As we all know, SEO is a very labor-intensive job. It can be really easy to let some things fall by the wayside even if you know you're supposed to do them but don't have the time. It gets even more complicated when you forget to do them in the first place! Now, Danny is an awesome man of many strengths, but he can be a bit of a forgetful grandmother at times - even he knows that it is important to write down processes so they're easier to replicate in the future. As an early holiday gift to us all, he has decided to spill the optimization beans and share his SEO checklist with the community he loves. Get the details in today's Whiteboard Friday video, and the more general checklist in the post below. Please feel free to share what's on your checklist in the comments below!
Cloaking - Is the page Cloaking? (It shouldn't be)
Capitalization - Does the page render with Capitalization in URLs? (It should redirect to all lowercase)
Trailing Slash (URL) - Does the page render with a backslash and without a trailing slash in the URL? (It should redirect to the non-backslash version)
Are all of the images in the optimal file format and web optimized? (Files should be as small as possible)Do links without editorial value include nofollow? (They should)
Video Transcription
Hello everybody. My name is Danny Dover. For today's Whiteboard Friday I'm going to show you something very, very special that we use internally -- the SEO cheat sheet for launching a website. We couldn't really think of a good name for it, but that's not the important part. The important part is the content that's on it.
Let me explain where this came from. We have a VP of Product here, his name is Adam Feldstein. A lot of times he'll come to me and there's a page that just needs to go up in the next two hours. Sometimes it's a landing page of some kind. Sometimes it's a marketing page of some kind, and SEO comes in right at the end. Usually what happens is I look at Adam like, "Well, Adam, you know you shouldn't do that." And Adam's like, "Danny, why are talking like your grandma? I don't understand that." The idea is that SEO, of course, is supposed to be within the entire process, the entire product process. You work it in every step of the way. To be fair, at SEOmoz I think we do a good job of that, but sometimes we get these side products, it just has to happen at the end. In those cases, this is the cheat sheet that I go through to make sure everything gets covered.
So, I've very deliberately broken this into two category groups. I have critical on my right. And I have not quite as critical. So, I'm not saying that they're not critical. I'm just saying that they're not quite as critical. So they're all important. The things on my right are the most important things to cover. So let me cover these one by one.
The first one is targeting. I can't emphasis how important this is. A lot of times when I get mockups back, there will be many, many ideas expressed on a single page. Sometimes they'll relate to each other and sometimes they won't. The idea here is in a perfect SEO world, which we don't live in, but if we did, there would be one idea per page. The reason I say that is because from a search engine perspective if there is one idea per page, it is very, very easy to figure out what that page is relevant for. A lot of times when I get these mockups, part of the process is figuring out what is the bigger idea that is trying to be expressed and then targeting that phrase. So, if it is Justin Bieber, if we're working on his home page for example, I don't know why we'd do that. But, if we did, it would be just him as a celebrity or his personal reputation. The concept of Justin Bieber, that is what the homepage of him needs to express. That would be what we would target will all these different things. I'll cover all of these. Once you figure out what you're trying to target, then you need to go through and target them in the correct ways.
The first thing that comes up is the content. We talk about this a lot in SEO. The content is king. If I had a dollar for every time I said that, I'd have a lot of money. Content is extremely important. The content that is on the page is why search engines are indexing in the first place. They act as middlemen or as middle machines as it would be. Real human beings go onto Google and they search with real queries and they are hoping to find content that is written for them about whatever it is they're searching for. So the content needs to be about whatever you're targeting.
Underneath the content is title tag. We've done a lot of research on this at SEOmoz. We've found that title tags are extremely important when it comes to on-page ranking factors. In fact, they're the most important on- page ranking factor from an SEO perspective. Titles tags, what you want to do is have the keyword phrase at the beginning of the title tag. I'll link in to a post below that explains all the intricacies and all the subtleties of title tags so you can get a good idea.
After I check over the title tags, I look at the URL. So the ideal, I want the URL to be as short as possible and as semantically clean as possible. Does it make sense? Something like a domain, a category, a subcategory page, and the content. Does it express an idea that makes a lot of sense? So if it's something like SEOmoz.org/content/blogpost/whatever the blog post name is, that would be a cleaner URL. Just for example.
Under the URL, I have meta descriptions. Meta descriptions don't help you at all from a ranking perspective, but they help you with click through on the search engine result pages. This is kind of a way of doing a free ad for search results. I always make sure that we have one of these and it includes again this target phrase that we picked earlier. If it is in there and there is an exact match or a near match, it will bold it in the search engines, which helps which click through rates.
Underneath that, I have rel=canonical. This one has come up a lot more in the last I'd say three months or so than it has before. This is very important. This is a resource the search engines have provided for us that we can tell them what the canonical version of a web page is. You'll see this a lot where tracking parameters mess it up or a trailing slash versus a non-trailing slash on a URL will mess this up. It is very important that you include this here. Actually, at SEOmoz, we're trying to push an initiative through dev to have this on every single one of our pages. That's something that I think will come here in early Q1 201l, making sure that we're making it very clear which is the canonical version of every single page on our website so that link juice doesn't get distributed unevenly.
