The Google Mobile blog announced that Arabic and Hebrew mobile searchers are now able to see universal results directly on their mobile devices. Although not related, this comes soon after Google Translated recently added 9 new languages.

I know a bit of Hebrew, so I tried to test out the search results on Google Israel on my iPhone. Unfortunately, I did not see the same mobile friendly results, with universal search, on my iPhone. Maybe it was related to me searching on my iPhone or searching in the states, I don’t know.

What it should look like is:

What I see on my iPhone:

Heck, it doesn’t even look iPhone friendly. To give Google some credit, the iPhone is not fully supported yet in Israel, so maybe this is not functional on the iPhone yet. Maybe it is only working on supported mobile devices?

Search and online marking firm 360i published findings from a small but structured test of how the “Caffeine” search infrastructure changes could affect results and ranking on Google. As Vanessa Fox wrote when Caffeine went public earlier this month, there apparently would be a number of changes in the ordering and content of results. As she observed in that post, “Google Caffeine will cause quite a kerfluffle in the web developer and search engine optimization world and many will dive in to try and figure out the changes.”

This is precisely what 360i has now tried to do. The company evaluated 40 retail-oriented keywords/searches to compare results before and after, and the potential SEO implications of Caffeine:

[A] sample set of 40 retail keywords. We looked at ten major retail brand names (keywords), ten retail head terms (single keywords), ten retail torso terms (two-word phrases) and ten retail long-tail phrases (four-word phrases) and compared the search results on the first three pages of both engines (standard Google and “Caffeinated” Google).

With the caveat that “everything could change” before Caffeine formally rolls out 360i offered six observations or “things to expect” if/when it does. Here they are:

1. Domains and rankings will fluctuate.
2. The index size, or “competition,” of single keyword search relevance will increase.
3. You’ll see a boost in relevance for long-tail searches.
4. You’ll get results (SERPs) in half the time, on average.
5. Blended results will increase.
6. There will be a social jolt.

To read the larger discussion of each of these takeaways go to the 360i blog post.

There’s a feeling in the air of change — from one era to another and, more precisely, from one type of media culture to another. The hearings on Capitol Hill yesterday on the “future of journalism” reflect this transition in which traditional media are under intense pressure and fighting for their lives in some cases. These changes are not simply being brought on by the recession but by the pace of technology development and the corresponding tectonic shifts in the consumption of media.

Just as Microsoft was and is synonymous with the PC, Google has come to be synonymous with the internet in popular culture. But with the rise of mobility and “the cloud” the question arises: ten years from now will Google still be on top?

There has also been a paradox around search. It can’t and won’t remain the same in this dynamic environment, and although there have been some UI changes (as well as infrastructure changes to support those), it has been hard to imagine what the future would look like exactly. Sergey Brin, writing this years’ Founder’s Letter, addressed the state of Google but also looked into the future of search:

I think it will soon be possible to have a search engine that “understands” more of the queries and documents than we do today. Others claim to have accomplished this, and Google’s systems have more smarts behind the curtains than may be apparent from the outside, but the field as a whole is still shy of where I would have expected it to be. Part of the reason is the dramatic growth of the web — for any particular query, it is likely there are many documents on the topic using the exact same vocabulary. And as the web grows, so does the breadth and depth of the curiosity of those searching. I expect our search engine to become much “smarter” in the coming decade.

So too will the interfaces by which users look for and receive information. While many things have changed, the basic structure of Google search results today is fairly similar to how it was ten years ago. This is partly because of the benefits of simplicity; in fact, the Google homepage has become increasingly simple over the years: http://blogoscoped.com/archive/2006-04-21-n63.html. But we are starting to see more significant changes in search interfaces. Today you can search from your cell phone by just speaking into it and Google Reader can suggest interesting blogs without any query at all. It is my expectation that in the next decade our searches and results will look very different than they do today.

