We reached our 400th test for this week and the test is on Solo Structure - linked supporting pages vs non linked page. This is the 2nd test of our Silo Structure Series, the 1st test being a test between the reverse silo structure by Kyle and the non-reverse silo structure. Clint also discussed if a link out to a Google Approved News Site increase rank and which is better, a link out to a .gov or a .org site?
As our usual flow, he checked on some test suggestions and answered some questions that came in from other members. For this week, he also reviews a roofing site in St. Louis - stlroof.com - which is in the number 10 spot. He looks through the site and see what he can find and how the site can be improved.
Tests this week
Test 397: Can a Link Out to a Google Approved News Site Increase Rank?
Test 398: Which Provides More Benefit, Linking Out to a .Gov or to a .Org site?
Test 399: Silo Structure – Reverse vs. Non-Reverse
Test 400: Silo Structure – Linked Supporting Pages vs. Non-Linked
January 31, 2020 Weekly Dossier
Transcription below:
How y'all doin today? If you're good let me know that you have audio. Awesome! Perfect. So, we are in 31 January. Month has gone by quickly. I was laid out for a couple weeks which is horrible but at least I don't have the kung flu virus..
This is the weekly webinar. Welcome! As you all know your participation is appreciated. If you're watching the recording just make sure you come on join live. This day of the week we picked it on purpose because we're not competing with everyone else's webinars, etc. Maybe if there is an overwhelming request to change the day and/or time we might be able to accommodate that.
Just keep in mind when you asked I do other two shows, Tuesdays and Wednesdays, so I don't know if those two days would be optimal. I'm pretty sure not but Mondays, Thursdays are still open. If Mondays and Thursdays work better for you guys and you can get here live let us know so we can accommodate the biggest crowd for the live episodes so I have someone to talk to other than Dori and James and John.
With that being said, let me share my screen. Here we go. So, we have four tests today. Typically, I only do threes but there's two tests that just didn't make any sense presenting them on their own. They're just peas in the pod kind of thing so we're going to show four this week. We have some new test requests.
Obviously, new questions and answers. Some good ones on there today and then we'll round everything out with the site review. If you guys good let me know if you like the site reviews or you want to see particular niches, etc. That will just help guide what we go on with that. Otherwise, I'll just continue to drive on but I don't want to waste anyone's time if I'm doing it in that manner.
Plus, they take about two-three hours to set up for every episode. Again, if you're not getting value let us know and if you – or would like to see some additions to it let us know that as well. That would be perfect.
The test today is 397: Can a link out to Google News site increase ranking? 398 is which provides more linking benefit? The .org or .gov when linking out? And then the two silo tests. These two I was talking about. Just makes sense, peas of the pod kind of thing. You should see them together.
Without further ado let's jump on. 397, this is James Goldsmith's test. Again, he's been doing – like he started with the Wikipedia test and he's done a whole bunch others linking out. The idea is that linking out to authority sites actually helps improve your rankings and/or establishes you in Google so you can meet some requirements, like if you think that a-z thing, these will help meet some of those requirements and prove that you are an authority or that your content is there in the research.
The basis for James' test is to see if one sets of or one type of links that you link out to would provide any more value than another and in this case... I brought up the question with him in our talks in the SEO testing group that we have, is Google News because you have to go through an approval process to get into Google News.
So, then, ergo the idea would be that these are considered more reputable by Google. So, there's only what – probably fifteen hundred sites inside of there? The last time that I saw ~ count. I'm sure there's probably more than that now at this point but you build the site, do all the technical stuff, populate the content consistently and then do the application to be Google News approved.
My idea was that if you're being approved by what appears to be a manual process then perhaps maybe that there in itself gives more trust to that site so linking out to it should be better.
So, he set that up. Google News site linked added and as you see he's chosen #3 like always. Then, they duplicate things here and then the for rest of the pages he added another link. He just found the domain authority according to MOZ of the Google News site that he picked and then he found a similar one for the rest of the sites and he wanted to see which count better.
You see that Google News had dropped the position 4, then it dropped the position 5 and then pretty much stayed at position 5. Until there, went the position 4, 4, 4 and that's it. So, I would call that a wash. I would even say that the drop was significant since it was only one place but I would call this test a wash.
Like, it doesn't make any difference whatsoever if you're using a Google News site. It kind of disappointing to me because if you're looking for – if you're following that EAT concept you're looking for authority sites to link out to so you can prove that you're better and figuring out the definition of what Google algorithm is considering an authority site would be a game changer.
