The Noisy Channel

 

Google Offers “More And Better Search Refinements”

March 24th, 2009 · 11 Comments · General

Fresh news, hot from the Official Google Blog:

Starting today, we’re deploying a new technology that can better understand associations and concepts related to your search, and one of its first applications lets us offer you even more useful related searches (the terms found at the bottom, and sometimes at the top, of the search results page).

For example, if you search for [principles of physics], our algorithms understand that “angular momentum,” “special relativity,” “big bang” and “quantum mechanic” are related terms that could help you find what you need.

A couple of reactions. First, Google has offered related searches for a while, so I’d love to know what makes these “more and better”. I can’t tell from playing with it, and the suggestions I see aren’t as good as, say, Kosmix. Second, if they believe that this feature can improve user experience, why are they putting the results at the bottom of the page (at least on all of my queries)? Surely they know from their own logs that only a minority of users look to the end of the results list.

While I see this enhancement as a step in the right direction for Google, I wonder if they have their hearts in it. Google used to promote refinements–actually faceted search refinements–on their product search site, but pushed those to the bottom too. It seems very hard for them to get away from the primacy of those ten blue links.

I’d like to get excited about Google embracing HCIR, especially after they were so kind as to let me lecture them about it. And perhaps I’m being too harsh a critic. Their post concludes:

Even if you don’t notice all of our changes, rest assured we’re hard at work making sure you have the highest quality

It seems to me that they go out of their way to make sure that changes aren’t noticeable to users. I suppose their conservative attitude might cost them the occasional designer, but hasn’t hurt their pocketbooks.

Come on, guys, you’re the market leaders! Don’t be so timid.

11 responses so far ↓

  • 1 jeremy // Mar 24, 2009 at 3:05 pm

    Second, if they believe that this feature can improve user experience, why are they putting the results at the bottom of the page (at least on all of my queries)? Surely they know from their own logs that only a minority of users look to the end of the results list.

    I’ve heard them comment about this. They say that the reason why they put things like this at the bottom, only, is that they don’t want it to interfere with your normal successful searching.

    Their opinion, if I understand it correctly, is that if a query is a successful query, and you find exactly what you want in the 1st or 2nd rank, then Google’s opinion is that having the facets at the top or at the side only gets in the way of you getting to that top result as quickly as possible.

    It is only when the query is an unsuccessful query that you start scrolling down further in the ranked list, so that you eventually wind up at the bottom of the list. At that point, you see the query suggestions, and only at that point do they become relevant to you, i.e. when your query has failed.

    This is Google’s story, and they appear to be sticking to it.

    Now, on the surface it sounds reasonable. But I don’t believe it, for two reasons:

    (1) Long ago they started putting ads above the top-ranked result, instead of just over to the side at the right. So they already throw up junk in front of the top ranked search result, and make it harder for the user to find/click that top result.

    (2) As a searcher, I don’t always use exploratory facets or query suggestions in an explicit way. Often I use them to conceptually orient myself in the searches that I do. I read over the facets to get an understanding of what the space is about, without actually clicking on them. They aren’t always directly useful, but they are extremely valuable in helping me establish a context around the rest of the information that I am seeing.

    Of course, Google doesn’t realize this, because they can’t measure it. If nobody clicks one of the query suggestions or facets, they have no way of knowing whether those facets were useful.. even if they were extremely useful to me.

    And so having facets at the top, even if I never click them, doesn’t interfere with my normal, successful searching, anyway, as Google fears. Even if I never click anything, seeing that information is valuable to me.

  • 2 jeremy // Mar 24, 2009 at 3:08 pm

    Notice how in the Official Google Blog writeup (the link that you give), they don’t actually show a query in which any advertisements are shown.

    One could read a whole lot into that.

  • 3 Otis Gospodnetic // Mar 25, 2009 at 12:03 am

    I didn’t look into this yet, but note how their examples (the ones you quoted) don’t include any simple substring matching. The related searches I used to see on Google’s SERPs *tended* to be *mostly* text-wise similar queries.

