Google: Caught in the Act of Balancing

And yet another post on the recent changes to the Google SERP. This time, Chris Knoch sent me an interesting screen shot of a Google beta test that eliminated the side sponsored ads. I pondered the implications to the user experience. Also published Nov 18 on MediaPost.

In last week’s column, I talked about the number of changes I was seeing on the Google results page, and, in particular, how they might maintain the delicate balance between driving revenue from the page and maintaining user trust. No sooner did the digital ink dry on the column than I received an email from an old friend, Chris Knoch, formerly of Omniture and now VP of Marketing at Ready Financial. In his email, Chris included a screen shot of a rather interesting beta that Google is running:



It’s hard to say, given Google’s love for beta testing, how widely spread this test is and how indicative it might be of future ad presentations, but there are a number of fascinating implications to consider here. For today’s column, I’d like to focus on one of them: the elimination of the side ads.

Side ads generate a small percentage of the sponsored clicks from the page. For most results, the top two or 3 ads generate over 80% of the paid clicks on the page, with the 7 or 8 running down the right rail splitting the remaining 20%. That’s a lot of real estate to devote to underperforming ads. Will Google’s expandable alternative, with the user choosing to see 8 more ads, generate more clicks? I suspect so. Here’s why.

We judge the relevance and quality of blocks of information as a group, rather than consider them individually. The first ad in any block will dictate the performance of the block as a whole. If it’s a high quality ad, it’s saying to the user, “I’m relevant. Chances are the rest of the ads in this group could be relevant too. At least, you should spend a few seconds deciding for yourself!” But if it’s a low quality ad, it sends the message, “Don’t waste your time here. I’m not relevant, and everything below me is even worse.”

For side ads, this means that the top ad determines the depth of scanning engagement with the entire block. The position and visual treatment of the ads reinforces that it’s a “sidebar”, of secondary importance to the main purpose of the page. We won’t invest a lot of time scanning here, and if the first ad sucks, the rest of the block is doomed.

Google’s treatment provides a compelling alternative to the user. It restricts the number of ads shown to only the highest quality ones (you’ll notice that this presentation appeared on a broad query, where there would be sufficient inventory to provide high quality ads). The ads should be just as relevant to the intent of the user as the organic results, and given the query, probably more relevant. The user should be hooked. The presentation of two ads (I’d bet big money on the fact that Google will be testing both 2 and 3 ad presentations above the “more ads” button) gives a ready-made consideration set for the user. We’ve known for some time now that users “chunk off” a result set in groups of 2 or 3 results (maximum 4) and consider them as a group. There are natural visual barriers (the related search suggestions) that reinforce the visual presentation of the top ads as a group. What this means is that the user will judge relevancy, and if the first two (or three) ads pass the test, there’s a high likelihood that the set will be expanded.

When the set is expanded, the entire visual balance of the search results set is changed to the benefit of the advertisers, but the user initiates it. The user has given the ads an implicit vote of confidence, and by doing so, all organic results are pushed down out of visual scanning range. My guess is that this will result in much higher engagement with the ads, virtually eliminating the side bar blindness that has typically plagued right rail ads.

It’s a perfect example of maintaining user trust while driving more revenue. Based on this beta, I’d have to say, “Well done, Google!”

Google's Recent Changes: Here There Be Monsters

Here's the follow up to the Google Places Page post. The recent changes to Google's pages looked strangely familiar. In fact, it took me back to a research paper we did 3 years ago, titled, ironically: "The Future of the Search Results Page: 2010" (available free on our site). Prescient? Perhaps. This was first published at MediaPost on November 11, 2010.

Something's brewing in Mountain View. Google's geared up the SAR (Screwing Around Rate) of its results page to unprecedented levels. We have Google Instant, Place Search and Google Previews all rolling out in the last few months. And from around the blogosphere, there's rumors of testing that allows users to show 11 sponsored ads on top and also the telling switch of the label "Sponsored Links" to simply "Ads." So what do Google strategists have up their sleeves? 

The recent changes at Google prompted me to dig out a research paper we wrote a few years ago called "Search Engine Results: 2010." In it, I interviewed Marissa Mayer along with a dream team of search pundits and usability experts. A lot of what we're seeing today was hinted at in those interviews.  

For example, Mayer said: "If you imagine the results page, instead of being long and linear, and having ten results on the page that you can scroll through -- to having ten very heterogeneous results, where we show each of those results in a form that really suits their medium, and in a more condensed format. When you started seeing some diagrams, some video, some news, some charts, you might actually have a page that looks and feels more like an interactive encyclopedia."

Michael Ferguson, who was the UX lead at Ask, which had just rolled out Ask 3D (which, in hindsight, was well ahead of its time), went further: "There might be a time you might see people advertising and providing content not just on web pages and blogs etc. but with short discrete self-contained video answers and audio answers that come up either as sponsored or relevant content. So you might have a breaking down of search marketing that takes some of the things that have been learned like optimization and designing good text ads and seeing how that would work when you're delivering an audio 20 second pitch or delivering an audio content that drives traffic to your site."

