Usually, when you click on a link in a set of search results at Google, the search engine will deliver you to the top of a web page. But what if it didn’t? What if it brought you instead to the place on a page where your query terms appeared, or just above those words?
For example, say you searched for [pizza 94043], and the page appearing at the top of Google’s search results included a list of pizza places, including one pizzeria at that zip code halfway down the page. How would you feel if, when you arrived at the page, your browser brought you to the part of the page where that pizza place showed up?
A patent application published today from Google explores how the search engine might insert artificial anchors into pages to deliver searchers to destinations within web pages, pdf files, word files, spreadsheets, and other documents, rather than just to the tops of those documents.
Below the Fold Design Implications?
How might you feel about visitors arriving at the middle or the bottom of web pages from search engines if you design web pages and are concerned about the appearance of pages and what viewers see when they visit your pages?
An old newspaper term that many have applied to web pages is the phrase “above the fold.” It describes the part of a newspaper that you might see before you purchase it in a coin-operated newsrack or on a store shelf. Many designers often refer to the top part of a web page that viewers first see when they arrive at that page as being “above the fold” and attempt to provide important information in that viewing area.
With large web pages, many designers may strive to have the area “above the fold” render more quickly so that visitors don’t have to wait to see that content when they first arrive at a page. An example is the pages at Amazon.com, which can often be fairly large pages, but the top parts of those pages tend to load quickly so that visitors don’t have to wait to see the content above the fold.
Would you design pages differently if search engines starting delivering visitors below the fold?
Named Anchors Could Be Used Too
Some pages, such as Frequently Asked Questions (FAQs) pages, often list a set of questions at the tops of those pages and then provide links from each question to anchors on the same page where the answer appears. Those anchors are often referred to as “named anchors” because they have a name associated with them like this:
<a name=”location”>
It’s possible that Google’s method could also deliver searchers to a named anchor when the anchor’s answer contains the query terms searched for. Google has described how they might do this in an Official Google Webmaster Blog post in September in response to some queries – Using named anchors to identify sections on your pages. They provide some tips on making it more likely to have links to your named anchors appear in search results.
But, the method described in the patent filing will work even if there aren’t named anchors like the kind you might find on a FAQ page:
With systems and methods described herein, mechanisms are provided to generate or simulate links with artificial named anchors and to allow the browser to recognize the artificial named anchor and navigate directly to the desired specific part of the target webpage even when the author of the webpage has not created a named anchor at the specific part of the webpage.
In particular, the systems and methods described herein simulate the general functionality of the named anchor and the HREF link to provide links containing artificial named anchors that allow navigation directly to a specific part of the target webpage even when a named anchor does not exist at the specific part of the target webpage.
Any webpage can utilize such links to provide a link to a specific part of another target webpage. In particular, such links containing artificial named anchors can be handy for search result pages returned by search engines.
The patent application also mentions that it might highlight or bold the query terms in question to make them easier to find for a searcher. The patent is:
Artificial Anchor for a Document
Invented by David P. Marmaros, Benedict A.Gomes, and Krishna Bharat
Assigned to Google
US Patent Application 20090287698
Published November 19, 2009
Filed July 24, 2009
Abstract
Methods, systems, and apparatus, including computer program products, for linking to an intra-document portion of a target document includes receiving an address for a target document identified by a search engine in response to a query, the target document including query-relevant text that identifies an intra-document portion of the target document, the intra-document portion including the query relevant text. An artificial anchor is generated, the artificial anchor corresponding to the intra-document portion. The artificial anchor has appended the address.
Snippets and Artificial Named Anchors?
Sometimes the description shown in a snippet in search results contains content from a meta description on a page, but it’s just as likely that a snippet may show text that actually appears on a page. We’re told in the patent that one place where searchers may be delivered to on a page is where the snippet has been taken from.
Unfortunately, that may not always be the place where many site owners want visitors to arrive, especially if the text from a snippet comes from something like user-generated content after an article or blog posts on a page, such as comments or reviews.