One of the most striking changes that has happened in the past few years is that search results are no longer just web pages. They include images, videos, books, maps, and more. From the outset, we realized that to have comprehensive search we would have to venture beyond web pages. In 2001, we launched Google Image Search and via Google Groups we made available and searchable the most comprehensive archive of Usenet postings ever assembled (800 million messages dating back to 1981).

Larry Dignan thinks the reference in this letter to “a search engine that ‘understands’” is an indirect reference to Wolfram/Alpha. (it could equally be about claims made by Kosmix, Hakia, Powerset [Microsoft] and a couple of others.) In terms of the future of search, Twitter and other “Help Engines” could also represent a compelling new direction the market and industry could go.

Back to Google: today the company held a press briefing before Google’s annual shareholder meeting. It featured Dave Drummond, SVP, corporate development; Susan Wojcicki, VP product management; Kent Walker, general counsel; and Marissa Mayer, VP search products. They fielded an array of queries on everything from Eric Schmidt’s continued membership on the Apple board (he hasn’t considered resigning) to YouTube (”eventually [it will] be a successful and profitable business”). Here’s a factual rundown from both AllThingsD and TechCrunch (I wasn’t present).

There were no revelations apparently, though Eric Schmidt again pointed to the netbook arena as one that would see some Google-friendly developments in the near future.

In a blog post today, Google confirms what we and others reported last week — they’re now showing local results on non-local queries.

Google says they’re guessing the searcher’s location, “in most cases” by using your IP address. As I pointed out last week, this isn’t always the most accurate solution. And, as Greg Sterling pointed out, Google will eventually replace IP-targeting with improved triangulation/GPS right in the web browser/device you’re using.

Meanwhile, Google Software Engineer Jim Muller replied to the questions we emailed last week when I wrote the original article on this. A few days have passed, but here’s what Jim had to say about the new expanded local search results.

Matt: Has this been rolled out for all Google users, or are only some people getting these results?

Jim: Yes, this is now available to all users worldwide.

How many terms have you targeted for this expansion of the 10-pack? (I don’t get the 10-pack when I search for “cars”, for example.) Are we talking dozens or hundreds of terms? More?

Many many more. We try to include all the world’s local information in the database that we search, so all that’s really required is a match to a place that’s near you. For example if you search for specific stores or specific street addresses near you, we will often show the map. The map can appear in all the usual configurations: in groups of 10, groups of 3, or alone.

But as you noted, we don’t always show the map. We try to show it whenever our algorithms determine that it will be most useful.

The 10-pack on these broad queries never seems to show up at the top of the results — not even “pizza.” Is that by design, or might it appear at the top for certain queries?

For these queries, the user’s intent is less explicitly local than for a query like [pizza baltimore], so our ranking algorithms tend to rank the map in the middle of the page. For the highest quality single results, our ranking algorithms often place them higher on the page, and in particular street address queries often appear on a map at the top, as we would with a strong user intent query.

Google is rewriting the local search space. They’re now showing local search results — a map, business listings, and more — even when searchers use generic terms that don’t include a local word. This was spotted in London last week, written about on Saturday by a California florist, and spread widely yesterday when Mike Blumenthal wrote it up on his blog.

This has potentially huge implications for searchers, local business owners, big businesses with a local presence, and search marketers, too. As I wrote on my own blog yesterday, Google is changing the game where local search is concerned. As Andrew Shotland asked, is every search local now? No, but we’re getting there. Google must be very confident in its ability to identify local intent, and its ability to minimize the ongoing map spam problem.

An explanation below of how it seems to work and thoughts on what it means, but first try it yourself: Do a Google search for pizza, and you should see something like this, only tailored to your area:

Even though I didn’t add a city name to my query, Google is recognizing that my search probably has local intent and shows me the “10-pack” of local results with a corresponding map. The results appear to be IP-based; the image above focuses on Kennewick, WA — it’s not my hometown, but is where my ISP is located, about 10 miles away.

How It Works

1.) Singularity / Plurality sometimes matters

Local results will show on a search for “attorneys”, but not “attorney.” Ditto for “real estate agents,” but not the singular version of that. In other cases, local results show for both — “plumber” and “plumbers,” for example.

