This is our first dossier and webinar for 2017! For this webinar, we talk further into detail on the tests for this month's dossier and how you can apply it to your site or your client's sites to help you move rank.
Test 86: Will Google Index Content Hidden By CSS?
Test 87: Mythbusting: Keyword Prominence Is A Ranking Factor
Test 88: Are Favicons A Ranking Factor
Test 89: Can You Increase The Strength Of A Link Through Traffic?
Test 90: Does A Link From A Live Stream Video Send More “Power” Or “Juice” Than A Link From An Uploaded Video?
Test 91: Are Local Search Results Immune From The Duplicate Content Filter?
Test 92: Indexing A Page With Images Only Does Google Know The Topic?
Test 93: Does A .city Domain Beat A .com In Local Search?
Test 94: Does A Link From Relevant Content Count More Than A Link From Unrelated Content?
January 2017 Dossier
Transcription Below:
Will Google Index Content Hidden by CSS?
How this got to me was that my brother does web development and a while back, he showed me a landing page where as you scroll down the page, you follow the line and it shoots from left to right as if you have gone to another page. But if you notice, the URL doesn’t change and it is on the same page.
I started thinking of the content over to the right and I was thinking that Google would see that as the same page but let us confirm it.
What I had him do was set up a test page with an arrow to the right that when you click, slides over. You can notice that the URL does not change. It is on the same page. That is just an action that looks like you are sliding but you are really not. It is all just one page that is laid out left to right rather than top to bottom.
The keyword pagothodens and guyawnloribuys are separate keywords so what we did is to see if they index. The pagothodens indexed just fine and you can see that it is right there in the URL. What is excellent is that the next keyword also indexed on its own.
Interestingly, Google took the h1, because they are both h1s, and put it as the title in the SERPS. You can see that the URL is exactly the same. What was kind of fun was that, from the description, it took from the left hand side, it took from the visible text.
So while the hidden text, if you will, indexed and the h1 showed as the title in the SERPS, the description was actually from the visible text.
Pretty cool results. When you have clients that do not want text on the page but it will ruin their pretty design, this type of thing is an option. I am looking at it in a practical sense. It looks nice, it slides nicely, it is pretty, and it works. What is interesting is if you look at the Google Webmaster Guidelines, it specifically says that you cannot hide text in the CSS.
What is interesting is that when you have something like this – the arrow – it clearly indicates that you can do a little bit better. It clearly indicates that it is something that can be clicked. In my personal opinion, it is something that we can get through the guidelines. If you have to explain, it is not ugly, it is not a miniscule pixel, it is a real design type feature and I think doing something like that will get you through a Google manual review because you are not necessarily hiding it, it is a design element.
So if you are looking at – will Google index this in the first place, the answer is yes. The content is crawled and it indexes on its own and indexes in addition to other things on the page. You can put SEO content if you will and it will index just fine.
Myth busting: Keyword Prominence is a Ranking Factor
I really like this test because this because when you read Backlinko, Bryan Dean’s Top 200 factors, this is always in there. It is the concept of you need to put the keyword in the first paragraph or the first 100 words.
Your keyword in the body has to be up at the top as quick as possible and that is a ranking factor. I like this kind of tests because they are old rules of thumbs, old wives’ tale about SEO and this is what you need to be doing in order to do it right. So let’s see if that is true.
Derek Booth in our group put this test together and what he did to format this is the middle ranking page has the keyword in the first paragraph. The other pages had the keywords added to the third to last paragraph so it is below the fold. All these pages did not have the target keyword in the body until he added it.
Almost immediately, it shot to number one – the page that has the keyword in the first paragraph. When you think about your main keyword for the page, get it in the first paragraph of your content. That is what I have been doing. It is something that you do as a matter of course, as you are writing the copy about a particular thing, you talk about it quickly although you may notice from time to time, for design purposes, you may not get it there.
You may have a little blurb here and some images and the next thing you know, the keyword is not in the body until the lower parts of the page or below the fold. Do what you can to get it above, exactly as you want it, as close to the first paragraph.