The patent filing also mentions that it may highlight the query text as shown on the page in some manner, with a formatting change that could be presented a number of different ways such as:
- Underlining,
- Bolding,
- Italicizing,
- Foreground and/or background color changing,
- Font and/or size changing,
- Border drawing,
- Text animating (e.g., “marching red ants”, etc.),
- Aligning,
- Kerning,
- Style editing/adding/removing.
We are also told in the patent that information from the target document might be changed in some way to add, remove, or edit relevant or irrelevant information, such as by:
- Altering or adding a link,
- Adding an image,
- Deleting surrounding aspects,
- Adding new text,
- Adding a popup or hover window,
- Adding and/or executing JavaScript instructions and/or other computing instructions in other languages.
Showing Visitors to Cached Copies of Pages?
The patent filing describes many different approaches to delivering visitors to sections of pages where the query terms they searched with appear on those pages, and one of the alternatives would have visitors arriving at the search engine’s cached copy of a page, scroll down to the appropriate spot, rather than the page itself.
I could possibly see doing that when someone clicks on the “cache” link that often appears with search snippets, rather than visiting the page itself, but the patent filing doesn’t tell us that it would show a cached copy of a page only in that instance.
Conclusion
As a searcher, would you find being delivered to the middle of a page helpful if that’s where the terms that you searched for appeared? I know that I’ve sometimes had to use the search function on my browser to find the words I’ve searched for to find a web page.
Would it disturb you as a designer or site owner? Would it disturb you if visitors arrived at your pages below the fold? If the search engine changed the look and feel of your site to accent where query terms appear on the page? Or if your visitors ended up at Google’s cached copy of your page instead of on your site?
Like you, Bill, I’ve often had to use my browser’s “Find” function to locate the part of a document in which I’m interested, especially with long PDFs. As a searcher, I can’t wait for Google to take me directly there.
When I write web copy, I scatter calls-to-action throughout and I’m often writing independent of the design, so I’m not too concerned about visitors landing below the fold. I’ll be glad they’re getting straight to the content that interests them.
If Google implements this, though, it could spell the end for certain black-hat and gray-hat SEO tactics, like stuffing background-colored keywords at the end of a page or buying off-the-shelf keyword-rich articles that don’t pitch the product.
Hi Anna,
Thank you. The point you raise about including multiple calls to action on a page is a good one. I’m not sure how much harm this would do to the black/gray hat methods you mention. I’m more concerned with the design implications behind it. For instance, if a main objective behind a site is to quickly let visitors see a phone number to generate phone leads, an easy to see phone number at the top of a page might not have the same impact if visitors don’t see it.
I’ve also seen snippets for certain queries created by Google from comments and other user generated content at the bottoms of pages, and those sections of pages may not be the ideal place for visitors to see first on a page. While I think that site owners should be paying attention to their analytics to see what terms bring visitors to their pages anyway, and check to see what snippets appear with those terms (and possible add or rewrite some content in response), that’s not something that a lot of site owners do, and it can be very labor intensive on larger sites.
I don’t think I’d like this as a searcher and I definitely don’t like it as a site owner.
As a searcher I don’t have that hard of a time finding what I need on a page. I just search within the page. It’s pretty easy. With these artificial named anchors we’d be trusting Google even more. It’s not hard to imagine Google taking you to the wrong part of the page. What I think you’d see happening is people clicking through to the page and if what they want isn’t immediately in front of them they’ll be off to the back button.
It’s possible it could make things harder to find since you’re putting more trust in Google and you’re thrown into a page without first having a chance to orient yourself.
As a site owner I absolutely hate it. I don’t think Google has the right to mess with my site and creating new URLs and highlighting parts of my page clearly crosses a line. I guess Google wouldn’t mind if we all started creating pages showing Google results, but without the ads since our readers don’t like the ads.
Hi Steven,
Thank you for your thoughts on this topic. I think we both agree on the points you’ve raised.
I don’t mind searching within a page myself either. I don’t find myself convinced that using artificial name anchors like the patent filing describes is necessarily a good user experience.