I think James is on the right track to keep going through this process in selecting more and more sites but I think also that once he starts doing the multi-site tests which is essentially these tests with multi-sites you can remove all the questions and answers that some people speculate about whether this work or not. So, it was a good stuff. This is good work for me in my eyes and, hopefully, he continues these more and more.
Again, the results of this is Google News had no advantage or in my opinion a disadvantage because if it only dropped to one place. So, that's pretty cool.
I would probably – if James is listening, what I would do is redo this test but not put any links inside of the control pages just to eliminate the idea that maybe you got lucky or unlucky in this case and picked out some sites that were actually considered more trustworthy. I didn't look around and check this. See these danced around some – not a whole lot.
The rest of them didn't dance around a whole lot. See, this postulate was started at 1, drop to 2 and back at 1 again. It's a reach for me to even suggest it but it'll be probably a good add-on test. Just the Google News that linking out with nothing else. No links going out on the control pages. Maybe that'll make a little bit difference in the test and the results would be interesting to see that.
His other one is the linking out to a .gov versus a .org. Again, you're dealing with relevancy and authority. What does Google think is more authoritative when you're linking out and, hopefully, you get the boost. Same set up. #3 linked out to .gov and then the rest got the .org links. As you can see it dropped all the way to position 5.
So, being a .gov versus a .org in itself doesn't matter. This is another one where I would do another test. I would test just the .gov by itself, no links going out on the control pages to see what it does and then maybe even reverse that and do one with the .org, no links going out. See what it does and maybe one to one where you have 10 control pages, a .gov and a .org test page each and see where those land in regards to your control pages.
This is a good idea especially considering like in the rehab market Google manually curates or was – I haven't checked recently, but they were manually curating the #1 spot for a lot of terms and they were putting in that .gov site.
Then, someone got slick over there and found someone at Google when they actually put in affiliate sites got called out. Switched it back to that .gov site, switched it back to an affiliate site, got called out. Switched it back to the .gov site. Not related but it seems to me that Google would think that a .gov would be more authoritative than an organization. But then again what type of organization are you talking about here?
This is an interesting test and this is one that would do really well when you're switching over into outside of the fake keyword testing and into a local competition, maybe even a longtail terms and checking the value of that too. Because if you're thinking about entities and LSI and TF-IDF maybe that plays into the selection with the types of site that usually count. I'm not sure. Completely guessing at this point right there but that's a great question to be asked for continuation testing for James. He's doing a great job.
This is Dori's. These are good. So, one question that we had. The reverse silo, as Kyle Roof calls it, and then the non-reverse silo which is, I guess, probably my version. Then, Dori thought, "Hey, let's check out one or the other. Let's see how it goes." I think most people typically do this and then Kyle Roof's is he links back to the target page.
I was always of the belief, you know, why would I do that? This is the page that I'm ranking why would I want to send this back. He's got ~ to say that this process works for him which is awesome. So, we wanted to test it out, which one was stronger because if this one is stronger then why not. It's adding one more link back so no big deal, right? We could try it out and push it in and increase rankings.
So, they were both set up in that process on these two pages. Silo A and silo B. As you can see, silo A 1. This silo 1 versus this silo in the single one-off test that one. I will say that as we were watching the rankings, because Dori had this one sitting a while and then seeing what was going on, A and B bounced back and forth.
Then, finally, silo A just locked in at #1 and never moved again, which in and of itself is pretty interesting. You're looking at indexing and other stuff like that. How was it affecting, the way the silo pages... were they still indexed and some drop out or not. I'm not sure that that was looked at. Could have justified that but then if that was the case I would expect to see one of the silos take out that first position and then the other ones dropped it like three, four, or five, right?
In this case, that didn't happened. They both locked in, danced around for a while and then silo A ~. This is not to say that this one is better versus the other one. I'm not too sure but from my perspective if I'm sending all the juice and I want to rank this page why would I send it back into here.
Obviously, when I create my silo pages – I'm not sure how Kyle does it, but when I do mine I pick topics that are supporting and then also have traffic and search volume of their own and I try to rank these and I build links to these to push that up to increase ranking and traffic in these pages as well which in turn carries off into this one.
