MayDay - What Are the Quality Signals?

Google is a mess at the moment. Trying to understand what they are doing at this moment will only drive you crazy. What works now might not work ten minutes from now. Crazy s... Google is doing right now. It all started in early March. It's as if they found the formula to keep SEOers, blackhaters, etc. at bay. I never put any significant effort trying to make them happy. I overwhelm them with so much force (lots of content) they have to return something in return. In other words, something always has to stick and it sure works well for me.

What bhjmc said is pretty close to what is going on. The results you see are varied. The other day even I was browsing google images and suddenly I realized it was a WHOLE NEW system.. it was really nice.. images came up in a sort of collage and you mouseover to get the large view.. I was wowed by it and start telling some others to check out the new google images.. only to find out on my very next visit there.. IT'S GONE! No matter what I tried I couldn't get the cool new one back. Random serve.

You have to keep this in mind when doing any kind of competition analysis or keyword research. My trust in the current tools to offer any kind of accuracy is right up their with expecting government to spend less. It won't happen. Rather, consider what you gather from all this data more like bold brush stroles from an artists brush to give you an overview, but not the complete picture.

You best research is your own from your own traffic, using your own analytics, and interpreting the data you recieve in a way that makes sense to your market.

As for what you are seeing TODAY, like bhjmc said, it might not be what you see the next day.. or next week, or the next hour. Google has openly declared their intent to rid the internet/google (to them its the same thing) of mini niche sites. They gave them a chance over the years, and all their data shows that users don't like them. Really, is that such a big surprise? A piece of crap is still a pice of crap no matter what you call it. It smells and nobody wants it on their lawn. Thus the demise of the minisite is a trend you can count on.

Perhaps this helps you understand what you are seeing. I also know nothing about these various keywords you are searching out. They could be absolutely low competition terms in which there is really nothing else relevant enough to match. It's sorta like saying.. 'minisites work, I just put up orangemonkeybuttslippers.com and I am #1!!'

Others are welcome to offer suggestion/observations on this.

Efex1.0 said:

Nice thread MrGemini.
I'm following to read more discussion on this issue. It's interesting to be updated on the Google thinking.

I just have some doubts in these regards.

I've read everywhere in the forum about Mayday, falling rankings etc for one page keyword targeted sites. People is saying they don't work anymore because G outrank them etc etc.

I've been doing some targeted keyword research to launch several of these sites and everywhere I look on my keywords there's always those kind of result on the first page.
I have seen a lot of Xfactor sites on #1, #2, #3, #4.....#7 positions for many many keywords. Sometimes they are just one page long and some times a few more.

Why they haven't been affected by all this new changes on G?

I'd love to hear your opinions on this guys as I'm new to this and those results made me think about.

Click to expand...

Time to put up my minisite that will churn about 1000 relevant pages a week : ) Focus will be on 'brandable themes' on and off SEO : ) Excellent thread BTW.

I'm trying to post a message and the system won't let me

EDIT: I used this message to reply bellow

MisterGemini said:

You best research is your own from your own traffic, using your own analytics, and interpreting the data you recieve in a way that makes sense to your market.

.........

Perhaps this helps you understand what you are seeing. I also know nothing about these various keywords you are searching out. They could be absolutely low competition terms in which there is really nothing else relevant enough to match. It's sorta like saying.. 'minisites work, I just put up orangemonkeybuttslippers.com and I am #1!!'

Click to expand...

Well, but I will need to trust some tool to initialy research and set up a site so then I can analize my traffic and analytics right?

So, the only thing to do now is trial and error?
That's a waiste of time and money

......

If orangemonkeybuttslippers is what people is searching at G then why not?
I'm always looking for high Global search count and low competition...as everybody else I asume right?

Efex1.0 said:

I'm trying to post a message and the system won't let me

EDIT: I used this message to reply bellow



Well, but I will need to trust some tool to initialy research and set up a site so then I can analize my traffic and analytics right?

So, the only thing to do now is trial and error?
That's a waiste of time and money

......

If orangemonkeybuttslippers is what people is searching at G then why not?
I'm always looking for high Global search count and low competition...as everybody else I asume right?

Click to expand...