2.) It’s not always the 10-pack

On some generic searches, you may get the older “three-pack” that just shows three matching businesses. In my area, searches for “bowling” and “ford dealer” only show three matching local results and a map. Makes sense, because I live in a smaller area and we don’t have 10 Ford dealers here.

3.) It’s not just commercial terms

The search marketer in me focuses on business-related queries, but this extends to non-commercial terms, too. A search for “parks” brings up a mix of listings — some community parks, local government listings, mobile home parks, and even a funeral home.

4.) Google’s local targeting is debatable

A search for “restaurants” or “italian restaurants” probably has local intent, and both of those generic terms show the local 10-pack for my area. But I also get local results on a similar, but not necessarily local search for “italian food.” I get local results on generic words like “liquor” and “burgers,” both of which might — or might not — mean I’m looking for something local. It’ll be interesting to watch if/how Google tweaks its algorithm to perfect its local targeting on generic terms.

5.) The local results are never the best match for a generic term

In all the searches I’ve done — as well as all the searches done by a half-dozen friends in the local SEO/SEM field — these local results never appear at the top of the results for a generic term. At best, they’re showing up in the fourth spot — and sometimes further down the page.

6.) Local results show internationally

Fellow search marketers have reported seeing the local results on non-local keywords in London, Canada, and all around the world. After I wrote about this on my own blog, Pieter van Schalkwyk from FlowCentric Australia said he also sees local Sydney results on a Google search for “attorneys.”

What it means

1.) This is great news for small/local businesses

With this change, small/local businesses will now be getting exposure on at least hundreds, and probably thousands of prime keywords. As search marketers, we often tell our small business clients that they don’t want to rank for terms like “lawyer” or “doctor” because they’re too generic, and the competition for those prime terms would be beyond their reach. But, with Google showing local results on this prime real estate, a big door of opportunity has just opened up. A doctor in Topeka can get visibility on the term “doctor,” but only when local folks type it in.

Likewise, this should be good news for local search marketers who understand the ins and outs of optimizing local business profiles and web sites and can get their clients listed in the 10-pack or 3-pack. It also further kills the value of ranking reports, because rankings are now even more tied to geography and even your choice of ISP (see below for more).

2.) Searcher behavior may change

We recently reported that search queries are getting longer, and these generic, 1- and 2-word queries are declining. Perhaps some of that is due to searchers not being happy with the results after using a generic term like “restaurants” or “real estate agents.” All of that might be out the window now. Google’s AdWords keyword tool says there were about 6 million searches for “restaurants” in February. How many of those searchers will be happy now that they’ll get local results for such a broad term? Enough to influence the overall searcher behavior? Possibly.

3.) Suddenly, your choice of ISP matters

With the results based on the searcher’s IP, it now matters who your ISP is — and more importantly, who your customers use as their ISP. Here are the results I get from home on a search for “lawyer.”

Meanwhile, my wife’s office about 10 miles away uses a different ISP. They’re also based in Kennewick, but here’s what she sees on the same search from her office.

Notice that the order is slightly different: results B & C are switched, as are I & J. What would happen if one of us was using an ISP based in Pasco or Richland? We’d probably see an entirely different set of results because the IP address would point to a different city. How different will results be if your ISP is in Minneapolis, but you’re living in St. Paul? One Chicago-based search marketer told me on Twitter that he was getting pizza results for North Carolina! (His IP is based off a corporate server.)

4.) Google Maps will be a big winner

The added exposure of Google Maps and its local business listings will surely increase traffic to Google Maps. As of last January, Google Maps only got one visit for every 45 visits to a Google property. That should start to increase dramatically in the coming months.

The losers are likely to be companies like MapQuest, whose traffic will probably decline as even more searchers are exposed to Google Maps. And city guides / Internet Yellow Pages sites (like Citysearch, Yelp, Superpages.com, and so forth) may also see a loss of traffic, because their pages often ranked highly for some of these generic terms.