Dori: Do you know if it was the first word or if its just in the first paragraph?
In this case, it was the first word. At least get it in the first paragraph.
Are Favicons a Ranking Factor?
The favicon is that little image. If you out of the box create a WordPress site or a Magento site, you will see these two images right at the top. Those are unique images that are put in the browser tab. Where this came from was, there was a whiteboard Friday chat where rand was talking about things that you need to worry about right now in SEO and on page.
Overall, I liked this particular whiteboard Friday but he oddly throws in favicons. In the whiteboard he talks more about conversion factor or maybe trust factor for users and I was actually really surprised about that. I was – what if this guy knows something that we do not know so let us test and see if favicons are really a ranking factor.
In the test, the star is the favicon and in the other pages, they just had the default pages with the dog ear and there was no movement. The thing about that is, if the favicon was something that was crawled by Google, we would have gotten at least a freshness update because we did not do anything else in the pages.
These pages indexed, none of them had a favicon, we added a favicon to the third page, at the very minimum we would have had a freshness update and the page should have moved. We did not get any movement at all, not even a freshness movement. We can safely say that a favicon is not a ranking factor.
Can you Increase the Strength of a Link Through Traffic?
The concept here is that we wanted to fire a link and get movement and once we have movement, we wanted to send traffic to the page that was firing the link, but not click on the link. What we were doing is, we wanted to see if we can increase the juice that is being passed through the link simply by sending traffic to the page, the page becomes more active, the page that is sending the link but nobody is clicking the link.
The reason is that, the patent from Google that talks about what makes one link stronger than another, it really boils down to if a link is more likely to be clicked. The higher likelihood that a link is to be clicked, the stronger it is, according to Google.
If a page is receiving a lot of traffic, it has a higher propensity to be clicked, in a sense, could we then increase the value of this link if it is a page that is receiving more traffic because then, it has a higher change of being clicked. That is the basic concept of what we are shooting for.
In this test, we used Google doc and fired a link. What was hilarious is that ads showed up and I love that when they show up on test pages. I would love to see the AdWords account when they are going through search terms of what triggered the ad and they get to see our fake keywords.
The test page moved from 5 to 4 with the Google Doc link and then we fired traffic at it. Initially it dropped from 4 to 5, which is a general indication that you have hit transitional rank and we see that a lot in test environments, and then the test page moved to number one. Pretty cool. The thing about this is that, we really scrutinize our own test and we really get in and make sure that what we are providing is the actual thing.
One thing that we realized after the fact is that we may not have waited long enough once the page moved from 5 to 4. That initial movement, we may not have waited long enough so that we could officially say a 100% for sure that it is the traffic to that page. So we started rerunning this test, it is going to take a month or two but what we are going to do is to get the movement, get it locked where it locks then we are going to wait for about a month and then, send the traffic.
Actually, we have a 26 day policy that we are working with internally on what we can then feel good about when results are locked in. So we wait that 26 days and then fire the traffic and then confirm it. We feel good but we want to work on it a little bit more. Nevertheless, we like the initial result and to be honest, it can’t hurt. If you are sending traffic to a page that you have a link on, it can’t hurt things.
Dori: Actually, that is more stealth, it is like a whole different two tier level of strategy.
Actually when you get a really nice link, don’t you get links to that and send traffic?
Dori: The common notion is to send links to it, not traffic.
I would send traffic, even just crowd searching traffic or if you really want to be less black, send Facebook traffic. Send ads to it. If you want to be really sneaky, what I would do is to build those tier two links, send traffic to those tier two links and then they click through because then it makes that tier two link more legitimate.
Does a Link From a Live Stream Video Send More “Power” or “Juice” than a link from an uploaded video?
There is this concept that live YouTube videos are stronger than uploaded videos and there are tons of software out there that allow you to game your recorded videos so that they look live. From an anecdotal perspective, I think we have all seen that. You do an on air or a live YouTube event and you post it and as soon as you finish, within minutes, it’s on the first page of the SERPs.
