It was just over a year ago when I chatted with Hanan Lifshitz of Palore for the first time. We've spoken a number of times since then, and he and the Palore team have done some great things in providing local business data. I'm very happy that Hanan has agreed to do an online interview.

Paul: First of all Hanan, thanks for taking the time. It's been awhile since we have spoken. In a nutshell, what have you and Palore been up to and can you give the Search Engine Guide viewers an overview of your services?

Hanan: Paul, as always it's a pleasure speaking with you. Palore is a data provider in the local search space. We collect information about millions of SMB's from hundreds of local search sites, IYP's, vertical sites and news sites. After aggregating and normalizing this data, we syndicate it to companies that need access to large volumes of rich and accurate local business information, quickly and inexpensively.

Our data services range from simple collection of local business data that appears on government sites, to advanced analysis of ad spending by SMB's online. Needless to say, all the data is collected from sites that allow scraping through their robots.txt file and in line with US copyright laws. 

Paul: You provide a couple of different services. One of them is content for site enrichment and the other is content for sales leads. Can you explain the difference between the two?

Hanan: Initially, we provided a simple web data extraction service to companies that preferred to outsource this activity instead of dealing with the technical challenges in-house. Typically, these companies wanted to enrich their sites' listings with information such as opening hours, reviews and price, in order to improve their user experience and SEO.

In recent months we were approached by a number of SEM, SEO and IYP companies that wanted to use our web extraction capabilities to determine which local businesses are advertising online. This is how it works - as we process millions of local records, our system analyzes the different characteristics of each business listing, such as background color and logo size, to determine whether it is a free or paid listing. We take into account the specific advertising program offered by each site, their pricing model and other relevant data such as which sites are affiliated. The result is an accurate report of which local businesses advertise online, how much they spend and on which sites, on a nationwide, multi-vertical base. Our clients then use this data to improve their sales forces' productivity and to conduct competitive analysis.
 
Our clients have found that it is much easier to sell services such as SEM to landscapers and roofers who already advertise online than to convince someone who has been advertising only on print YP over the last 20 years and who doesn't understand what's wrong with that...

You can see some of our high level data reports about local business ad spending on our Palore blog.

Paul: Since I last spoke with you, Palore transitioned from a consumer service to a data syndication service. Can you explain this transition?

Hanan: Indeed, we started with a consumer application that used our local business data. The application had a good uptake with some 150,000 downloads during the first few months, but then we were approached by a number of companies that wanted to buy our data. Initially we refused to give it away, but when we were told by one of the largest local search sites that we have the most comprehensive database in the world about local businesses, and that they were willing to pay handsomely for that data, we decided to make the transition into becoming data providers. I think this was a good decision since it enabled us to focus on data extraction and really boost our capacity.

Paul: I've known Greg Sterling for a quite a few years now and he is one of the people I look up to the most when it comes to local search. He's also on your advisory board. How did this relationship come into play?

Hanan: We have been reading Greg Sterling's Screenwerk blog for industry news and analysis ever since we entered the local search space. Since Greg was right across the Bay Bridge from us in San Francisco, we asked if he would meet with us, and he did. Greg was so helpful and friendly in that meeting and in subsequent meetings that we invited him to join our advisory board. Greg has been an endless source of good advice and insights and we feel very fortunate to have him on board!

Paul: I've always considered local search as bringing users from their computers to a physical business door. For Palore, what are the main differences between online and offline content sources?

Hanan: Palore is a child of the continuing shift from the offline world to the online world in local search. While traditional content providers have relied on print directories, government records and other offline sources, Palore is a pure online play. This has several implications:

  • Freshness - our content is always up to date. For example, in our ad spend service we can provide a list of plumbers who have advertised last week on a hundred different sites. The offline equivalent of that is a list of plumbers who have advertised on a print YP last year...
  • Unique content - our content is a reflection of the new content types that are available online. Things like reviews, online reservations, company websites and online ad spending have become very relevant to our clients and they can be collected almost exclusively online.
  • Scale - from an operational perspective, collecting online data using crawlers is much more scalable and cost effective than transcribing print pages in Costa Rica...
Paul: Local aggregate data has been pretty huge for companies who provide this information to portals such as Yellow Pages. How does Palore differ from companies like InfoUSA, Acxiom, and Localeze?

Hanan: I think all these companies have done a great job in providing content to this growing industry. As a technological start-up, Palore has taken a different approach of focusing just on online data. While we don't see ourselves competing with these companies on base data such as name, phone number and address, I think we can complement their data with new information that is available only online (see my previous answer).

Paul: I often consider myself technologically inept, which is strange since I make my living on the Web. What does it take to collect information from literally hundreds of sites?

Hanan: Extracting online content about millions of businesses from hundreds of sites is not an easy task, and has required many man years of R&D work.
 
The process starts with identifying the right (and copyright approved) content sources and deciding what parameters we want to extract from each web page. We then use our crawler infrastructure which consists of dozens of servers, spread throughout the US internet backbone, to access millions of pages from the chosen sites. Reaching this information, also known as the "deep web" (http://en.wikipedia.org/wiki/Deep_Web) requires special techniques such as automated address form filling.
 
Next, each business page is analyzed and the relevant data is extracted from it. We then merge the data from the different sources, reconcile any discrepancies, clean it up, QA it and normalize it to a unified format.

While many web extraction companies deal with similar challenges, the local nature of our industry makes this especially difficult. Many of the hyper local sites were designed using obsolete technologies, with strange implementations of navigation methods, text encodings, cookies and Ajax that are hard to deal with. But this makes it ever more satisfying to complete a complex data aggregation project from multiple sources and to deliver a unified feed that really helps our clients.

Paul: Hanan, thank you very much for taking your time to answer these questions. Local search has come a long way in the last five years and will only continue to grow. We look forward to hearing more about you in the future.




 




 


June 30, 2008





Paul Jahn is the owner of Localmn Interactive Marketing, a Minnesota-based Internet marketing company. Localmn focuses on helping local companies large and small leverage the most out of their search marketing campaigns depending on their respective needs, including SEO, PPC, local search, and social media campaigns. Paul also runs and contributes to the Localmn local search blog.

With over ten years of Internet marketing experience, Paul has particular expertise in search marketing, with extensive knowledge of local search.






Comments closed after 30 days to combat spam.


Search Engine Guide > Paul Jahn > Local Search Interview with Hanan Lifshitz, CEO of Palore