One of the reasons that I become interested in search engines and SEO from my early webmaster roots was a concern over something like this patent application describes – where search engineers might have good intentions that could possibly have unintended results. In this case, possibly causing viewers of a page to experience it differently than its designer intended.
The details in the patent filing about transforming the look and feel of content on pages bothered me as well. I suspect those changes would probably be minor, but I’m not sure that it’s a line that Google should cross.
I guess, It is not up to us to decide. If Google wants to implement these changes I think it has every right to do so. Personally I would not mind about having internet users landing on the footer of my web site, if this the part of my web site that actually contains the most useful information. I always believed that the the main purpose of web sites and search engines is to serve in the best way possible users’ information needs. This might be the best way to do it.
Aloha Mr. Bill, in regards to using artificial name anchors for snippets, it still boils down to predicting users search intent which is still haphazard at best. But I donรขโฌโขt necessarily see it as a bad thing, especially if youรขโฌโขre an ecommerce site utilizing tabular page segmentation. If they รขโฌลwere toรขโฌย direct the user to the correct table at the head of the conversion funnel, its possible that it could offer a better user experience and improved conversion or delivery of information, based on the users query of course.
But thatรขโฌโขs what I gleamed from the patent, though you and Dave are much better at dissecting those than this little ole shirt maker is.
I am trying to remember the blog, but the designer had a clever trick that I wanted to figure out. As you scrolled down the article, a menu bar on the left side scrolled down with you. It was unobtrusive, because he used small icons for the navigational elements. I think if designers begin to realize that this may happen, we may see more of such elements. How about a 150px header that stays fixed at the top of the window as the content is moved in the browser?
Hi Stefanos,
I agree with you that it is in the best interest of site owners and search engines to work together to serve information to searchers and site visitors in the best way possible. I’m not so sure that search engines and site owners are always in agreement as to what that best way might be.
Remember the public uproar over Microsoft’s “Smart Tags, in Internet Explorer 6, which would highlight and make clickable words found on any web pages, so that viewers could click on those words and see search results related to them? Microsoft has reintroduced a more subtle variation with their Internet Explorer 8 Accelerators, from which you can highlight content on pages and perform searches or other actions on those. That doesn’t seem to created the same kind of public outcry that smart tags had, which actually transformed the appearance of content on web pages by making them links.
Google may have the ability to make changes like the ones described in the patent application, but there may be some question about their right to do so, or how designers and searchers might react to those. Would a negative reaction from a large percentage of site owners harm Google?
Hi Charles (Mr. Aloha),
Helping people find helpful information at the right places on sites may just be a good thing. But I do think that if Google started delivering people to the middle or bottoms of web pages, we may find a lot of people deciding that they need to redesign their web pages. What changes would you make if Google were to do this?
I’m also concerned about Google using the same or a similar decision process that Google uses to decide which snippets to show to searchers to also decide where to place artificial anchors. I’m not always very happy with the choices that they make in choosing snippets.
Hi Frank,
I’ve seen some sites with the floating menus that you mention. I believe that most of the time I’ve seen those that they’ve been implemented with java script. I have to confess that I’ve found those distracting on most of the pages that I’ve seen them upon, but like you, I could see more sites using them if people started entering pages from search engines at places other than the tops of those pages. A fixed header like you describe might be another possible option we would see more of.
It’s possible that we might see something similar when it comes to serving advertising on sites that rely upon ads as well.
I’m with Steve Bradley on this one. To me it just seems like going beyond what a search engine is meant to do i.e. deliver you to a page of relevant content, not to specific portions of a page or in some way meddling with the content of that page. For SEOs and web developers, it certainly has implications re page structure and navigation. Too big botherish.
“What changes would you make if Google were to do this?” I can honestly say that I am not sure if I would make any changes. Though placement and incorporation of broader calls-to-action should enable you to compensate for it. I am certainly on your team Bill, and I agree that site links and snippets are pretty messy, but given there are no click metrics to measure their impact, its really not worth agonizing over. But over all donรขโฌโขt you think that if Google single handedly interrupted or even corrupted user engagements that the users would defect from Google and not the site?