Need to change your perspective friend. If you go about the whole process and you gather the information and analytics necessary, then you haven't wasted time or money, you have bought yourself a piece of very valuable data.

Here is the thing, even the data that gives you losers, still gives you data towards your winners. You just have to be willing to accept that's the way to go. Business is done this way everyday, only millions and hundreds of millions are spent on what could be total flops. The companies that absorb this cost certainly don't regard it as a loss. They use all that data, and they guard it fervently because they understand its value. The major difference here though, is maybe a month and $50 to $100?

As I said, all those stats are guides. They are by no means exact numbers to base your projections or expectations on.

This thread is great and I learned quite a lot from it which I'm step by step putting in place on our sites. 1 thing that I found through my own testing and not mentioned here, is the Url of the page. I noticed that the pages on my authority sites rank significantly higher when I put them in a long url like "site.com/productname-description-and-offers.html" then when I put them in a short url like "site.com/productname.html".

We are now in the process of nearly changing all of the urls of our products and using 301 redirects to rediredect the site.com/productname.html to the site.com/productname-description-and-offers.html... Results are quite cool so far.

Hope that helps

@Efex1.0
There is a saying "I didn't fail, I just found 10.000 ways, that didn't work.

@Raphael
Are you sure about that?
Till now the right way was "put just keyword in every tag, title or URL, nothing else".

Well it could be true the way you are are saying, normal website don't use just keyword in title and url but ussually something more descriptive.

But, if you now put "description" and "offer" or "buy" keyword into the URL, it is possible that your ranks for "keyword" didn't rise, but you are getting
more traffic, because now you rank for "product offer" and "product description" too? Did you in calculated or measured that?

The biggest thing I have noticed is in one of my oldest sites. It is not a micro site, but more of an authority type site. It has been sitting in #1-2 spots for most of it's keywords (roughly 10-20) for over 4-5 years now.

It dropped yesterday to page 9...and today I gave up looking for it, lol. It's still indexed just not anywhere in the top 20 pages for anything. This site never made too much money as it was focused on a celebrity.

So who took your positions and why?

universalboy said:

The biggest thing I have noticed is in one of my oldest sites. It is not a micro site, but more of an authority type site. It has been sitting in #1-2 spots for most of it's keywords (roughly 10-20) for over 4-5 years now.

It dropped yesterday to page 9...and today I gave up looking for it, lol. It's still indexed just not anywhere in the top 20 pages for anything. This site never made too much money as it was focused on a celebrity.

Click to expand...

From what we have tested with our (authority) sites all our ranking have risen quite significantly for their keywords (some jumped to page 1 and other are moving up) since the change. We have now done this to all sites and see nice results from this 1 change (we do although other changes but this was one that we tested alone). These keywords are all keywords with big traffic volumes. Also notice what Cutts says in the video about Wordpress and how it is naturaly good for SEO because it creates "speaking Urls"...

You can put things like buy or price comparison in the Url to make it more attractive for the reader and get more clicks. Jerry West talks alot about this and it makes stand out for the competition...

I believe as we move forward, this branding theme is going to become more and more prevalent. All of a sudden, the whole keyword match domain thing will be over after google has identified all the sites that have created their own brand, and drop all keyword match domains out in favor of the brands that target those terms.

Click to expand...

It makes sense for exact match domains rank higher. Think about it, if you search for paypal, the most relevant site is paypal.com. If someone else builds a lot of quality links for the keyword paypal and outranks paypal.com, google loses.

polocolto said:

It makes sense for exact match domains rank higher. Think about it, if you search for paypal, the most relevant site is paypal.com. If someone else builds a lot of quality links for the keyword paypal and outranks paypal.com, google loses.

Click to expand...

while this may be true for direct brand searches like PayPal or Coca Cola, it may not be the case for other searches that include the brand. Searches like "PayPal phishing", "Coca Cola calories", or "Best Mercedes Benz car ever" do not necessarily need to have the brand's website at position one in the search engines, especially not when it comes to opinions, reviews or ratings. (unless they are clever enough to add those pages providing an overview of those on their site).

Brands are important, but there is no reason why Amazon or other popular shopping destinations should outrank an exact keyword domain for a keyword for instance.

