Local Search, Meet Facebook’s Graph Search

Excerpts from my latest article at Practical eCommerce: “Facebook Takes on Local Search.”

Consumers use the Internet to search for local businesses, events, and topics. Google is a major player in local search, but so is Bing, Yahoo!, Yelp, YellowPages.com, Foursquare, and other sites. Local search now has a new competitor: Facebook Graph Search.

Launched this week, Facebook’s new search product focuses on finding connections between people, photos, places, and interests. But the most interesting aspect of Facebook search is its potential in the local search space.

Recommendations from friends are a powerful inducement to try something new, and Facebook search makes it easier to discover the places and things your friends like.

Facebook has had the pieces for this new product for a while — Places and Pages to house business information, the ability to friend or follow, and the ability to check in and comment. All of these are vital pieces of the puzzle.

But each interaction with a Place or Page was merely dumped into the user’s Facebook News Feed or Timeline. The result was a fleeting stream of information that might get noticed but would probably be overlooked or quickly replaced with something newer. Those are all missed connections, missed chances to discover.

With Facebook search, these connections are discoverable at will….

Read the article in full at Practical eCommerce »


Web PieRat logo.

Originally posted on Web PieRat.

Facebook’s Graph Search: A Different Kind of Search

Excerpts from my latest article at Resource’s weThink blog: “Facebook Launches a Different Kind of Search.”

Facebook’s much anticipated new search product, Graph Search, launched to a limited set of users this week with a mission to help users connect and discover. With a reported 1 billion people, 240 billion photos and 1 trillion connections to create uniquely Facebook search results from, is this Facebook’s play to win search dominance from Google? Not yet, at least.

Founder Mark Zuckerburg stressed in his announcement speech that “I think what you’ve seen today is a really different product from what’s out there.”

The difference is in the dataset. Traditional web search is designed to receive any textual input via the search box and convert that to search results. Google, Bing, and other traditional search engines use armies of crawlers to index as many pages of the Internet as possible. From that index of content, they apply their own algorithms to determine context, relevance and authority, and produce a set of search results.

Facebook’s new Graph Search is different in two important ways: Facebook doesn’t index the Internet, and search is based on controlled filters rather than open-ended search queries. In place of crawling, Facebook relies on its user base to create or discover content that they find interesting across the Internet and share it with their Facebook friends….

Read the article in full at Resource’s weThink blog for more on how Facebook search works and what it’s ramifications are for users »


Web PieRat logo.

Originally posted on Web PieRat.

Opinion: Facebook Search Is Nifty, Not Necessary

My personal opinion, because of course I have one as a frequent Facebook user, is that Facebook search is a nifty toy but not a compelling feature by itself. I suspect people will try it out, find that the results aren’t as rich as Facebook promises and mostly won’t bother to use it again. Even more than traditional web search, Facebook search relies on network participation and data input.

What percentage of Facebook users have a large, active network with many friends that actually share similar interests? Actually, I’ve never seen statistics like that from Facebook. But unless it’s a majority of users, most users will find their results rather anemic. Few overlapping interests or few active friends and really there’s not a lot of personalized data to return in the search results.

Here’s an example. I’ve lived in the same northwestern suburb of Chicago for two and a half years. I’m Facebook friends with exactly four people who live here. One of them is my husband who pretty much frequents the same places I do. One hasn’t posted on Facebook in over three months. And two are frequent Facebook sharers, but mostly of adorable photos of their kids and lives. In addition I’ve Liked perhaps 10 local businesses and organizations. So if I ask Facebook Graph Search to recommend local businesses based on my friends’ likes, the results are going to be pretty meager. Sure, Facebook will pull in recommendations from other Facebook users near me, but at that point my results are just like any other generic resident of my town. Hmpf.

Now, when we’re talking about photo search I think it gets more interesting. I do see myself using Facebook’s photo search to find that one photo that had my best friend and me in it that at my wedding posted by our best man last summer. It’s a pain the tuchus to scan through his timeline or my photos to find that one photo, but if I can remember who was in the picture, who posted it and approximately when, the list of search results should be nicely targeted to the photo I’m looking for. IF someone bothered to tag me and my best friend. If not, well, all I have to go on is photos posted by the best man last summer. Still an easier task than scanning through them all, but you can see how effective Facebook search requires effective data input.

And that goes to the root of the issue. Google and Bing have spent years trying to find other signals to understand context, relevance and authority. They’ve come a long way but there’s still a long way to go, especially with photos and other media.

Really, it boils down to data input. If I post a photo on a blog with a numeric name like 123.jpg and don’t include any textual information about what that photo is of or who is in it, and there’s no tagging for date or location, that photo is essentially useless to the world of search based on the lack of data input. Garbage in, garbage (or nothing) out. Facebook has a leg up in that they at least have automatic data about who posted the photo and when, but unless it’s tagged with location, date of capture, people involved, and a useful description of what’s going on, that photo will be almost as useless to search as any generic data-less photo posted on the web.

And here’s the big issue: The average Facebook user doesn’t understand this and doesn’t want to. It takes time to input data and think of a reasonable comment. That’s why there are so many comments and descriptions out there like “What’s happening here?!?!” and “OMG.” Garbage in, garbage out.

Unless Facebook can train more users to thoughtfully input more data, Facebook search isn’t going to be as rich as their marketing videos indicate. When a Facebook employee looks at his search results, of course they’re rich — he knows a lot of people at Facebook who probably spend more time than the average bear on thoughtful data input when they share. Because they understand the value and quite possibly just a little nerdy like that. I also enjoy data input when I share, because I’m a little nerdy like that. “Oh, an optional field to fill in, well sure I’ll do that!”

Most people on Facebook and in real life are just not that into data input. And that’s the main reason why Facebook search is nifty rather than truly necessary.


Web PieRat logo.

Originally posted on Web PieRat.

Four New Year’s Resolutions for SEO in 2013

Excerpts from my latest article at Practical eCommerce: “SEO New Year’s Resolutions for 2013.”

With a new year ahead, it’s time to think about New Year’s resolutions. What do you want search engine optimization to do for your site in 2013? The most likely goals for any ecommerce site revolve around driving more traffic and converting more visitors. Let’s look at some steps for analyzing, planning and implementing stronger SEO programs in 2013.

  • Drive More SEO Traffic
  • Convert More SEO Visitors
  • Implement More SEO Actions
  • Build Better Relationships

Each resolution includes details and links to articles for more practical SEO tips. Enjoy!

Read the article in full at Practical eCommerce »


Web PieRat logo.

Originally posted on Web PieRat.