So, if that doesn't make any sense to you basically what I'm doing is I build this page, I build this silo and then I focus on ranking these. I want to get these up in the top ten and I'm building backlinks to all these and once they get up into the top ten, inevitably, what happens is this target page comes up all by itself without any backlink building on its own aside from like the occasional shares or the IFTTT networks if I have those set up or occasionally if this is a like a hard term I know I'm going to need to build some backlinks but typically this stuff works for local right out of the gate. Just like that.
So, if I'm sending all that juice intentionally in between all of those into that target page I don't see the value of sending that back. Again, I would test this out on your own, test it out in multiple sites and just see which one, at the end of the day, wins.
Also, test it out in the real-world environment too against your competition. Maybe one term is looking for that target back to close the loop there versus the other one that isn't. It's Google. It's an algorithm. It's Math.
So, you could play with it but I like silo A. It's what I always used. It's really up to you if you see the value of sending the juice back. Knock yourself out and just complete the circle of life. But to me, I think I'm going to go on the dark side here and say that all things go to the emperor. Yes, I get Star Wars in my head.
Any questions about that? I should've asked about the other ones too. I didn't asked about the other two tests if you guys have questions. No question. Good, alright.
The next one is linked supporting pages versus not linked. There's two silos. You can see that the setup got a little bit janky here right after the beginning but, at the end of day, it didn't really hurt too much. So, basically, what was tested this time is I have all these supporting pages, if I link them together or if I don't which one of those actually gives more push for the target page. This is a really good one too. It is like, am I wasting my time?
Because, like, this one says linking to this one to this one to this one to this one, right? But I linked mine all together. So, this one links to all four of these. This one links to the other four and they all link to the target page. That's how I do it. So, this test shows – is designed really, am I wasting my time? Looking at it, silo A kind of moved up just one.
So, when you look at the janky the way this started it only went up one. Then, silo B went up a lot more. So, there could be an implication that there's more power in there and Google like this page better. Then, there's more power in the site B page or for whatever reason the silo was more powerful but it wasn't powerful enough to kick it up in one. It's an interesting idea.
To me, this one needs to be redone. But looking at the test results as the way that they're presented this didn't provide any more value than doing this so maybe we are wasting our time placing those together. That'd be a fun test to do in the real world too and I think it would be interesting because interlinking them, to me, is if, let's say, I do some backlinks here and I do backlinks to all of these and page 1 got better backlinks than page 2.
But if they're linking together all that juices flowed around between the whole collection and they're in so much as now all these pages are sending equal amount of power all at the same time versus this one sending more power than this one versus sending more power than this one.
So, what you miss out on is which one of these did you really need and which one do you don't. If, like, this page got some weak backlinks and these four did and this page is the one that's ranking but these four are not do you focus on just these four, now? But all these ones had all their backlinks. These are the ones that have to get backlinks and these are the ones that are ranking this. But this weak one is ranking because you picked a nice long tail easy competition word and you think there were more at this is the issue and really in fact this is your outlier.
So, cool questions with this one, I think. Also, from an SOP standpoint this is way easier to set up than this is. Just looking at it from a workload perspective. So, I'd like to see this redone in the single variable and then we'll probably redo it. I'll get Hunter to redo it. Some testers are working on some other cool stuff with interlinking and I don't want to send them backwards.
We'll get Hunter to rebuild these in the single variable again and then we'll have them rebuild it in the multi-site using Ted's testing tool at the same time to make sure we get that consistent result back and forth. Which is why that tool is so cool. Once Hunter gets a handle on it, him and I can knock out test left and right.
We're looking at, on average -- once we get the process down we're going to be probably starting to at least two tests each day. More depending on what's going on. Like, Mondays, Thursdays or Fridays there's a lot of sitting around time for me. I'm doing link building and waiting and Tuesdays and Wednesdays is those two shows and Fridays is this show but I can set up probably four or five tests a day.
Plus, we get members like Carolyn Holtzman and Jesper and John that are sending in test ideas. So, as long as we have something to test we can do it all day long.
I'm looking forward to this, seeing the variations that come out of this. We can test different silo structures and see if we can find the best and then test them all using Ted's tool against each other and stack them into orders too. So, that'd be awesome. A lot of implications in this test and a lot of further information would be good.
Anyone have any more questions on or have any questions about this one? No questions. So, two new test requests this week. This one's going to be pretty quick to set up.
Do Twitter mentions help rankings? This is one of the member's. He pointed it out and did his setup in the Slack group. So, if you guys aren't in the Slack you need to check that out. But he was able to create a Twitter mention and then actually rank that. It ranked almost instantly for him. ~ again and those from dofollow.
