstraction, etc.). If you have comments or ideas on what you'd like to see, please let me know.
Classified publishing seem to be popping up all over the place. We’ve recently seen them surface in a number of online communities – both small communities such as Boof as well as big ones like MySpace and sometime soon Facebook (if the rumors are true).
We're also expecting classified listings pop up in blogs. A few people (myself, Rohit Khare, Assaf Arikin) recently proposed a microformat called hListing that would faciliate this. Microformats are a set of simple open data format standards used to implement structured blogging or web microcontent publishing. Part of the idea behind microformats is to “pave cowpaths” – i.e., “standardize and codify emergent, popular behavior on the Web.” Given all the activity swirling around online classifieds over the last few months, seemed like a good time get the conversation rolling on something like hListing.
With hListing, classified listings can published in a way that's easy for search engines to reference them -- whether the listings are published in a marketplace platform that supports microformats or a blog. To give people a sense for how this might work in a blog, Assaf created a WordPress plug-in for the draft spec. This is clearly pre-beta stuff but it helps visualize where this could go…
Received an email alert from Classified Intelligence this weekend. Looks like GoogleBase is now delivering their users to jump pages from the results page instead of to the listing on the source site. This Google hosted/created jump page displays all the attribute information and it's striking how much thislooks like a listing.
It will be interesting to see how this is received by the folks that provide GoogleBase with a feed. Oodle and other vertical search engines (SimplyHired, Indeed) made an explicit decision not to do this. We follow traditional search engine etiquette and send traffic directly to the listing from the results page.
This isn't the first time Google has pushed the envelope. As I mentioned in a previous post, I was surprised when they decided both to point off to third-party listings and take their own (putting them in competition with their listings partners).
The Internet continues to churn our more and more choice for consumers. At the same time, it’s getting harder and harder for consumers to find what they are looking for. New techniques that seem be doing the most to address this issue provide more context to the discovery process. Here are three that jump to mind:
1. Vertical Scope. Vertical search engines take advantage of a narrow context (e.g., looking for an apartment in San Francisco) to offer more relevant functionality to the problem at hand. And when done well, they should look and feel more like an application anchored in search than what we think of as a stand-alone search engine. Shopping.com does this for online commerce. Kayak is doing this for travel. Oodle is doing this for classifieds.
2. Tagging. Tagging allows users to provide context either for themselves or for others. Obviously del.icio.us provides such a filter for finding web sites. Reviews are a form of tagging, and as such, InsiderPages provides a good filter for business directories. SPAM reporting is also form of tagging, and is very successfully used in the classifieds world by Craigslist to keep its market from getting clogged with SPAM and porn.
3. Social Networks. Social networks allow context to be established based on the people behind the tags (versus the content in the tag) whether it be people like me or people I know. As such, social network needs to appropriately align their context with context of the tags that they support. LinkedIn provides such social context to its tags which are online resumes & contact information. Fremont will soon introduce social context to classifieds market, recognizing that the person behind that roommate listing is a important piece of selection criteria.
Am I missing any? Let me know.
In a previous post, I poked fun at all the Web 2.0 hype and jargon. I do think, however, that we have entered a new and distinctly different phase of the Internet. IMHO, it’s being driven by the fact that the primary business model for most sites on the web has changed. And it’s this change that is aligning the economic interest of sites to do things like support mashups.
Web 1.0 was dominated by media driven business model – i.e., deliver ads to a captive audience. The goal was to capture users, lock them down (in a “walled garden”). To do this, businesses spent tons of money negotiating multi-million dollar distribution deals with portals (walled garden versions of a search engine) and developing their brand with display advertising.
Web 2.0 is dominated by a lead referral model – i.e., delivering prospects to your advertisers. The goal in is to entice users to your site. To do this, businesses are becoming more open and interconnected within their ecosystem. They are syndicating their content, opening APIs, offering affiliate programs and embracing search. They succeed when they find ways to bring more prospects to their site. As I previously posted, it’s no longer about lock-in, it’s about “love-in”.
Most ecommerce sites are far down this path. Classified publishers are just starting.
It looks like MSN will soon release a free classifieds service with some social twists. This idea has a lot of promise. Classifieds are most definitely a social phenomenon in that they typically result in an offline, face-to-face interaction with another human. Also, sometimes consumers want to trade in smaller communities where they know or have some sort of connection with other people (e.g., on college campuses, on company bulletin boards).
I'm not surprised that it's free. As I've mentioned in previous posts, the market price for a basic classified listings has already been established as free. Any consumer can post a free listing in Craigslist and now a growing list of sites including GoogleBase, Backpage, Geebo, LiveDeal, etc. They can also post in on their website or in their blog and pay Google, Yahoo to advertise it.
I'm most interested to see what kind of biz model they employ or at least what model they hint at in this initial release. I'm also excited about adding another quality source to our index ;)
I guess the old adage is true -- no press is bad press. We have been making great progress with the build out of our classifieds index in part thanks to the increased visibility we gained as a result of this issue.
We now have about 6M active listings in the index (which spans 50+ metro areas in the US). This brings us above the listing account we had about a month ago when we indexed Craiglist. And there are a lot more listings out there. I suspect we'll double our index in the next few months.