Hi Bullaman,
When search engines start bringing visitors to places on pages where query terms might appear, are they potentially skipping past information that those visitors should see before they get that content. Most designers are going to assume that a visitor will see the content of a page starting from the top. While the “above the fold” content could include logos, calls to action, advertisements, navigation, and other information that could potentially be important to the designer and site owner, there might be additional information that should be seen as well, that can impact what is seen further down a page.
For example, if I had a page that described and solved a problem with a specific product, and it only affected one particular product, I might make it a point at the top of the page to specify that the solution on my page only applied to that product. If visitors arrived on my page after that warning, and missed it, there may be a potential for harm if they try to apply the solution to a similar but different product.
Google has stated in their September official blog post that they may start delivering people to parts of pages when they see named anchors already on the page, and I think its important for designers to keep that in mind on those pages. It’s less likely that a designer will consider that possibility when Google is the one inserting artificial named anchors into pages.
Hi Charles (Mr. Aloha),
I wonder, given your point about analytics, how we might learn about visitors who arrive at artifical named anchors. Would Google start telling site owners about these visits in Google Analytics, or Google Webmaster Tools? Should they? How would other analytics tools be able to tell us about visitors who arrive in the middle of pages?
Making a good first impression can be important for many sites, especially ecommerce pages. When I search, I’ll often right click pages found in search results and open a number of pages in separate tags, so that I can quickly compare the pages that were returned for my query. If I were to see the middle of a page, with information that didn’t look relevant to what I was looking for, I might not explore the area above where I landed. While the way I search and look at pages might be different than how many other people do, I wonder if people looking at one result at a time might also not scroll up.
Would people defect from Google if the search engine started doing this, or would searchers more likely leave the site they arrived upon, and continue to use Google? I’m not sure. There are only a limited number of search engines that I use, and Google tends to provide results that seem at least as relevant as the others. If I had to guess, I would think it’s more likely that people would continue to use Google.
I don’t have a problem with them doing this, if they do it RIGHT. If they are going to do this, they had better make damn certain that it actually works most of the time.
I think that Google has WAY too much confidence in it’s algorithms. Some of the things the current SERP algo does are already ridiculous – such as assuming that an older domain contains better content (giving it a higher SERP ranking because of domain age).
There is no substitute for human eyes. Google right now does a pretty bad job (for how much credit they get) of listing pages in order of content value right now, at least for anything that isn’t incredibly mainstream. The problem is that Google’s algorithm is designed to figure out what is popular, rather than what is actual, good content. Google just thinks that if something is the most popular, then it must be the best. That logic is truly broken for A LOT of search terms.
I really never like the sounds of any of these changes when they first come in but as a rule they often seem to work quite well but this one is a little scary because its unclear where the line will be drawn.
It looks like there is potential that they might actually be changing the website/page that they are taking you to and I think that this might be overstepping the mark depending on exactly what they change, its where to draw the line.
Bolding text or changing the formatting might seem reasonable but when you really consider this and think about the effort that goes into the design of some sites its a little underhanded. There will no doubt be a robots option to prevent this but even so how many novices bother with robots?
Hi Bill
The web is already competition intense and from my hindsight as a SEO, I feel that it is time we give more importance to quality, highly content. Perhaps what I would do is that produce decent content and have a table with a distinct design at the middle portion of the page (not preferrably at the middle as the patent says) highlighting some of the main points of my articles and that would be the most relevant part of my post/content that would invariably tell to the search engine, the answer to a particular search query. I would go about with this once Google has any intentions of implementing this rather useful/strange algotithm . Bill your thoughts on this! Again excellent post! Kudos!
Personally, I don’t like the idea for some of the reasons mentioned above, the biggest of which, as a site owner and operator, are: someone else manipulating my page in anyway, possibly causing loss of branding, advertising, and overall revenue.