The Twitter mentions are dofollow. So, I'm going to set up the single variable tests for that and then the multi-site test and a couple different ways. First, we're going to test do they work? Do they move the needle at all? That'd be the single variable with a fake keyword and then we'll confirm it with the multi-site test.
On top of that Peter Drew has a piece of software that's out. It's a tool you might as well see if that one works. I'll fire that up and shoot that. Add some test pages and see what happens if we scale that into a process that people can replicate pretty easily.
Twitter mentions in themselves not really blackhat either so if you don't know the process, basically, you do a tweet, you create a Twitter mention, you bring that tweet into the mentions and then that turns into a dofollow link to the target page that you set the link to. Pretty simple.
You can have as many Twitter mentions as you want and if you have a thousand black Twitter accounts you can go nuts if you have the right software. I've not played with Twitter mentions in quite some time but the tests excites me and, frankly, it's pretty easy to set up. So, we're going to fire off that test. I wanted it done yesterday but I'll probably get it done today. Let's see.
Then, the other one is title case versus sentence case. I've saw this one and that's a pretty good point. If you don't know what he's talking about the title case is the title and every word is capitalized versus sentence case where every word is not. Just the lead sentences and it's closed out with punctuation but in a title tag you don't do the punctuation.
I'm sure you could especially if you're doing like questions content or how-to content or whatever use that punctuation and close it out. This is going to be a pretty cool test I think. You set that up in a single variable and multi-sites and then see where it goes.
I want also to test – this one would be a little bit harder, can you improve your CTR by switching from one to the other? So, that would be an interesting test to see especially, again, for those informational terms such as how-tos, what is, kind of questions that you could target for informational traffic. So, this would be a cool test too and pretty easy to set up I think. I'll probably get Hunter working on this one right away.
Let's get into the questions and answers. This is a good one. This one I like and the reason why I like it is because there's a lot of talk these days about 301s and Google are saying you can send so many 301s, you can't send as many 301s and when does the juice stop and how many redirects is too many redirects i.e. a redirect train, etc. In this case, almost two weeks ago he bought old domain and made it 301 to his site.
The domain has Wikipedia links and a lot of authority links which is awesome. This is one that I look forward to. Redirects fine and set to "This address moving" section. Everything fine but he does not see the effect on the SERPs and it does not see links Ahrefs or similar tool in search console. Is this normal?
So, let me get the pen out so you can see where I'm referring to when we're talking about parts of this question here.
1) Is this normal? Yes. Ahrefs and similar tools, the response in whether you see those links kick in there is really up to you – when and if they crawl the sites, where the backlinks are. There's nothing you can do about it. You can't force it and just because they don't find it doesn't mean that the link's not there. I'll show you another good example of that in the site audit. These two I don't even worry about these when you're doing 301. If you're doing 301s, these tools, "I don't see them in there yet." That doesn't mean anything.
This one, "I did not see any effect on the SERP". My guess, by reading this question, is he set up the redirect, made sure everything was working and then link to the pages he wanted to. All good steps. What I don't see in here is that him saying I submitted the backlinks to this site, the domain that they bought to an indexer or a crawler in order to get them recrawled so that Google would see my new backlink that I got now through the 301 on these old backlinks here and give my new site – my new page, this one right here, the credit for those old backlinks.
Hopefully, that makes sense. I know there's a little – like listening than seeing in this version is a little harder but if you buy these domains and you do 301 and you just do the redirect you're looking at 60 and 90 days, maybe even six months, before you'll see the true effect of them if you just put them and let them sit. That's assuming that the links that you got really are authority links and Google is crawling those sites at a regular basis.
In most cases, though, they're not crawling them at a regular basis. So, you're looking 60-90 days up to six months to see an effect from a 301. But if you use a tool like an indexer, like ~ Omega Index here that's out now and I love that thing. That's great. That's one cent a link or Colinkri to force recrawl or building some new links to these links. It kicks that in a little bit better and you can see the results a lot faster.
So, when I see two weeks to me is indication that that part of this whole process wasn't done. The member needs to go back and get all those links recrawled and stuff and then you'll start seeing that juice pass. Also keep in mind, if you build links like I've seen testing now, I'm doing the SEO AutoPilot test that I told all of you about before, I'm doing those now and I'm seeing it.
All the links that I built with anchor text dropped the test pages down and I'm expecting a three week wait turnaround time before I see the benefit of those backlinks. Two weeks one is not enough if you're following that model at all anyway. So, you still have some time to push off and do the rest of those. Hopefully, I explained that clear enough if I didn't drop a question or statement inside of the questions box real quick and let me know.
