You are here:  Home  >  Data Education  >  Current Article

Advice for Semantic Web Startups: Embrace Evolution

By   /  September 27, 2010  /  No Comments

Photo credit: Flickr/kevindooley

The Semantic Web is so very much about adaptability – adaptability of data to serve new purposes, adaptability to top-down and bottom-up approaches, and delivery of a whole new Web space that users will adapt to, without necessarily even realizing the mechanics behind the change. Semantic Web entrepreneurs are discovering that adaptability matters to their own business models as well, when the field is still so green, many people still aren’t 100 percent sure about why the Semantic Web might matter to them, and potential big customers may be skeptical about the street cred of an emerging company in a space that may still feel a little blurry to them.

Take the case of startup Bueda, co-founded by CEO Vasco Pedro. It originally envisioned that its tagging technology would come in handy for scenarios such as helping content sites rich with video and images better monetize advertising opportunities around their user-generated content, as well as generally help publishers with support for improved recommendations and search accuracy. The idea got people’s interest, says Pedro, but also left them a little confused. At its matching engine API’s launch a few months back, “we had an interesting set of use cases,” Pedro says, but acknowledges it was too general and diffuse for users to easily grasp onto. There was a lot of input about how to enhance the API, but to what clearly understood end? “Unless there’s a very clear motive for using it people are just going to dip their toes in. So we had to eat our own dog food and come up with an application that uses the API.”

It faced other struggles as a small start-up, too: On the one hand, some content providers were reluctant to pay an unknown quantity to help them better target advertising for video and images. Other potential users were themselves small players and it was difficult for them to find the bandwidth to support integration with the API, while some other larger prospects had to go through a lot of internal layers before they could sign off on anything. And in other cases where there was interest, “we didn’t have the bandwidth to support the data they wanted to run through,” Vasco says. “We had to build the infrastructure and get to the point to how to tackle that.”

What’s A Start-Up To Do?

Probably by now other entrepreneurs are nodding their heads in sympathetic agreement. They may have faced similar problems in their own past – some are perhaps trying to solve such situations right now. How to solve it?

When Bueda went up against these factors, it came to the conclusion that it had to focus on finding more direct opportunities. Its thinking turned to the fact that there’s tremendous value in leveraging semantics with social media content because it’s so noisy and there’s so much of it. And small businesses, in particular, are having a hard time trying to realize marketing opportunities there because they usually don’t have enough resources and time to build social networks and make sense out of the feedback they can offer.

So, the company decided to adapt its own technology to new ends: Delivering a social media marketing service that uses its semantic analysis to help match marketers with the people who are most receptive and open to hearing their message. Thus, Five by Bueda was born, to serve as a platform for making social media promotion faster and easier, starting, of course, with the Twitter network. Basically, the idea is that Bueda will analyze what message an SMB wants to promote, and what goal it wants to reach, and match it up with the right people – to the tune of five relevant leads a day.

The service creates a semantic profile of what the SMB wants to promote using an initial set of tags Bueda provides, with the SMB user specifying their location and the semantic type of customer they want to attract. Or the service can choose for the SMB the types of individuals likely to be receptive to a particular SMB’s services by semantically analyzing its web site content and then determining who its relevant customers would be.

Matchmaking happens when those tags meet up with and disambiguate around the ones defining what people are talking about on the social web and who those people are. So SMBs targeting musicians won’t miss, for instance, the social networking enthusiast who defines himself as a bass player for a band (rather than a musician per se) who wants to buy a low-cost, high quality speaker – and will skip right by that social networking enthusiast who is talking about going bass fishing. Bueda is still experimenting with types – what to do with those who name themselves in their profiles as hipsters rather than by occupation or other well-understood definition?

hipster.png You can join the experiment at its www.hipsterboxing.com, where you’ll be shown two pictures of potential hipsters and you then click on the one you think is closest to actually being a hipster. “It’s fun but also we want to be able to get some data on some subjective terms and how to use our platform to analyze them,” says Pedro.

Users can respond right in the system to the leads generated with reply or follow messages, or pass if the person doesn’t seem relevant, and the system learns from that. A response message has a link embedded in it so that the service can help customers track whether the message is spreading and whether they are connecting with the right audience. “They should become your advocate, to propagate your message. We want to connect you with people who are already talking to you but you don’t know it yet,” Pedro says.

Smooth Sailing From Here?
So, problem solved right? This Semantic Web startup has tapped into the right vein, found its new audience (about 200 users signed up so far for the subscription service), and it’s all cake from now on.

Heck, no. Pedro says Reducing the amount of social promotion work an SMB has to do from hours to minutes – and enabling them to do it in mobile fashion, too – has people excited, he says. But then they realize that it still is a task they have to do every day, and like any other daily task that isn’t absolutely required to keep the business running, SMB users with a lot on their plates get tired of performing it after awhile, and use tapers off. “We haven’t really figured out a solution per se – maybe it. has to be something that has them use it for awhile and once the system has learned enough, then maybe we have to automate some part of it.” And, in addition to enriching and returning content, he’d like Bueda to be the engine that helps add content for SMBs – tell them where their potential customers hang out on the web, for instance, so they can go there to raise awareness of their brand. But there’s still a problem with data sparsity to enable that efficiently.

The service is starting to see traction with advertising agencies, whose job it is to help their clients reach customers – so spending a few minutes a day following through on leads for them isn’t a chore. There’s potential that the service could find a home as a more integrated tool within such environments.

But that’s the exciting part about being a startup in the Sem Web space – there are frustrations and dead ends, but also many opportunities are there, as long as you have the energy and wherewithal to follow up on them. Bueda is funded by Pennsylvania’s InnovationWorks, which invests in tech companies in Southwestern PA., and its funding sources include the state of Pennsylvania as part of efforts to drive its technology economy. “We’ve got to keep demonstrating progress,” says Pedro. “The next few months we’ll start raising the next round and the goal is to clearly demonstrate that the first step was the Semantic Web and now we are moving to the idea of integration of social and semantic information.”

Says Pedro, “We have to stay capital efficient until we find the business model we can scale up.”

About the author

Jennifer Zaino is a New York-based freelance writer specializing in business and technology journalism. She has been an executive editor at leading technology publications, including InformationWeek, where she spearheaded an award-winning news section, and Network Computing, where she helped develop online content strategies including review exclusives and analyst reports. Her freelance credentials include being a regular contributor of original content to The Semantic Web Blog; acting as a contributing writer to RFID Journal; and serving as executive editor at the Smart Architect Smart Enterprise Exchange group. Her work also has appeared in publications and on web sites including EdTech (K-12 and Higher Ed), Ingram Micro Channel Advisor, The CMO Site, and Federal Computer Week.

You might also like...

Property Graphs: The Swiss Army Knife of Data Modeling

Read More →