Underneath rel=canonical we have Alt text. Alt text is a textual representation of an image that you can optionally apply. This is helpful for search engines and for human beings as well. So, human beings that need screen readers or that are using some other alternative method of visualizing web pages, Alt text comes in handy. We have found this, along with title tags, to be an oddly, highly correlated search metric. Alt text, again you're going to want to use the target phrase you chose in the first step.
The last one is internal linking. This is the easiest one to screw up. When you have a new page that is about to launch, it is important to look at all of the things that are on the page itself, but it is also important to look at what pages are linking to it. Guess what? Again, you want those anchor texts to be using that targeted phrase that you came up with in the beginning. If it's the Justin Bieber thing again, you're going to make sure that all of those pages on the rest of your domain are linking to it with the correct anchor text, which is whatever you targeted.
That's the critical stuff. That's what I look through in the first five minutes of going through a new page that is about to launch. The other things that I go through have a lot more subtleties to them. So I've put them into a different category. Sometimes you want them and sometimes you don't, depending on what the situation is. I'll go through each of those independently.
Meta robots. Meta robots is a tool that search engines have provided for us that allows you to either have a page be indexable or have a page not be indexable. It's sort of like robots.txt, although it is a lot cleaner implementation. Let me explain that. With meta robots it gives the option, like I said, of being indexable and non-indexable. It also gives you the option of having all the links on a page followed or no-followed. So whereas with robots.txt you can say, "Google, don't crawl this page," and it won't, with meta robots you can say, "Google, crawl this page. Don't index it, but have all the links on it pass juice." That way any link outward from your domain that you do or any link to a different page on your domain on that web page will still be indexable and will still pass juice. So, I never actually recommend using robots.txt unless you absolutely have to, because the benefits of using meta robots just outweigh it completely. Again, I'll link to this and explain it more fully.
Underneath that, I have meta keywords. Meta keywords are an older SEO tag. In fact, I think it's almost completely useless at this point. This year it came out that Google was not using it, Bing was not using it, and Yahoo came out and said that they were not using it, although Danny Sullivan proved that they actually were using it. What happened now is that it doesn't really matter that Yahoo is using it because all Yahoo search results are getting transferred over to Bing search results. Meta keywords are actually not going to help you. It's one of those things where not only will it not help you, it could also potentially hurt you. You're giving your competitors valuable information by providing meta keywords. You're saying these are the keywords that are important for me to target on these pages. You don't want to do that. You're spending time to help out your competitors, which is something you want to avoid at all times. The best argument I've heard against this is that with meta keywords sometimes they can help you on some social sites. I really haven't seen it on any of the big social sites. On some, like, some very niche sites, meta keywords can help you. Really, the bottom line for me is it is not worth your time to go through and add those on each of your pages. So, I never recommend doing that.
H1. H1 is one I go back and forth with a lot with different SEOs. The idea behind an H1 is that you are using HTML headers to explain to the search engines and to different protocols how information relates to other information on the page. So, with H1s we found from a pure ranking perspective they actually don't help you very much. We think this is because they've been abused a lot in the past. But the problem with just going out and saying that and making that the best practice is that H1s are actually very helpful for users. If you go to a blog, it makes a lot of sense that the H1 will be the title of the blog post. This is what the entire page is about, so this will be the H1. I totally agree with that. That is how it should be. From a strict rankings perspective, H1 is probably not going to help you very much. Maybe not even at all. But for users it helps a lot. I recommend you include them, but don't put a huge amount of emphasis on them.
Underneath that is cloaking. Cloaking is something that usually comes up by accident, although some mischievous people do it on purpose. Cloaking is showing one thing to search engines and showing something completely different to normal users. This comes up a lot on our website when we have one version that is being shown to logged in users and one version that is being shown to non-logged in users. A search engine cannot log in. It doesn't have credentials and it can't operate that way. So it is very important to figure out exactly what the search engines are going to see and make sure that other offline users are also seeing the same thing. If you're ever in a situation where you are targeting something based on user agent, say in Googlebot, you probably don't want to be doing that. Try to avoid it in most cases. There are some very limited hyperlocal exceptions to that, but feel free to ask questions in the comments if you want me to expand on that at all.
Capitalization. So, I'm actually going to group capitalization together with trailing slash. These are again talking about subtleties of URLs. Capitalization is if you are going to include capitals in a URL, and trailing slash is if you have something like www.SEOmoz.org/WhiteboardFriday/. If you have that trailing slash, that page will render, at least on our servers, that will render along with the page without the trailing slash. This is a mistake. This is something that we are going through and fixing. I think a lot of people make this mistake, actually. The problem with this is you're creating duplicate content. The same thing can happen with capitalization. If the URL has some capitals in it and you can also render the same page at a version that does not have the capitals in it, you are going to have duplicate content. These are two things I take a look at when I'm going through a web page that needs to go out into production into the next hour or so. I make sure they do not have these two things.
If you have any more questions, you want to expand on this at all, or you have anything I've forgotten, please comment in the comments below. I appreciate your time. I'll see you next week on Whiteboard Friday. Thank you.
Postări
Niciun comentariu:
Trimiteți un comentariu