Does anyone have a general formula/spreadsheet for forecasting organic clients traffics and keywords?
Maybe you've heard me say this before but I'm a consummate ~. I like to get, take courses. I like to buy courses and I take ideas from xyz and push that all together and see what's going on. I do that with spreadsheets a lot of the time. This is an old spreadsheet. This is my old brand that was on there but this is like in the view that I questioned and you can see what I did. I put the keyword and use the keyword volumes and I get the percentage of on-page conversions to call.
The average percentage is going to be an appointment and the average percentage is going to be a close and then apply the marketing costs and the average value of the customer. So, if this customer if my average cost of what I'm targeting is $22,000, that's my average sale price, then this traffic on average based off of these numbers bring me a $24,000. That's looking at I get three calls.
My conversion rate is only six calls out of this here. I'm only going to get appointments with three of those and I'm only going to close one. As you know, that is pretty too... One out of three at $22,000. How you can do that all day long? But if you're talking about a lower end customer where this value is, let's say, at $200 this gets a little bit more expensive especially if you're charging $1,500 a month.
But looking at your keyword selection and your close rates and everything you're saying you're only getting a hundred dollars a month. This is a good way to do that.
Another way is you can plug it in the same thing and change the numbers and your capture rates and stuff and illustrate it this way. You can look at your monthly search volume from all of this. The estimated traffic capture that you're going to get. Roughly 25% if you get page 1 plus or minus at any... I like to always lowball these. It was 25%.
Nowadays, we're looking more like a 15% just a low bar. We don't want to set expectations high and then coming up short you want set the expectations a little bit lower. So, you want to do that way.
Then, you can look at the conversions and play with all these numbers and turn out a pretty good estimated return of investment for your clients. That's just another way to do it. I'll look at and see if I can pull these back out again and then I'll share them in the Slack group with you.
But, again, if you look at it I get one, two, three different versions of it that all work pretty much the same. It's the same process and those helped a lot with sales. So, that's how you do that.
Question from Jesper. So, what you're seeing is some sort of Google dance for SEO out of public tests?
I'm seeing it in all of my link building. Specifically, it's being illustrated because we have Sweet and what Sweet is going every day and checking the search results. I can literally watch as he's updating that information. I can literally watch and see where it is bouncing and stuff.
All of the ones that have anchor text, I'll tell you right now, that every link that I built that has an anchor text in a contextual manner, a contextual backlink every one of those drops out of the index. Like, the page is noindex and then they all started coming back and slowly they're coming up to their positions and I've seen that on a daily basis with a link building, especially contextual link building.
Well, specifically contextual because that's usually mostly all I do for clients. I see that all the time. So, we'll build a set of links, wait three weeks, the links kick in the effect. When we're not there yet we build more links. Rinse, repeat. That's just being illustrated really nicely for me inside of this test, just using the SEO AutoPilottool because I wanted to test those different modules that are inside of there.
When I finish the write up like that I'm going to have to figure out how to present that data in such a way that you can see that play and the time that is taken to do for Google to give you the juice from your new backlinks. Also, I'm going to repeat that test but I'm going to use the SIA network that you guys should have access to depending on your membership plans.
There's a link building tool inside of the SIA section so you can use some of dori's networks. She's also getting ready to release a new network to you for link building so you guys be ready to to jump on that.
Let's see, questions. Last one. Running POP (Page Optimizer Pro) for just a competitor above you, if you're positioned 2 and they are position 1. So, there's two schools of thought here. 10 running pop to get to page or position 1 versus position 2 there's absolutely nothing wrong with it especially if you're like an agency plan and you get unlimited credits then why not.
Maybe there's something in there that will kick in to 1. Maybe you missed something. I'm going to go on a limb and say, I highly doubt it. There's probably not going to be something in that on-page that's going to kick you from 2 to 1 unless you're missing something glaring like somehow you got away with not having the keyword in your h1 tag or something like that.
What I would do is start... Once I'm in the top 3 it's time to look at my CTR and I want to get the natural clicks and if I get more natural clicks by having a better title, rich snippets and a better meta description than my competition Google's going to reward me with that #1 position off the back. That's a much easier process to manipulate by testing your title tags versus – comparing one versus the other. More than likely you're also dealing with backlinks.