You get that freshness update factor to it and you are immediately on page one. It might drop off after a while but you are there. If you have SEO’d it you actually get it to stick or close there, too. At a minimum you get it to stick in the video section on the first page.
So the concept is, if we can do that, does that type of video send more juice or a regularly uploaded video? In this test, we set a Livestream up and an uploaded video and fired links from them into our test environment. The idea of, if the Livestream is stronger, it would leapfrog the test pages receiving the uploaded link.
Carolyn Holzman put this together. She has the video she made via PowerPoint uploaded and the Livestream one. She did the Livestream and immediately uploaded the recorded video and the recorded video actually hit the SERPs first and then the test pages that were receiving the link from the uploaded video ranked immediately above the live stream.
As we were going through, both videos hit and you can see that the recorded one was on top and the Livestream was below. There were no movement between the two – the Livestream video was still below. The uploaded and live video survives up until last week – no movement.
The page that is receiving the link from the live video is not beating the uploaded video. Interestingly, the uploaded video is still beating the Livestream in the video results at the top of the page.
Dori: That is interesting. I heard people sweat that it is the opposite.
Yep, that is what I always thought. So we weren’t necessarily testing that. That was an interesting thing that resulted from doing this test. In general, I would have thought the Livestream would have leapfrogged the uploaded video. It didn’t. It could be because they were not uploaded at the same time so it might not the liveness but the timing of the upload.
Dori: But the link from either one was not significant.
Exactly, that is the bigger thing. There is no edge, if you will, if you create a whole bunch of Livestream and put links on them to your pages. You are not going to gain.
Dori: That is good to know. Do not waste time on that.
Are Local Search Results Immune from the Duplicate Content Filter?
There is this concept if you are making local specific pages- geotagged pages. This could be for every city in your area, neighborhoods in your city, any kind of geo modifier. The idea is you could use duplicate content and Google will give you a free pass so long as you have geomodified your URL. This is the concept that permeates local SEO.
Our thought was that, we can test that. We know that Google has a duplicate content filter. Pages from the same site that are seen as the same content get filtered out. You get the little blue bar that there are other results but they are not being shown to you and you could click here if you want to see them. If the geomodifier gives you a free pass, then we would get past that filter.
As it turned out, that didn’t work at all. This experiment was run by Ryan Hough and did an excellent job. He was put in an identifier for this particular place in Western Australia and they are not passing the duplicate content filter. If you search for a particular area, the page will show.
For this example, fake word and basewater, the basewater page shows. But if you just search for the term on its own, like plumber, they do not pass the duplicate content filter, all of the pages won’t show. What I think this is giving us is a general idea of what duplicate content is. If you are doing a general search, they are seen as duplicate. If you are doing a geomodified search, then the page can show.
If you are going to do this technique, you have to make sure that you are after the geomodified term and that if you want just the term of your own, you are not going to win it unless you have a page for it.
Dori: You can have 51% different.
That is the other thing, do that and then you will be fine.
Dori: If you wanted to show up for Plumber and there are a lot of cities around your area and if you wanted your modifiers to show up for that, that would be something to do.
Then you are good to go. You have the chance to win that geomodified search and the general search.
Indexing A Page with Images Only, Does Google Know the Topic?
This is a Josh Bachynsky inspired test and Derek Booth who is our resident image expert from the gif, jpeg, png test ran this one. What we are doing is within the image, we are putting texts. Where this is coming from is, Josh saw that Google recently purchased technology where they can read images and understand what images are a little bit better.
He was thinking that with that technology, then can read what is inside the text on these images. They can really understand what these things are because they can not only process the image, but they can see the text that is on there and understand what it means. What we did to see if this is true id we put this into test pages and the idea is we will run them through AdWords keyword planner, where if the keyword planner can then suggest terms related to these, then we know they can read the text that is inside the image.
Each of these are on an individual test page. It’s got this image, you can see some gibberish, then the online accounting courses on the image. Google has crawled the page so we know that it’s been there. We put it into the keyword planner to see what it would suggest as keywords for that page and it does not come up with online accounting courses. We tested it with all of them and it wouldn’t come up with any of those.