There is one reason. If your domain itself is not a brand.

While this might not be the case currently, I firmly believe this is the direction Google is headed. They have been pointing the way for years now, and this year they have come even closer to that ultimate goal.. ranks for the brands and to hell with everything else.

There is a simple reason, the so call trustrank really factors in more along the lines of moneyrank. A branded site is a real operation. Has resources and a level of reliability, accountability, and dependability which Google so wants to deliver in their results for numerous reasons. You don't think the way the Internet works now is going to last forever do you? You don't think one day there is going to be legal ramifications for Googles index and what results it delivers? I don't know, but you better believe Google is not going to risk getting sued by their shareholders for not seeing a potential threat to the business model.

Besides these reasons though, putting trust in a branded domain make a whole world of sense from every quality perspective. Think about the world as it is now. Look at the top search terms and sites related to them. Are they keywords? No, they are brands.

Thus this is why, I seem to be the only one standing against the current saying.. don't buy keyword match domains... buy a brandable domain. If you are looking to turn out a peice of junk that will last a year or two at most and you don't really have plans to really build something of it, then sure, register your keyword. However if you want to go long term 3 + years on whatever you are creating, then brand it.

As per your competition with the keyword match domain. The hierarchy of your site navigation should reflect the importance of your documentation. Thus to beat out a match keyword domain, you simply create a matching document that is top level with all the necessary elements that include your url, title, description, and H tags. Not to mention other minor elements like your html coding. Then of course you make your internal linking reflect the keyword of that URL. Presto, you are in business. If this doesn't work, then I don't know what I am looking at everyday where I see exact matches showing mostly in the #5 position, and everything else taking the positions above that. Occationally an exact match domain does show up higher, but only because it had more of the other elements necessary to put it there, NOT because of its domain name.

If BHW is still around in a few years, or maybe even sooner, you can come back to this post thanking me, or screaming up and down over your keyword domain that disappeared like a zit off the face of Google.

radicalseo2 said:

while this may be true for direct brand searches like PayPal or Coca Cola, it may not be the case for other searches that include the brand. Searches like "PayPal phishing", "Coca Cola calories", or "Best Mercedes Benz car ever" do not necessarily need to have the brand's website at position one in the search engines, especially not when it comes to opinions, reviews or ratings. (unless they are clever enough to add those pages providing an overview of those on their site).

Brands are important, but there is no reason why Amazon or other popular shopping destinations should outrank an exact keyword domain for a keyword for instance.

Click to expand...

@MisterGemini

I believe you are quite correct about the direction that Google is going with respect to how it treats branded vs. non-branded sites, and the fact that the internet is 'not going to work like this forever' - but I respectfully believe you may be off base about keyword domains. Let's not throw the baby out with the bathwater; at least not yet.

I think that branding is so much more than a unique non-generic keyword. Things like you mentioned: brands = resources, accountability, reliability, dependability, etc.... But what are the signals that trustworthy brands send? What are the signals that Google picks up on that tells them: "Hey, this site is a trustworthy brand"?

Some of the biggest brands in the world are generic words/domains - but they have been actively branded:


While it is clear that many more huge brands have a more unique and 'branded' name, image, and domain name - I don't believe that having a generic keyword domain will prevent you from branding your site.

This is the key, though. In order to have a branded site - you have to actively brand it.

How do you brand it? Well you have to do all of the things you mentioned, and more.

You have to be accountable, dependable, etc. But these things take time. These things are actually the result of other actions (which potentially signal to Google that you're site is a brand). So, what are the actions that signal resources, accountability, dependability... to Google?

You can start by doing all of the things that a brand does:

1. Have a creative and unique logo and site design.
2. Have a way (or even better - more than one way) for customers/site visitors to contact you: email, contact form, facebook, twitter, phone number, address.
3. Generate buzz; get people talking about you. One huge quality signal could be if people are talking about your brand/site in the text of conversations, posts, or articles:


4. Be mentioned in news media, press releases, and blogs social media, and websites - basically have a wide diversity in your source of links and 'buzz'.
5. Use high-quality vendors and suppliers like big brands do: Dedicated servers from top-tier hosts like Rackspace, use Markmonitor domain registrar, Akamai CDN, etc.
6. Trademark your brand name, domain name, and/or your logo.
7. Etc....