We still miss Craigslist. Unfortunately, I don't have much to report in with respect to getting them back. I'm still hoping to get a chance to talk to them. I have lots of ideas about how to address concerns that they've raised in a few press articles.
I’m sorry that I haven’t posted in about a month. In the last few weeks, my role of CEO and father have seriously cut into my blogging time.
Now that I have had a chance to play with GoogleBase it's safe to say that it's definitely more than a classifieds platform. It will be interesting to see all the different ways this product evolves. A simple to use, hosted database in the sky is a very cool thing. I was a big fan of Quickbase, a similar service Intuit pioneered about five years ago.
That said, that GoogleBase -- or more precisely the classifieds solution that will be built on GoogleBase -- will definitely rock the world of classifieds. GoogleBase is simply the first of several platforms that Google may tie together to provide a classifieds solution and perhaps the least interesting. Two other future platforms, which have been written about in the blogosphere, will be much more interesting: Google Automat (which enables a CPC like model for listings published in GoogleBase) and Google Purchase (which could enable Google to participate or at least track a transaction between two parties).
Google’s moves are significant in that they will push the market to adopt a new revenue model for classifieds. This model is poised to change from a classic media model (where the advertiser pays an upfront fee to reach a captive audience) to a performance-based model (where the publisher gets compensated for delivering prospects to the advertiser).
This shift has occurred with all other forms of online advertising but it hasn't yet happened with online classifieds. Why? First, none of the existing players -- newspapers, Yahoo, vertical sites like Cars.com -- want it to change (perhaps a rational decision on their part). Second, and more significantly, no renegade company to date has forced them to change.
I used to think this would be Craigslist. Craigslist certainly got the ball rolling by finding a way to make the free classifieds model work (through community moderation). But it has failed to act as a catalyst to drive industry wide change from a business model perspective. I think this is true for two reasons. First, many of the existing players outside of the Bay Area still aren’t getting hurt badly enough by Craigslist. Craigslist mainly targets private party ads and most existing classifieds publishers make their money through listings generated by small businesses (car dealers, real estate agents, property managers, hiring managers, etc.). Second, Craigslist -- when it chooses to charge – still employs a classic media model.
Google will act as the forcing function for this change. It doesn't have an existing classifieds business to protect and it's demonstrated a clear focus on developing the long tail of advertisers (classifieds advertisers being the right end of that tail). Having said that, I certainly don't think this is game over for existing players. It's just finally time for them to change their model.
Today, a few articles have surfaced that Google will enter the classifieds business. Very cool looking product. We’ll definitely include Google’s listings in our index (unless they stop us which I very much doubt) . The more sources the better.
Based on previous reports, it now appears that Google is pursuing a dual strategy with respect to classifieds. They will both a) become a classifieds publisher by directly taking listings and b) incorporate classified listings published by other sources into Google Search, Google Local and Froogle. This dual position – as both a publisher and a search engine – is likely to spark a big debate.
Search should be a win/win game. Publishers get free traffic (their advertisers in turn get more prospects) and consumers get a better user experience. When a search engine sells their own listings, they begin to hurt their partners by competing against their core business. For this reason, Oodle doesn’t take listings. It will be interesting to see this play out. If nothing else, Google is certainly exhibiting portal-like behavior. Stay tuned, I'll be tracking the debate...
We’re very excited about where things are going. Oodle provides a valuable service to consumers by enabling them to see all their choices. We’ve seen over 25% growth in users and traffic to partners every month since we launched. Last month alone we sent Craigslist well over 1 million referrals (i.e., users sent to their site). We also continue to have the largest classifieds index (Craigslist represented about 20%) and will continue to add millions of more listings to our index in the coming months. This week, we received over a dozen requests from classified sites wishing to be included.
This whole Craigslist thing has unfortunately mushroomed into quite a big deal. They certainly have every right not to be in our index and I’m genuinely sorry that we appear to have upset them. We continue to be great fans.
In a recent news article, Craigslist CEO Jim Buckmaster made a couple of comments about our actions. It’s never good to see these sorts of things play out in a public debate but we wanted to provide our point of view.
We have always been up front with Craigslist. Before we launched, we sent them a note asking them to check out our site. A month later we requested a meeting to discuss what we were doing and how we were accessing their site. Both times they politely declined to meet with us.
We do our best to act as a responsible search engine. We strictly follow the robots.txt protocol and check it every time we visit a site. For those sites that don’t want to be included, they simply need to add a single line of text to their robots.txt file (as of today, we are still not in Craigslist’s robot.txt file).
After Jim let us know that Craigslist wished to be removed from Oodle, we complied within two business days. During this time, we tried to establish a dialog to address their concerns and are still interested in doing so.
We always try to be very mindful in the way we index a site. We go to great lengths to minimize our impact on our partners’ servers (and visit most sites in the middle of the night when the cost to crawl is near zero). To put it in perspective, Google and Yahoo pull about three times the data from Oodle's servers each day than we pulled from Craiglist. It’s helpful to know the issue at hand is the load we place on their servers. We have a number of proposals on how to address their concerns. I hope we get a chance to share them.