So, if that guy in 1 has better backlinks to you but you're #2 and you get better CTR Google's going to reward you with that #1 position because you're proving to be the better in everything anyway. So, that's where I would do that. Now, probably I can guarantee that I wouldn't waste my time running a pop or even Cora if I'm #2 and they're #1. Unless I really really want it really bad. It's likely I mess something up but I can't say that I would miss something up that significantly where it's going to have that much effect.
All right, so that's it for the questions and answers. You have any more insights or anything you like to ask? Well, you have me here. We're rocking right along at twenty minutes left to do the site audit. I don't see anything so let's go on a site audit.
So, there's a new post on Raven tools about SEO for roofers and I am may be having a lack of imagination. So, I search for roofing St. Louis and got stlroof.com. #10 in here so it's pretty cool. We're going to go through and see what we can find and maybe help them rank.
As always, what is it? What's it about? What's in it for me? So, I search for that term I've came here: roofing and exteriors and then I know I'm in the right place. St. Louis roofing. Good. Right there is my city and I get the exteriors. What do you mean exteriors? Then, you look at this content and it's not – until you get down here, it's new roof installs, roof repairs, home exteriors and stuff.
So, what is it? What's it about? What's in it for me? I'm roofing, roofing. I'm good but I need a new roof. ~ new roofing and you ~ . You had to scroll around and get that information. This section right here, I will actually move up and put it up here.
Let's look at their h tags and stuff, see how that's built out. I got two h1s which is interesting. The phone number is wrapped in h1 and there's no keyword inside of the h1 tag. So, that needs to be fixed. H2s, there's roofer, roofing. That's good.
I'll get rid of this one, roofing that's good. I'm going to keep it. There's no h3s and h4s. There's a missed opportunity there. You can get those. The canonical tag is not set, leaving them open for negative SEO. If you've been to the SEO rockstars you know the negative canonical SEO trick that I've shown before at previous SEO rockstars.
That stuff works like we implemented that the other day for another reputation management client. This is the #1 reason why it works so well if you don't have that canonical tag set properly. It works if you do but it works a lot even better if you don't. It makes it way easier so set that even on your home pages. Title tags says a little long - we call you back.
This kind of blah and this is overdone, in my opinion. You have St. Louis you don't need this too. This call to action needs to be fixed now. There's no phone number in here. Clearly, their primary call to action on their website is getting a phone call so put the phone call in here. Put the phone number in here.
Who cares if they go to your website if you still get the phone call, right? So, put up the phone call and your call to action inside of here and then it'll increase the value of the website for you, in general.
Surfer SEO is little... This is something that happens with all the tools and I wanted to point this out. This is to remove your exact match terms in here and then at the same time it tells you to increase your keywords. So, you're actually dropping the page density of your content out. I have two keyword density test to show you that Google is actually counting keyword density when it's ranking. It's a duplicate of two test that Ted Kubaitis did.
We're just retesting it, make sure that the results are still the same but I wholeheartedly expect those to go up just like his did where it was zero to 90 and 90% keyword density ranked #1 across all his test pages. So, if this tool is telling me to remove keywords and stuff, or any tool for that matter, don't do it. Especially, if it is telling you at the same time to add keywords or add terms or add words in paragraphs.
In this case, it's words into paragraphs. You can see here the h2s, the h6s. The tool just ~ looking at my experience told me that was going to happen. It was saying you need to do it and then using the strong tag inside of the content on this page. We get 31 to 136 seems a bit excessive to me and that's probably coming from the fact that there's some yelp, angie's list and other stuff inside of there.
So, common sense would dictate that I won't do that. It picked out the exact keyword in title, exact keyword in h1 and the h1 is not there. There's plenty of opportunity to flip the inverse inside of that title tag if they wanted to do that. In particular, that h1 needs to be addressed since there's no variation of it whatsoever.
Then, this speed. It's the speed test here I think they're using a lighthouse. Not exactly the best speed test tool. You can see functions and stuff but the end results you need to be using a webpage test. I would always confirm ~.
We also look at Page Optimizer Pro just to be fair and it pretty much the same thing. Meta title on the h1. Paragraph. Again, this one is also telling you to reduce... I always never reduce so this can give me a complete bunch of reduce, reduce, reduce. If it's all reduce then I would do it but in this case I wouldn't. I don't reduce at all and then on top of that I want you to reduce and increase variations.
So, what I would do is just go back through here and add variations and I would not reduce anything. So, it is the standard practice. I never take away from a website. I only had two unless I've blown it out and I just don't know what's going on anymore.
