In yesterday’s post, I discussed the human connection in relevance. This post deals with the discovery angle, which basically says that relevance can be expected or unexpected.

Discovery is a place where we can find true delight in life. It may be gratifying to find what we’re already searching for. But it is superb to stumble across the perfect thing when we didn’t even know we needed it.

Exceptional design always encompasses discovery. Nobody knew they wanted an iPod before it existed. Nobody knew they wanted a fax machine or a rubber garlic peeler thingy. Once people saw these things, though, or realized what they could do with them, they couldn’t imagine ever having lived without them.

In search, discovery can make the job of the engine easier and more difficult at the same time. Easier because results aren’t limited to what already exists in the conscious mind of the user, but more difficult because the search engine has to know the mind of the searcher better than he knows his own mind.

Just as the human connection has its own set of qualities or expressions, so too does discovery. I’ll touch on a few here, and I invite you to contribute to this ever-broadening definition.

1. Happenstance discovery
   You’re walking down the street and the perfect pair of shoes catches your eye. You’re reading the business section of the New York Times and spy a headline about someone you went to high school with.

   Happenstance discovery is entirely unintentional on both sides of the equation. Yes, the shoe store is hoping their shoes will catch your eye, but it’s not really your eye they’re hoping for, is it? After all, they don’t know you, didn’t know you’d be walking by that shop at that time, don’t even know if you wear shoes at all. And a newspaper is putting out content they want people to be interested in, but they’re not trying to reconnect classmates, at least not overtly.

   Happenstance discovery is the ‘Can you believe it? What are the chances?’ type discovery. (That’s what I say to my sweetheart when I come home with a new pair of shoes.)

2. Passive discovery
   Passive discovery is intentional on the part of the person or company serving up the content, but one-sided. The user may know it’s happening, but doesn’t participate in its creation. Passive discovery is what’s behind most recommendation engines: ‘Did they buy a big screen TV? Sell them some carpet! 42% of people who bought a big screen TV changed their carpet within six months!’

3. Collaborative discovery
   Collaborative discovery is possibly the most powerful form of discovery. Essentially, collaborative discovery is when the user participates in the process:

   I am going to take your hand, and you and I are going to walk in that direction. Neither of us knows exactly where we will end up, but I know you, and you know yourself, and together we will make decisions along the way that will allow us to find experiences we can’t begin to imagine.

   How can it still be discovery if you are participating, you may ask? Doesn’t that mean you’re aware of what’s going on? Well… no, not really. Haven’t you ever chosen a dish off a Chinese menu without having the foggiest idea of what you were ordering?

   Collaborative discovery is when the intimate knowledge you have of yourself comes together with someone else’s intimate knowledge of content. If you don’t truly understand yourself, how can you define what you’re after? And if the person on the other end doesn’t comprehend the intricacies of what she’s offering, how will she know what will suit you personally?

   Collaborative discovery is where VortexDNA lives.

What I noticed as I was writing this piece is that all of these qualities that I’m attributing to relevance are interconnected. It may seem that there’s some repetition, for example, between the assumed interest I spoke about yesterday and the delivered discovery I spoke about today. The truth is that we’re weaving a web here; bringing together tightly woven concepts to create a cohesive understanding of relevance in search.

What do you think? Do you think I’m making arbitrary distinctions here? Or do you see some value in breaking relevance down into its composite parts? I’d be delighted to hear from you.

This entry was posted on Monday, July 30th, 2007 at 9:17 am and is filed under VortexDNA, Relevance, Search. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.