There’s a script I posted Sunday on my website that I believe should keep visitors landing on the top of the page for those who are interested…
Hi Buzzlord,
In the Official Google Post that I linked to above, it’s mentioned that Google might only deliver visitors to actual named anchors that they find on pages some of the time. It’s interesting that the announcement was made by someone on Google’s “Snippet Team.” It’s also interesting that this patent filing mentions that it is possible that delivering visitors to “artificial” named anchors might be related to snippets being displayed as well.
I’m not sure that we can make the statement that Google favors older domains with any certainty. For many queries, it seems like Google seems to favor fresher content, regardless of the age of the domain.
It’s true that Google (and other search engines) may have a bias towards more popular sites and page, but there are many other reasons why great content may not be ranked as well as it can be, including being located on pages and sites that just aren’t very search engine or user friendly.
An hypothetical example of that bias – a scientific journal focuses upon a very narrow niche involving the effect of gravity on light around black holes. It’s filled with many great articles on the topic, but doesn’t attract much in the way of links or visits because of its very specialized nature. A more mainstream journal on Astrophysics published an article for mass consumption by the public on the effect of gravity on light around black holes. The article from the much more highly visited and linked-to Astrophysics journal ranks highly for the term “black holes” and the articles from the little known and highly specialized journal are nowhere to be found in search results.
This kind of bias appears in the kind of citation analysis that PageRank is based upon, and it’s quite possible that this problem also exists in the major commercial search engines.
Hi Jimmy,
I hadn’t tried to find an example of the links to named anchors that was mentioned in the Official Google Blog post, but figured that it might help us think about how links to “artificial” named anchors might work. The W3C pages on HTML elements seemed to be the kind of page that Google might use the named anchor approach on, so I searched for [w3c html colgroup], and I did get the kind of result that Google mentioned in their September blog post:
I actually like the “jump to link” approach that they use for that query. In the context of my search, I think that’s a good result. But that’s for a same page link (named anchor) that the web site owner put on the page himself or herself. I’m going to assume that if Google is the one inserting the artificial named anchors like the patent application describes, we might see something similar.
I don’t like the idea of Google making changes to the way that a web page appears, but I guess that if they did, it might be a very minimal change. When I follow the link within the snippet showing in my screenshot to Column groups: the COLGROUP and COL elements, Google didn’t make any changes to the way that page appears. Will they when they are the ones creating the artifical anchor? I don’t know.รขโฌลฝ
Hi Shameer,
Thank you. I’m beginning to think that if someone wanted Google to insert these artificial links in the middle of their pages, that a suggestion in the Google blog post for getting Google to link to already exiting named anchors might be a start. The one I’m thinking of is:
It’s going to be interesting to see how this is implemented if Google moves forward with it.
Hi Justin,
Thank you. I visited your site, and read your lengthy and thoughtful post (Block or Remove Google’s Artificial Anchor Tag Insertion) about the the script that you developed. I have a lot of the same apprehensions that you do about people arriving at the middle of a page. It’s nice to see someone thinking proactively about what could be a potential problem, and developing a solution.
Hey Bill,
I came across the use of the “Jump to” link in the search snippet back in October and documented it here: http://www.matthewsdiehl.com/google/google-losing-money-with-search-snippet-jump-links/.
I definitely commended Google on thinking about how they could improve usability especially when they are sending users to lengthy pages. However, I took another perspective on how this could negatively impact Google, webmasters and other companies from a revenue standpoint. The “Jump to” link in my example bypassed on-page ads, which as a user doesn’t bother me, but from Google’s standpoint could hurt them as they need AdSense ads to be seen by as many people as possible.
Hi bill,
To be true its not about whether we as a webmaster or site owner like people to visit our pages about half way, I really think if visitors going to directly check out what they are really looking for in a page then why not? Its all about users in the world of web 2.0, so I am not really against anything that will benefit from users point of view and a good webmaster should always provide a better platform for his users. I came across a page where this guy really broken his long article in to distinct parts, so even if a person arrives at the front of his page, they always have an option to move to the section they need, just like what you see in FAQs, from a users point of view I like such pages where I can quickly navigate to the section I like instead of doing a search on the webpage, plus when search engines does the job for you then it makes my job a lot easier so I am completely in favor of it.