As a user, if you like this user interface that's awesome. They had some checklists and stuff. It's handy, I guess, but typically I always do this inside of the excel format. I'm not really a fan of this. I didn't do the ~ or the Google entities. I just want to save my credits for client stuff.
There is an opportunity there for you in pop and inside a surfer if you want to do it. You can also look at the same data inside the Cora tool. But what I would do here is look at the basic tunings: meta tags where I looked at title tags we addressed in headings. All those other tools pick that stuff up and then the intermediate tunings: keyword frequency.
As you can see, the other tools are saying take it out but the keyword frequency is actually higher in your competition so you need to be adding more into it. Then, some of these trust factors need to be addressed as well. I wouldn't really mess around with the amp but the open graph stuff that I would do and then social integration just having those out there. That goes along with your backlinks too. Just having a branded network set up. So, that's pretty good.
Then, the off-page stuff. You can see the backlinks are not too bad so it's really on-page and you kick in a little bit of an on-page work and fix that up it'll probably pop you into the top five. Top three would be my guess especially because you can see that page is at the top 10 already.
Then, I want to look at the tools here real quick so we can stay at the technical before I go into the backlinks. But, again, looking at response codes sort this the opposite way - highest to lowest, you see all of the pages are doing good. There's some cdn stuff in here.
It is kind of weird and the way that's set up, especially, considering how slow some of the testing tools are saying the site is but most of that – like, their images and stuff and this is a wix site. I think, there's definitely some work to do there. Probably not a lot they can do with wix.
Then, if you see the canonical this tool picked it up a little bit. They look kind of in place. I would look at that again and make sure that those are working right or at least set up inside of your optimization.
Rarely, the overview here is showing us that their SEOs and their on-page technical stuff was actually doing pretty good. Site structure is pretty narrow which is awesome especially for local. Then, you look at some of the stuff that they're doing. There's like some pages don't have any outgoing links and, in this case, it's only one but that's pretty much like par for course for all of their on-page stuff.
It is like one or two pages has an issue here but overwhelmingly though its URLs without or duplicate h1s or without h1. My guess is, if we look a little bit deeper and if you remember that phone number was all h1 that's where the duplicate h1s are probably coming from. So, something to look at and remedy and fix.
Let's look at the backlinks. Here's when I was talking about one service over and over when we run that 301 thing. Don't trust any one service over another and don't expect them to get all of your backlinks. Here's a good example. You see here the domain has got seven referring domains. But if you have Ahrefs, they actually have 39, they Ahrefs found.
So, you have to take all these tools with the grain of salt and, again, you can't force any one particular service to find more backlinks over another one. It's just where there bots crawling and they're not going to find it ~.
This is the pattern that I normally see between the two. Ahrefs finds more than Majestic but Ahrefs will also find more the spammy sites than Majestic well. So, if you're looking at it from a negative perspective you want to see if you have a little bunch of spam Ahrefs probably a better tool to do it. If you're looking for a lot of the better backlinks – I won't necessarily see it better.
I think it's the wrong term but the better sites these guys are focusing on sites that they find are better. They're Majestic million or whatever and carrying it off and ~ based off of those. So, that leads to less backlinks most of the time, in my opinion, on the Majestic side than the Ahrefs side. But we have seen, even in our site audit, the subtle difference between the both.
So, I typically look at anchors. URL is good. These are wasted backlinks. This "click here", these are generics. These are wasted backlinks, in my opinion. There's no brand and there's one exact match. So, I would build a few more of these exact match and then I would also build some more brand stuff to keep that ratio down just a little bit.
Obviously, you're going to do anchor text ratio analysis of the top 10 to see what Google is looking for. I have a spreadsheet that I do that. I get the software for that that's going out to all you guys sia. We're having issues with the licensing service. So, will talk to dori. We were probably going to strip that out. Because it's mine, I paid for it.
Dori's helped me out with it because she's done software stuff way longer and better than I have. But we're going to probably strip out the licensing of original one and just set it off to everyone because I've been working on it forever and kind of get frustrated with that project.
But all I have to say is that you should analyze the anchor text and see what the percentages are based off of classifications: URL versus brand versus exact match versus partial match, and I have like 20 of them. Then, you see what the pattern is. What Google is looking for and then meet and beat based off of that information versus just building willy-nilly websites and abouts and click here to your websites and stuff. I don't even build these anymore. Those are a waste of time. So, that's where that process is going.