We have an email in to Google with some questions and requests for more information. We’ll update this post or publish a new one if/when we hear back.

Earlier this month we reported on Google’s test of favicons in Germany and Poland in AdWords. Subsequently I was alerted by someone in email that this was happening in the UK as well. Dave Naylor noticed it today in the UK and has some screenshots. I was unable to find any conducting various searches on Google.co.uk.

In addition Blastmodo found them in AdWords in the US:

Source: Blastmodo

Previously Barry also noticed favicons being tested in organic results.

So why is Google doing this? A cynic would respond it’s to boost CTRs, which the Checkout icon has reportedly done in many cases for companies that participate. However I believe there are a number of issues converging here:

• The pressure to allow “branding on SERPs” (a la Search Monkey)
• The shift to a more “visual” SERP via universal/blended search
• And . . . the desire for higher CTRs

Assume that Google rolls this out. If everyone gets to do it and/or they’re inserted with all ads it could result in clutter and “favicon blindness” to some degree. But I believe that we will see favicons on SERPs.

Mike Blumenthal reports seeing Google test a new type of map within the search results listings. He said he is only able to trigger this map to show using Firefox with the Google Toolbar installed. So his theory is that Google is testing some geolocation technology in the Firefox toolbar. I personally cannot reproduce it, but I trust Mike is seeing this. Here is a screen capture of one such result:

Others reportedly have seen this type of map embedded within specific listings a month or so ago. It would make sense for Google to show maps in some cases, when you are looking up a specific company and the address may be relevant to the searcher.

Google frequently tests changing how maps are displayed in the search results, and with new geolocation capabilities being built into browsers, I would not be surprised to see more enhancements and experiments in 2009.

There’s an interesting post today on the Official Google Blog that gives a glimpse inside the company’s usability and eye-tracking studies — tests that help Google determine what their search results pages should look like.

The image above, for example, shows how users interact with a standard results page with 10 links. The deeper colors up top, Google says, show that most people found what they were looking for in the first two results on the page.

No surprise there, but in May, 2007, when Google introduced Universal Search to its interface, marketers began to wonder how the addition of videos, image thumbnails, and other visual cues would affect searcher behavior. Says Google:

We ran a series of eye-tracking studies where we compared how users scan the search results pages with and without thumbnail images. Our studies showed that the thumbnails did not strongly affect the order of scanning the results and seemed to make it easier for the participants to find the result they wanted.

The thumbnail image seemed to make results with thumbnails easy to notice when the users wanted them … and the thumbnails also seemed to make it easy for people to skip over the results with thumbnails when those results were not relevant to their search.

(There’s an image in Google’s blog post that goes along with that explanation.)

Overall, it’s not the most groundbreaking or in-depth thing you’ll read, but it does provide an interesting peek into how Google tests its search results pages and how we interact with them.

The Forrester Blog published a small but interesting study on how you can improve your chances, by 50 times, of showing up at the top of the Google search results. Their tip? Utilize Google’s Universal Search by creating videos.

Not only does Forrester share statistics and analysis on how to improve your odds with videos, but they also share tips on how to improve your video’s chance at being the top video in Google web search.

Let’s first discuss the odds. The blog post showed that a video has a “11,000-to-1 chance of making it onto the first page of results.” If you compare that to web pages for the same keyword set, you only have a “500,000-to-1 chance” of getting on the first page. So for keywords that show video results in Google, you have a “50 times better chance of appearing on the first page of results than any given text page in the index,” reports Forrester.

How can you improve those odds? Well, use some of the tips Forrester gives on how to achieve that top spot for your video.

