Last week, I contacted Tony Conrad for an interview about Sphere, a new blog search engine presently in private beta testing. Tony, who is the CEO of Sphere was gracious enough to agree, amidst a very busy schedule. I’ve been lucky to beta test Sphere and think Tony and his team are off to a great start. As usual, the interview was conducted via email. Enjoy.
Tony, tell me a little bit about your professional experience to this point. What were some of your past projects?
Prior to co-founding Sphere, I was a general partner in an early-stage venture capital fund where he led consumer-tech and marketing software and services investments. I served on the boards of directors for Oddpost (acquired by Yahoo), Iconoculture, MusicNow (acquired by Circuit City), and Centive. I also played an active role managing investments in Post Communications (NASDAQ: NTVS) and Stoneyfield Farms (acquired by Groupe Danone). My roots are in consumer marketing. I have worked with early-stage-growth and public-branded companies. I was director of mergers and acquisitions for Groupe Danone, a leading $30+ billion global consumer packaged goods firm, where I focused on brand extension opportunities through mergers and acquisitions in Southeast Asia. I also held several senior marketing management positions with Groupe Danone, the last of which was brand manager of new product development for Gervais-Danone France.
Describe your involvement in Sphere – do you work more on the technology or business side?
CEO. My principal job is to create an environment in which my two cofounders can focus on building great technology.
Considering all the other blog search engines out there, why Sphere? What is different about your approach? What makes it better?
We thought we could build a much better blog search engine, one that would find relevant content to match a blog reader’s keyword search query. It’s a very simple idea but really hard to do. The blogosphere is growing incredibly fast. When we started, there were around 5 million blogs. Nine months later, there is more than 18 million blogs. With an increasing number of people reading, writing, and commenting on blogs, finding relevant content has become increasingly difficult. For a variety of complex technical reasons, current blog search services deliver less-than-satisfying results. These reasons include an exclusive emphasis on freshness, or a too simplistic computation of a blogger’s authority. Our algorithm sorts through all those blogs, super fast, and finds relevant content to match a blog search query.
With those thoughts in mind, what’s your Sphere ‘elevator pitch’?
It starts with relevant results and fast performance. Our new relevance-based algorithm discovers new blog posts as they’re created, indexes them within minutes of being published, applies rich semantic analysis and makes them searchable by relevance or time. Plus, we’ve got a few fun, helpful features that we think make for a richer user experience.
How is blog search important to those outside of the blogosphere? Will Sphere attempt to reach out to those types of users or have any features that will cater to someone not familiar with blogs or blog search?
If you’re not interested in blog content, we’re probably not the best service for your needs.
I noticed on your blog and on the Sphere about page, you listed what you consider two types of users for Sphere. The second type being “publishers who would like to integrate high quality blog content into their websites.” Can you provide a little more detail about how this second case would work? Is that going to cost money?
Throughout our development, we’ve been in constant dialogue with a small group of highly respected publishers about their plans to integrate user generated blog content into their website properties. Our conversations lead us to believe there is an interesting market opportunity servicing the needs of publishers. Specifically, we will syndicate contextually relevant information and search capabilities to publishers, vertical portals, complementary search engines, and weblog hosting companies. Integrating relevant blog content into publisher website properties will result in increased traffic, more ad views, and higher ad revenue.
How is Sphere addressing technical concerns such fake blogs, splogs, and scalability issues?
Sphere’s relevance/authority algorithm rewards authentic blogs and penalizes splogs, fake blogs (flogs?) etc. by taking into a account a wide range of data points (metadata, link patterns, posting patterns) that we gather about a blog. Fake and spam blogs are a big challenge for any search engine that indexes them, because they’re so damaging to the user’s experience, but overly aggressive approaches can led to false positives. Our algorithmic approach has yielded great results so far, and improvements to our algorithm are ongoing. Re: scalability, our technology has been deployed in high-volume settings for several years and we’re positioned for a smooth transition to a scalable service after our beta period.
What was the biggest challenge you faced with Sphere to this point?
Securing the name! Creating our user interface was a difficult task, lots of tradeoffs.
Any rough timeline for when Sphere might show its public face?
Two areas need to be addressed prior to a public market launch: 1) we need the input on the quality of our website and search results from our beta users; and 2) we need to purchase servers to scale the number of users on the site.