There's a delicate balance that must be respected when you're combining the presentation of advertising and the way we search for information. As the results themselves become increasing rich and interactive, advertisers won't be very happy if the ads start to lag behind in terms of visual prominence. Mayer touched on this: "As you know, my theory is always that the ad should match the search results. So if you have text results, you have text ads, and if you have image results, you have image ads. So as the page becomes richer, the ads also need to become richer, just so that they look alive and match the page. That said, trust is a fundamental premise of search. Search is a learning activity."

It's this trust that makes the presentation of advertising a precarious proposition on the search results page. We're not there to find ads, we're there to find relevant information. If ads are highly relevant, we're receptive. If they're not, we'll skip over them. We accept ads not as ads, but as potential paths to relevant information.  

This is an important distinction. If ads start to look too much like ads we start to skip over them. And that decision is made in milliseconds, before the relevance of the information that lies on the other side of the ad is even considered.

This phenomenon is called banner blindness. Jakob Nielsen explains: "If they put up display ads, then they will start training people to exhibit more banner blindness, which will also cause them to not look at other types of multimedia on the page. So as long as the page is very clean and the only ads are the text ads that are keyword driven, then I think that putting pictures and probably even videos on there actually work well. The problem of course is they are inherently a more two dimensional media form, and video is 3 dimensional, because it's two dimensional - graphic, and the third dimension is time, so they become more difficult to process in this linear type of scanned document 'down the page' type of pattern." 

I believe that Google is now responding to the multi-screen search challenge. Search on a desktop needs to be different than search on a mobile device or on a tablet. Mayer's "encyclopedia" format makes sense here. But experimentation and the resulting change come at the potential price of alienating users.

Why have ads been the least changed part of the search page? It's certainly not because advertisers have been demanding that they remain as boring lines of text. It's because Google, along with Bing and Yahoo, are acutely aware of how important that trust is. The nature of our engagement with ads on a search page is far less straightforward than you might think. There's a lot of subtle psychology at play here.  In the words of Hector Barbossa, "You're off the edge of the map now mate, and here there be monsters!"

Google Defines 'You' On The Fly

Hopefully, over the coming holidays, I'll catch up with my blogging again. In the meantime, here are a few back posts that I haven't had a chance to get up on the site yet. Recently, Google's results page has been downright schizophrenic with all the changes it's been through. Here's my take on the importance of Places Pages. First published on MediaPost on November 4.

Google's ramping up of local results last week made me realize something: our Web presence is rapidly being taken out of our immediate control. Case in point, the Place Page.

Beyond the Walled Garden...

For over a year now, I've been pushing a mind shift to our clients, asking them to stop thinking of their online presence in terms of a "website" and more in terms of a portfolio of digital assets; some under their control and others either completely or partially out of their control. For every entity that lives online, there is a ripple effect. At the core is our website. Spreading out, usually with lessening degrees of control, are the "rings" of our presence: portal sites and extranets, mobile apps, information or products on channel partner sites, online ads, videos, interactions in the social space, comments, reviews, references and third-party apps that may access either our data or pieces of our functional infrastructure. The sum of all this is our online presence. As such, it is incumbent on us to be aware of what that looks like, and how visitors might interact with it.

The challenge is daunting for any company that has been online for a while. Even as an individual, according to Google I "live" online and in over 10,000 separate locations. And that's just what can be easily identified in Google's index. I suspect the number is even higher. Today's column will have its own ripple effect, adding to the collective total of what is "Gord Hotchkiss." My company's online presence is the sum of over 25,000 individual parts.

Bringing the Web to Your Neighborhood

Now, consider a tiny two- or three-person company in some small town somewhere in America. The odds are pretty good that they may not even have a website, or if they do, it may not have made much of an impact on the vast ecosystem of the Web. At least, that's been true up to now. But Google's Place Pages provides a prescient view of how our Web presence might be defined.

Place Pages aggregates at least some of the various pieces of a local business' online presence. The interesting part is that these Place Pages exist even if there's little or no input from the business owners themselves. It's an online presence defined by an algorithm -- or rather, multiple algorithms. It's a small digital snapshot of "you" as defined by Google. Google decides which parts of "you" it exposes.

Place Pages are important in Google's local search strategies because they solve a problem that restricted the growth of the hyper-local online market. People will only search if there's something there to find. Google had to create a scalable on-ramp model to give local businesses an online presence. The company did it by leveraging its strength: finding and organizing information. In this case, the presence is created from the information that defines the business on the Web. It's carrying a search results page one click further, making it specific to one company and structuring the data in a more cohesive way.

"You" on the Fly

This is interesting and important on two different levels. It shows that an online presence can be created through algorithmic aggregation alone, even in the absence of an official website. It shows how extensive our identities are online. Like it or not, we leave footprints on the digital landscape, and no one is in a better position than Google to gather those together to create online destinations on the fly. If this is true for the tiny Mom and Pop shop in Cannon Ball, N.D., it's even truer for bigger, more established entities, whether they be organizations or individuals. Will our online selves be increasing defined by Google, with or without our input?

The other thing to ponder is that this is scalable and driven by technology. Google has an open door to aggregate and present different types of information, specific to the type of company it is. I suspect a lot of what you see in the current Place Pages is simply a placeholder for new things to come.

The creation of Web destinations on the fly is quite probably a game-changer for Google.  It's a natural extension of the company's mission, organizing the world's information. It provides a new outlet for something that Google has been doing for well over a decade now: gathering together the ripples that define us online.

Learn more about Eye Tracking