I wanted to briefly respond to a few of the comments I've read on this issue...
I think it’s important to keep in mind what's best for consumers. And I think being open is good for consumers.
People tell us they want to see all their options. This is especially important for classifieds:
If I'm looking for an apartment and don't see the listing for my dream place, it's gone and I never even knew about it. If I am looking for a used Volvo, I want to see all the listings available as soon as possible to maximize the potential for finding what I want. If I collect rare coins, I want to be alerted on listings across as many places to buy as possible.
Comparatively, this is much less of an issue when I'm shopping for a new digital camera (where I can go into any one of ten stores and buy what I want). In this situation, search still has value to empower users in price comparison -- causing retailers to further differentiate in service, convenience and support (all benefits to users as well).
A similar argument can be made for consumers trying to sell items. They want to reach the biggest audience so they can get the best deal. Services like Oodle, that help bring prospects to a consumer's listing, are helping them not hurting them.
Craigslist certainly has a right to not be included in Oodle. That said, we believe that by helping consumers find listings on Craigslist, we positively contribute to their service.
We are no longer adding new Craigslist listings to our index. We received a request from them to do so and we honored that request. We’ll miss Craigslist. They are clearly a good service and we're big fans. We are working to understand and address their concerns.
As of last month, about 80% of our listings came from sources other than Craigslist. Without them, we still have about 4.5M active classified listings and that number will continue to climb by the millions in the coming months.
We strongly believe that search plays an important role in any Internet ecosystem, especially ones involving commerce. We’re glad to have had a good response from the 99.9% of the other classified publishers in our index and we’ll continue to work hard to make them happy. We send them free traffic. We don’t compete with them by taking listings. And we'll continue to work hard to make classifieds a better online medium for consumers trying to buy & rent things.
At the end of the day -- with our without Oodle -- consumers check multiple sources when using classifieds. Oodle makes this process less painful and helps consumers to be more successful. And if you're in the business of publishing classifieds, this is a good thing.
preparation for this week's show,
I thought it would be fun to describe Oodle using Web2.0 buzzwords…
Oodle indexes classifieds listings from across the web, many of them without even a URL to call their own. Once we find them, we tag them (e.g., pet listings with breed, gender and color); remix them with other data sources (e.g, expand abbreviations, normalize location); and create mashups with other web services (e.g, put them on Google Maps using AJAX).
Perhaps by next year we’ll even incorporate an architecture of participation ;)
I've received a few pings
on the news that Google may be launching a classifieds product as well as
entering the "classifieds search" game. Here are a couple
of quick comments…
I'm clearly concerned. As an entrepreneur, it pays to be paranoid about a lot of things. Google making a move into classifieds is something we need to watch closely. Moreover, I have tremendous respect for Google. They build great products.
Google’s entry into classifieds search is also not surprising. Google is attempting to index all of the world’s information. Online classifieds, with close to a $20B market, should be on their roadmap.
Google directly taking and publishing classified listings would be much more disquieting. As I mentioned in a previous post, I think this is a dangerous line to cross. Classifieds search can be a win/win, where publishers get free traffic and buyers have a better experience (also helpful to those selling classifieds). Once a search site begins to sell/publish listings, however, they hurt the partners in their index by attempting to steal their business.
At the end of the day, I believe Google and Oodle are on different paths. I strongly suspect that Google is approaching classifieds from a platform perspective. Google is a search company; online classifieds are another content source they can integrate into their search platform. Oodle is a classifieds company – search being one element of our value proposition. Our success rests on our ability to better focus on the whole online classifieds problem.
When we first launched, I read a few blog post that contemplated whether or not what we were doing is legal. Not only do I firmly believe that it's legal, I think the debate is somewhat moot. We hold ourselves to a much higher standard.
As a search engine, indexing the content on other sites is a privelege not a right. Publishers should clearly see value in the relationship. If they don't, they can easily ask search engines (all or a selected few that are ill-behaved) to stay away. doing this is as simple as adding a single file to their web site -- robots.txt.
As I've stated in other posts, web publishers (including commerce sites, etc.) increasingly view search engines as playing an important role in their market ecology. To play this role, however, i think companies need to follow certain unwritten rules (or net-etiquette):
1. Summarize and point, don't aggregate
I've seen oodle referred to as a classifieds aggregator. We are not an aggregator in that we don't display listings. Our job is to point our users to listings published on other sites -- and in doing so, get users off our site as quickly as possible ( love-in not lock-in).
To this end, we've provide (attributed) summaries that contain enough information for users to qualify their interest but not enough for them to take action. To respond to a listing, users need to click-through to the full listings on the publishers site.
2. Be a search engine not a publisher (or said another way, don't compete with your content partners)
It's a fundamental conflict of interest when a company attempts to both play the role of search engine and play the role of a publisher. Why would publishers let their listings be indexed by a company building a competing business off their back? It will be interesting to see how this point plays out in the coming weeks. Rumors are ablaze with Google's entry into the classified listings business. It's also rumored to be adding classifieds listings into Froogle. Will classified publishers stand for this? I'm not sure why they would.