Most of the biggest, best and most valuable generic keyword domains have been monopolized by domainers, have parking pages on them, and have had their value essentially wasted, but this is beginning to change. Domain parking doesn't 'work' anymore and most of the big domain portfolio owners are beginning to either develop or sell their best domains. Eventually these killer domains will get developed. For example, just look at some of the domains being auctioned by Moniker.com (another generic keyword domain) at an upcoming domain convention:


With the inherent type-in traffic, SEO benefits, and instant brand recognition (albeit generic) that these domains have -- these generic keyword domains instantly become huge brands the minute they are developed by someone that knows what they're doing.

While many of these domains are out of reach for most folks on BHW, and certainly to folks who are just getting started - the lessons here can apply to everyone.

Just to be clear - I don't have anything against domainers at all. I'm not one to knock a person's hustle - but it is clear that to park a killer domain like Stocks.com or Home.com -- is inefficient and a huge waste.

I guess the moral of my story is no matter what kind of domain you choose - generic-keyword or unique-brandable: you have to actively brand your domain and site. While choosing a generic-keyword domain may give you a boost in the beginning, you cannot get lazy and coast on the SEO benefits your keyword domain gives you -- because this is rapidly changing.

GS

Well put GS. You certainly went into more details regarding the aspects of branding than I did. Excellent contribution. I think the two of us are looking at things from a different point of view in terms of the practical outlook that is being taken with keyword match.

I'll also point out that the brands you noted are not ubiquitous to the terms they occupy. They gave them a whole unique meaning. Apple is not an apple company, but a computer company. Tiffanys a high fashion store chain, and Amazon has nothing to do with forests or hot sexy tribal women.

As you pointed out the ubiquitous domains might be up for sale now, and maybe that gives some people an opportunity to make something out of them, it's just that those domains are going for thousands of dollars, which I would say is unrealistic and out of reach to about 99% of the BHW population. Frankly, while sure, they have a match, I still have to say that the value of the match is far less than most precieve. Apple can get computers, Amazon can get shopping, and a girl name Tiffany must give way to a shopping chain. These were all done by some of the branding methods you mentioned in your post.

Consider this. If a person were to buy a domain called for example Girlsywear.com, and their site was basically nothing but a custom line of skirts, they would rank and outrank based on the merits of their site for every single keyword related to 'skirts', but quite possibly nothing to do with the phrase 'girl wear'. Why? Because they branded themselves as the goto for shirts. They gave themselves a unique label that tells Google and the world 'we are all about skirts', even though their domain says something else.

If this is the case, it is therefore of absolutely no value to keyword match your domain. The value is in branding your site for the specified function that it serves. I think that many people have taken this has a mistaken preception that the domain match is the cause for the ranking factors that some domains enjoy. As you noted, there once was a day when yes, the domainer did work. Not anymore however. It is the BRANDER that dominates the SERPs.

I think what my post was speaking to, is the format which keyword match has taken at large today. It is not like many domains you mentioned which as you pointed out are largely taken, or you must be taken (to the cleaners ) to have ownership of one. It's more like some of these for example:

shoes fashion boots
tennis shoes for girls
converse shoes ladies
children dress shoe
brown dress boots

All of these have a CPC of more than $1. So they are likely to come up in the typical niche search finder tool. Now the problem with some of these is one of three things.

1. You cannot do anything else with these domains. They are totally pigeon holed into one single thing. Any attempts to brand them beyond what they clearly state would be difficult.

2. Do any of these sound anything like a potential company? As far as domains go, no, they actually sound more like spam.

3. Even if you somehow managed to do something with them as far as branding goes and google signals, how will this domain be precieved to the end user. Again, the Internet won't be like this forever. A few more years, and you will be dealing with a whole new generation of "been there done that"s who will be looking for a 'real' brand/company to deal with, and will simply scoff such domains as these. This has actually already started to happen.

This is the type of keyword matching that I say is just not practical.

What you state as potential terms that could be made into brands is something else which in some cases, does lend to branding potential.

The overall point I think being, be a brand, don't be a keyword.