Plus you mentioned what if a person stumbles up on a comment area instead of the main content part, if that happens it only comes to prove the probability that actual content part of the site may not have relevant information what we are looking for.
I am looking forward for your response ๐
Regards
Update:
Sorry forgot to add the page url I mentioned in the comment here it is.
http://yoast.com/articles/wordpress-seo/
Hey Bill,
Glad you visited and thanks for the link…
One of the reasons I wrote the script is because I know quite a few people have the same feelings about people landing half way down the page, and I thought it was better to try and provide a way to keep visitors landing at the top of the page than to have one of those annoying floating ad boxes travel with me while I’m trying to read.
All ‘floating ads’ do is annoy me, so hopefully if people don’t like the idea of someone landing half way down the page, whether Google uses a real named anchor or an artificial named anchor in the SERPs, webmasters will consider and use what I wrote or something similar, because IMO it could very well be Google just added an annoyance to otherwise quality websites in the name of convenience…
Hi Bill
I’m in two minds over this one. While we would like to have as much control over how the visitor gets to our sites and how they see it, sometimes the method of search means that they should find a mid page section of a page. Possibly the best example of this is a blog. A blog could have multiple very unrelated posts all being displayed together. A search could return this page for a post somewhere near the bottom. In this case, jumping to the post would be the best result. How often I’ve found a page and actually had to do a search to find what I was looking for. I honestly believe that the average searcher isn’t going to search on-page, but simply hit that back button. Potential visitor gone!
Obviously there are numerous ways that this can go wrong, just look at how poor the snippets are at times, but this could be a good thing. I guess we’ll just have to wait and see just how Google plan to use this.
Hi Matthew,
I enjoyed your post, and I think you’re right about the potential for people visiting pages to miss Adsense advertising. I imagine that there are people at Google who may have some of the same thoughts.
Hi seoyourblog,
I’ve been tempted a number of times to include named anchors and different sections in longer posts here. I’m beginning to think that might have the best application when a page covers a number of related but distinct parts (like a FAQ page might). The Yoast page is a good example, too.
I think there are times when delivering people directly to a relevant section of a page is a better user experience, and it’s quite possible that those visitors might scroll upwards as well.
I have seen content from comments show up as snippets on a number of sites, and considering how the snippet team from Google is the one writing about delivering people to named anchors within a page, in the official Google Blog post that I linked to in my post, I’m guessing that they might deliver people to some comments when they might think the comments are good matches for a query. Actually, I expect to see that happen if Google decides to move forward with this. Fortunately for this site, I receive some pretty good comments from visitors – and I expect people to want to read more if they like the comment they see.
Hi Justin,
No problem. I appreciated your comment, and the work you went through to provide a solution. I hope we don’t start seeing more floating ads and menus.
Hi Robert,
I like your blog front page example, with excerpts from multiple blog posts. I think delivering someone to the appropriate post would be a good idea. But I still like visitors arriving at the tops of pages. Guess I’m of two minds about this, too.
This would be a huge improvement, I think. Will put a lot of thngs in order, and, like Anna says, it is sometimes a real pain with long PDFs.
Talking about search engine is always frustrating me. [newbie here..]
Now I don’t even care about how google will send visitors to my webpage.
But as a searcher, I would really like if google send me to the right place somewhere on a page that is relevant to my query. Even if it’s below the fold. But again, sometimes google lead me to the absolutely wrong page where there’s no readable content but a fulset of keywords.
Hi dekguss99,
I agree – better delivering me as a searcher to a useful part of a page, even if it’s below the fold then sending me to a page of meaningless keywords.
With the advances in web 2.0 and all that, I think Google now has enough tools to tastefully highlight the search string without actually forcing your browser to scroll to the required location… I think this would be better from a usability standpoint.
Hi Bill,
I would be great if Google inserts artificial anchors because sometimes it takes a lot of time to reach the desire term on the page.
Has Google implemented this yet?
Hi Max,
Google seems to have made this kind of artificial link available to users of the Google Toolbar through their Quick Scroll feature.