• Insert keywords into your video filenames.
• Host your videos on YouTube, and embed those YouTube videos into your own site. Google says its algorithms consider how many times a video is viewed, and any views embedded videos receive on your own site get added to the ‘views’ tally on YouTube. (And yes, nearly every video we saw Google blend into its results came from YouTube.)
• Optimize your YouTube videos by writing keywords into your videos’ titles, descriptions, and tags.
• Embed videos into relevant text pages on your site. The context provided by the text on those pages (which is hopefully already optimized for search as well) will help the search engines figure out what your videos are about.
• Also create a video library on your site, so Google knows where to find your video content. (Google Video Sitemaps can help with this too.) Write keyword-rich annotations for each video in the library.

For more details, you will want to check out the blog post at the Forrester Blog.

In some ways the “future of search” is a tired topic that has already seen hundreds (maybe thousands) of speculative articles. But it’s also a topic of critical importance for search leader Google, whose entire franchise is built upon doing it better than anyone else.

Just as Microsoft didn’t anticipate the rise of search and search advertising and Google didn’t anticipate Facebook, it’s possible that something as yet unknown could displace traditional search’s (and Google’s) centrality in the internet experience. Citing her blog post on the future of search, the UK publication TechRadar uses an interview with Marissa Mayer as an occasion to discuss the issue and the future of Google more generally.

Beyond universal search, Mayer emphasizes personalization, social search and location awareness as important elements of search’s evolution. She also points to Google’s “cloud computing” efforts and data storage as significant for Google’s future. As an aside, Google Health — specifically electronic Health Profiles, which is one of those efforts — will potentially get a big boost from the new Obama Administration, which wants to aggressively digitize heath records.

In the near term, Mayer also suggests some new interface changes will be coming this year:

Mayer reveals that the team has been evolving the interface design and user experience of the rich media-heavy search results since the launch in May 2007 and that we’ll see the fruits of this experimentation in the coming months.

The article paraphrases Mayer, saying that “while search is 90 per cent solved, the last 10 per cent will take decades to complete.” I would argue by contrast that search is about 50 percent solved and that it won’t necessarily take decades to see significant changes and improvements. Mobile will be an increasing influence on search for one thing, and that will play out in the next five years.

Location awareness in particular is much closer than the article implies. I agree with Mayer that personalization and being able to tap into social networks are two important, undeveloped areas in search. However, personalization and Google’s potential approach to “social search” are in hypothetical conflict.

Regarding “social search,” Google is inclined to incorporate anonymous, aggregated user data to influence search results; but users are likely to be interested not in the anonymous “crowd,” but in what their networks and friends have to say about things. That partly accounts for the rise of social networks. And, in some ways, “search” and “social networks” are opposites.

Google’s acquisition of mobile services Zingku and the Twitter-like Jaiku provide potential tools for the “network search” scenario. Google also owns Orkut, which has gained usage over the past year. (Google also bought DodgeBall a few years ago but doesn’t seem to be doing anything with it.) The challenge here is integrating these tools and services potentially with one another and search in an elegant way. A reader of my personal blog Screenwerk also predicted that Google will buy Twitter in the coming year. (Facebook has tried already and will likely try again.)

There are a range of emerging mobile services (i.e., Mosio, kgb’s Texperts, ChaCha) that, to varying degrees, allow mobile users to tap human networks for real-time “answers” or “recommendations” on the go. These are substitutes for traditional search. In a very basic way, so is text messaging.

(Beyond mobile, TV screens represent another frontier for search, and another challenge that may “evolve” search in interesting directions. That’s fodder for another post.)

The rise of the mobile internet is forcing search to be more precise rather than simply serving up a list of potential links for subsequent investigation. And personalization may well help in that effort. But more generally, search needs to remove noise and redundancy, in my opinion. Why, when I look for a restaurant or a business location for example, am I subjected to 15 links from competing sources for essentially the same information? That’s another topic for yet another post.

On her own future Mayer impliedly denies rumors that she’s about to leave Google as ValleyWag had suggested. TechRadar quotes Mayer:

“I’m a geek and Google is a great place for geeks,” she says. “I really love my job because I get to work on new problems and have new challenges each day. I’m currently working on our Geo products, Google Book Search and Google Health. They’re all things I’m excited to be part of.”