Right now, in terms of the image, we do not think that Google can index a page based on the image content.
Dori: I wonder if we just put plain accounting or football, will they put suggestions on that? What if we have a picture of people playing football and we put the URL in and would they recognize and suggest football terms. Is this technology even incorporated in the keyword planner?
The technology between the search and the keyword planner does not necessarily have to be one and the same but you would think in these situations where they are trying to figure out what the page is about, I doubt that they would reinvent the wheel. If they have something that could index the page and say what it is about, that seems to be the technology that would be used for he keyword planner.
But yes, the next variation of this would be an image alone of a soccer player or an accountant and see if it can do it based of it.
Does a .city Domain Beat a .com in Local Search?
The .city is one of the new TLDs and we are seeing that the new TLDs aren’t beating anything. I guess if you have a ninja academy, it doesn’t really give you too much of an advantage to go .ninja but what about, if you are doing a local search and you have a page that is specific to a city, can you get an edge against the .com. You have a page that is completely localized and is all about a particular city, can you win that local result, is the concept.
I ran this test from Phoenix because I live there and the pages where created from a Phoenix IP – I just used my regular IP. I have schema on the page for physical address, I had an image from Google – the map showing Phoenix. I got an image of Phoenix – I got this from Google and it was an indexed image so Google knows that it is a Phoenix image. I geotagged it with the longitude and latitude of Phoenix and in the page title and body, Phoenix was used.
What I did was I placed the keyword in and around the word Phoenix city and then I added authority outbound links to the Phoenix Wikipedia page and the Phoenix.gov page. Basically all the things that you could do to say that this page is about Phoenix. Everything was submitted.
One of the things that you run into in this situation is that we are just throwing two domains into the SERPs – a .com and a .city. You are a little bit more susceptible to randomness or factors that you can’t control when you only have two pages going on. We have two homepages, two domains. It is entirely possible that the results can be corrupted that is why we do not just run this page one time. In this situation, we ran it three times, so two sets each time.
In the end, what ended up happening. The .com won 8 out of 9 times and the .city just once.
At the end of the day, I would end up going for the .com. You see those anomalies from time to time where you see a .city is winning, I am sure that it is possible but if you have to make a decision, I would go with the .com.
Dori: This is single variable test results. We are putting this on an isolated situation so we know what is going to move the needle a little bit than the other. We are not saying that you cannot rank a .city and win. Great content, great pages, do everything right, send links to it, that is the kind of thing. You are then competing with other players in a competition on which would create a case study instead of a single variable test which these are which tells us what moves the needle more than anything else or if something moves the needle at all.
Does a Link From Relevant Content Count More Than A Link From Unrelated Content?
Just the content on itself on the page, does it match what you are sending that link to? In a single variable environment where we are just introducing content that is related or not related to where that link is going. Basically everybody will tell you that you need related content. What we got in this situation is we did not get any movement. The link from the related content did not overtake the link from the unrelated content.
In this situation, Caryn Handsun was running this test and she had real words about the topic. She has a page about the topic and she has a link from another page about the topic and then on another page, a link from a totally unrelated page. We could not get the related content page to overtake the unrelated link.
There are other factors when you think about other links coming to a site. They would be passing their juice or relatedness but just based on the content itself, we could not get any movement.
Dori: This is something that I have been doing since 2007 or even 2005 when I was doing megaton link networks. I did not care what the content was, it could be gibberish, I didn’t care, the links were coming and it would work. Ever since then, people has been more conservative. I am not really in the link selling business anymore. I haven’t paid much attention as much aside from listening. I think this is really interesting and I could dive deeper into this again.
With PBNs, you’d want to be linking out to other sites and this is an incredible test to lean on and say I can actually link to different sites from my PBN.
There is this idea that the domain may have a particular topic based on the links that are coming into it and that could be a factor. But if you are really stressing over getting a link from here but the content isn’t exactly right, you may not need to stress on that as much. We haven’t put this one to bed. We want to watch it a little bit more than we were able to by time of publishing but as of this moment, it is still the same.