  • 4 Max L. Wilson // Mar 25, 2009 at 9:41 am

    Daniel – Interesting post. I have also heard what Jeremy had heard, in fact, its possible Jeremy and I discussed it at JCDL. Anyway, it is disappointing that the HCIR elements are at the bottom. I saw a talk at a workshop at SIGIR07 (I guess it was), on eye-tracking studies over search engines. This was Yahoo, if I remember rightly, and there was talk of a golden triangle, esentially spanning from top-left to top-right to bottom-left. The strange assumption they had was that they should try to fit everything into that golden triangle. The question I asked, which came back with little response, was what can they do to make it a golden square. It seems like, as with many search UIs, making use of the right hand side might be a way to do that. If it was under a few sponsored links too (towards the bottom-right of the avg screen), then it would potentially make it even more square.

    Jeremy – we are about to promote this same position about getting context from facets without specifically interacting with them, at the sensemaking workshop at CHI. It follows on from our Backward Highlighting paper at UIST, which did try to measure it to some extent.

  • 5 jeremy // Mar 25, 2009 at 10:51 am

    This was Yahoo, if I remember rightly, and there was talk of a golden triangle, esentially spanning from top-left to top-right to bottom-left. The strange assumption they had was that they should try to fit everything into that golden triangle.

    Oh my goodness — don’t even get me started. I agree, that is a strange, strange assumption. It’s almost completely backwards. The statistical distribution of eye movement across the page is not independent of the design itself. Change the design, and you’ll change the golden triangle into an oval, square, vertical striped lines, whatever. It’s almost as if they assume that the golden triangle is an objective, decontextualized fundamental law of the cosmos, and design everything to fit into that. That doesn’t make any sense to me.

    (FWIW, it might have been Yahoo, but Google also does this golden triangle stuff:

    http://googleblog.blogspot.com/2009/02/eye-tracking-studies-more-than-meets.html)

    About your CHI workshop paper: That sounds quite interesting! Any way you could send me a copy? Or do you have to wait until it’s officially published? :-)

  • 6 Daniel Tunkelang // Mar 25, 2009 at 11:06 am

    I’m not an expert on eye-tracking, but common sense leads me to agree with Jeremy on this one. At most I imagine that we are predisposed towards the design elements we expect based on familiarity , but surely we react at least in part to the actual design put in front of us. Indeed, those expectations have changed over the past several years.

    Otis, you might be right that their previous related searches were strictly based on text-wise query similarity. Too late to check now! But I think that’s still largely true of the current ones.

    For example, compare:

    http://www.google.com/search?q=sigir

    http://www.kosmix.com/topic/sigir

    And, in case anyone is wondering, I have no vested interest in Kosmix, and more than in Duck Duck Go or any other web search engine. But I do make a point of plugging the folks who are pushing the envelope of exploratory search on the web.

  • 7 Taking the Google Wonder Wheel for a Spin | The Noisy Channel // Mar 25, 2009 at 2:35 pm

    [...] I blogged yesterday, I’m glad that Google is giving exploratory / HCIR approaches a shot. But I’m shocked [...]

  • 8 Much ado about nothing « The Mendicant Bug // Mar 26, 2009 at 2:19 am

    [...] post on it is worth [...]

  • 9 Is Google Diving Head First Into HCIR? | The Noisy Channel // May 12, 2009 at 10:18 pm

    [...] features they’d already rolled out, and that I personally didn’t find overwhelming (see here and here). Still, I’m pleased that their marketing language is embracing [...]

  • 10 Max L. Wilson - Max's Blog // Oct 17, 2009 at 4:06 am

    [...] was recently interested by a debate about why Google sticks its facets, and now its query refinements etc, at the bottom of the s…. The basic assumption that was proposed was that you only need to refine your results if you didnt [...]

  • 11 Why Google Is Tweaking The Look Of Its Search Results — Tech News and Analysis // Apr 4, 2012 at 12:54 pm

    [...] on more quickly to the sites they actually want to visit. Same goes for The Noisy Channel, which complains that the list of search suggestions is still, for the most part, at the bottom of the page, a [...]

Clicky Web Analytics