Let's looks at this one too just because there's different site links here. Empty anchor text. These are pretty cool. I see a lot of these and that's typically images that have those. Let's see, URLs there. No brand, no exact match, click to go to the website, blah. Those are boring, boring and absolutely no value whatsoever. There's a link.
Let me see how it's up, see if Majestic is right. It's nofollow, external, no opener, target blank, no anchor text. Interesting. I want to say that I've done some site cleanups recently and I've seen a lot of this. This best-rated site. It's pretty interesting but I will point out that don't get excited that you see it on there because a lot of the tools are downgrading this site and calling it bad.
I'm not sure where they're getting their information from but I think they're trying to do something like a home advisor, yelp kind of model but they're doing something in the backend. This is the link research tools, cognitive, privacy, SEMRush are all rating.
The site is bad so if you see it use your best judgment in whether you want to remove it or not. For right now, we're keeping it but we're definitely keeping it ~.
Question. Why do you think click here type of anchor texts are a waste of time?
Because it's generic. It typically goes with call to actions. It's typically visit this website, etc. So, you're doing that call to action and you go and that's the idea behind it. But if you look at normal linking inside of context, unless it's an ad or something, it is going to be a word. An anchor text, in this case, let's say roofing in seattle or roofing in St. Louis versus click here or this is my roofer.
My roofer is way better than click to this website to check out my roofer because you get that relevance at the same time and you're passing the relevance that you need along with the link versus trying to get the the click here, click the website and stuff. A lot of those SEO tools and link building tools have generic set - click here or websites and all that stuff. If you do disavow of all of those...
And I've seen it like time and time again, but I've done disavows of all of those just playing around and because most of those end up being the bad links anyway and the rankings went up. So, I would definitely... Unless it's like part of that site like for a citations, a lot of those are you have to do the click here or read more or whatever. You're just forced to do it. But don't intentionally go out and build those anymore.
You should be using stuff like the variations from Cora, for example. If you don't know what I'm talking about... And I have a script that I got from pablo. There's a SEO ruler now. I think he's got this built into it but I just use this and these are the variations right down here and I use those for my anchor text now. I use exact match, use brand or only use like St. Louis roofing with dori friend present St. Louis roofing or the best roofing company inside St. Louis.
So, it's really long anchor text and stuff and that's carries that off a lot better, in my opinion. I'm using the context and the relevance of the page a lot better and I'm using the power of the backlinks a lot better so I need less now. That's why I think they're wasted time.
All right and that was exactly one hour. I think we were rocking right along. Do you guys have any questions about anything that we talked about at this time?
All right. Since you guys are all members, let's show you some of the tests we got cooking in the pot, get your juices flowing. The SEO auto pilot tests are still fired off and doing well again. Some of these actually kicked it right to page one or position #1 instantly but some of them dropped it down. Doing some keyword placement tests, some image tests, the unlink authority domain inside a test.
All of these are all going to be some good stuff plus the testers have some other stuff that are tracking on their own that gets fired off. I got, like, 15 test ideas that I want to do and then we're also kicking off the retests to make sure that we're presenting you with the latest and greatest information. Then, finally I have the leaderboard test. I'm waiting on the indexing. That one's going to be a pain but we just put a-z, essentially.
This is some of the common html tags that are being used right now. See which one's rank higher. If you've seen the title tag is more important, then the h1 tag is more important, then the h2 tag. That test we're redoing that setting it up and going to be maintaining that consistently and figure out a way to get that to kelly so that it's added to the website and updated on a weekly basis.
Then, probably, what I want to do is figure out a way to... I have the automator. Again, the licensing is messing that up but the automator is running so we're going to start doing a Diff of on-page based off of running Cora and running a Cora Diff and seeing if we can watch patterns and stuff before and after the algo updates.
The hard part of doing that is it takes 10-15 minutes depending on your computer power to run one of those. I can do 100 keywords in 12 hours so we have to pick a hundred keywords that matter most to the members and we can see that Cora Diff and push those back and forth and check some cool stuff and maybe get some insights a little bit faster than other people otherwise would and so much as we can see based off of correlations.
Speaking of correlations or actually I'm also breaking down all the on-page ones in Cora and we're going to retest all of those as well. Maybe we get some testing this show that we're wasting our time with something and we're not spending enough time on another one. So, that would be really cold. That's it so thank you all for showing up. This has been the weekly webinar for January 31st in SIA and we'll see you in Slack group.