Mister Gemini

I understand the value of brands and all but all things being egual don't you think exact matches will always have an advantage?

If i have the domain shoes.com, i optimize it to rank for shoes and my optimization and links being the same as the second place dsw.com. Why would google not rank me first for the keyword shoes? Why changing the algo that much and risk giving the wrong match to privilege brands. Google can't recognize all companies all over the world.

Yes they can. By using the various signals as was noted in the previous post.

Again, it is not the keyword match domain that will rank, but rather the branding. This is happening all the time. I am more convinced that its not only the future, but the way to go today.

Search 'computers'

Dell.com
Apple.com

Where is computers.com ? Nowhere.

The brands dominate.

Build your brand, you dominate.

polocolto said:

I understand the value of brands and all but all things being egual don't you think exact matches will always have an advantage?

If i have the domain shoes.com, i optimize it to rank for shoes and my optimization and links being the same as the second place dsw.com. Why would google not rank me first for the keyword shoes? Why changing the algo that much and risk giving the wrong match to privilege brands. Google can't recognize all companies all over the world.

Click to expand...

This is some thoughtful stuff. Nice post.

Brands are the way to go, atleast in future when all G algorithm is changing.

Brands = trust + good visitor experience on websites = something G likes a LOT

MisterGemini said:

Yes they can. By using the various signals as was noted in the previous post.

Again, it is not the keyword match domain that will rank, but rather the branding. This is happening all the time. I am more convinced that its not only the future, but the way to go today.

Search 'computers'

Dell.com
Apple.com

Where is computers.com ? Nowhere.

The brands dominate.

Build your brand, you dominate.

Click to expand...

I gave some more thought to my domain name examples during my drive in to the office, about how the examples I used are generic words - but aren't necessarily keywords - and wanted to post a caveat to my argument, but MisterGemini beat me to it and did so much better than I could have -- thank you MisterGemini.

I agree totally that the smart money is on being a brand, vs. being a keyword.

My spin on it is to be a brand that just happens to have a core keyword in it.

Using the 'skirts' example, my choice for a domain would be girlyskirts.com or skirtgirly.com instead of girlywear.com. But then I'd be pigeonholing myself into selling just skirts...

GS

Hey.. great minds think alike . Haha.. this domain branding selection is exactly what I promoted as a method for choosing domain names in another thread not to long ago. I only cited the example of girlywear to avoid any misunderstanding about branding vs. keywords in domain.

For all of my projects I always combine either a partial keyword, or one word of a two word phrase, while making the other element of the domain something that sort of gives it that brandability and descriptiveness.

In the example we are using yours were good. I would also consider domains like:

skirtalicious.com
wewearshortskirts.com
skirtz.com
hotskirts.com
skirtpalooza.com

Just to give a few examples. I have no idea if these domains exist or not.

In terms of the dash ( - ) in domains debate, surprisingly, I am starting to lean a little more towards thinking its acceptable. Perhaps when you combine the dash with a proper branding domain, you have something that is effective. To take from the examples above, it doesn't always apply, but the ones that do would be:

we-wear-short-skirts.com
hot-skirts.com
skirt-palooza.com

The reason for this is because google will more easily define the actual words within the domain.

Why would google like this? Well look at it. It's easier for a person to read. Thus, user experience +1.

I think the dash got a bad rap only because people who used it were generally lazy in their site/seo builds.

In my experience, whatever you can do to make things easier for google, generally makes good for your rankings.

I have not tested this yet, but the good news is that I have an upcoming project where I will. Will see how those results turn out in a few months.

Mister Gemini

Gunner Steele said:

I gave some more thought to my domain name examples during my drive in to the office, about how the examples I used are generic words - but aren't necessarily keywords - and wanted to post a caveat to my argument, but MisterGemini beat me to it and did so much better than I could have -- thank you MisterGemini.

I agree totally that the smart money is on being a brand, vs. being a keyword.

My spin on it is to be a brand that just happens to have a core keyword in it.

Using the 'skirts' example, my choice for a domain would be girlyskirts.com or skirtgirly.com instead of girlywear.com. But then I'd be pigeonholing myself into selling just skirts...

GS

Click to expand...