2015 Google On-page SEO Ranking Factors List (Including Deprecated Factors)

[martinibuster]

I want to split off on-page from off-page and discuss solely on-page ranking factors, including the deprecated factors. What's your list of important on-page factors and those that are less important?

2015 Ranking factors
User experience metrics (all of them)
Shorter title tags
Original content
Engaging content that provides an answer, teaches, informs, is useful, delights
Original images
Quality site design
Descriptive meta description

Deprecated
Keywords
Focus on longtail phrases
Focus on ranking for specific keyword phrases
Lean code

[goodroi]

I said "mention the keyword at least once on your page"
You said "keyword phrases associated with search queries in virtually every title tag"

I think you might be comparing an orange to an apple.

[martinibuster]

Of course, you're right. There is a difference between on the page or in title tag. I consider title tag as on-page optimization. But I still have reservations about it dependent on how it is done.

[lucy24]

Should how be considered a stop word?

We need a third category: words that are meaningless in and of themselves, but that determine the interpretation of a search query. In fact "how" is a good example, because you could easily have a page that shows how to such-and-such without ever using the word "how". Similarly "history" "where" and so on.

[webcentric]

I can see how "how" might bring up results that contain a set of instructions whereas "when" might connect well with date and location information such as found in a calendar. "Where" smacks of locality and then there are the results of combining such terms. Not sure how many people naturally type "Where and when is the event" but I'm sure it happens regularly. I'm finding this an interesting foray into concept over keyword.

[nomis5]

Number 1 is User Experience Metrics.

I serously wonder what is going on when a third of the posts are about lean code. It matters nothing if the code is bloated and G cant distinguish that anyway.

If the user stays on a page longer than average compared to similar pages and the user then stops searching, job done.

[Nutterum]

@martinibuster : Is it wise to list the entire keyword phrase? What about a longtail phrases? Is it wise to use keyword phrases associated with search queries in virtually every title tag across an entire site?

Ask Amazon - they do pretty well, doing exactly that. I have seen it on some heavy authority websites too. They do it so that once you tybe search phrase matching/partially matching with the domain name Google would place 6-7 results in a row from that same domain name.

It is all a matter of how and where to use keywords. It is not a matter of if.

[dethfire]

I think quality links is the only real SEO factor. I still see plenty of complete garbage looking sites with no regard to any seo or design guidelines ranking very well in my sector.

[webcentric]

Purely as a writer I find it pretty boring to keep repeating the same words over and over again in an article or editorial piece. In the past and maybe to some degree still today, it's been an effective SEO technique but I still find it a tedious and unimaginative way to communicate. Maybe it's just me but my writing style has always tended toward

1. using the primary terms for my subject early on the page and in a prominent way (it is the subject matter after all)
2. using a wide variety of related terminology to avoid repetition (getting out the thesaurus if necessary to find a different way to illuminate something)
3. writing from a variety of perspectives (how red looks to a kid, how a color-blind person understands red, how red enrages a bull, and why red is an emergency color for example)

You could say these are SEO techniques but I'd rather like to think of them as fundamentals of good writing...something primarily interesting for humans rather than interesting to bots. Perhaps the moral is (feel free to disagree) that good writing should be enough where keywords are concerned. Effective writing should naturally address the "what, when, where, how and why" of things which may be the easiest way to address the interpretive aspects of search algorithms mentioned above. In the long run, it might just be wise to look at the fundamentals of writing and assume that Google has done the same. Which would make me wonder...

Does proper language skill factor into authority at all? I mean, can something scribbled in fifth-grade level prose compete with a doctoral thesis where authority is concerned? Might depend on the searcher and/or the subject. No fifth-grader wants to read a thesis but a grade-school educator might want to read what fifth-graders have on their minds these days. Makes me wonder if a searcher's reading level is, or will someday be, a factor in all of this.

[Sillysoft]

I serously wonder what is going on when a third of the posts are about lean code. It matters nothing if the code is bloated and G cant distinguish that anyway.

Im sure Google cant determine between bloated and lean, but it can determine between slow and fast. And using Adsense and webmaster tool, page speed is an important factor to Google when it comes to determining your current site optimization level. They even use it on your scorecard in Adsense. But we should not forget who matters here and that is the user, Google is a big part getting them there but if your site has bloated code and loads slow, you risk losing visitors because they aren't going to stick around waiting for your site to load.

But there are a lot of ways to improve the page speed of your site:
- Minify
- Using a subdomain to load things like images, css, js files etc (But not too many due to DNS lookup requirements)
- GZp/Compression/Caching
- I use memcache or Redis to cache results like view counts or data that doesnt change much
- Moving JS includes/code to footer
- Using asynch, case by case basis

I use YSlow to grade my site/progress and use it to see what areas I can improve in. And though I had been a senior developer for 15 years in the industry, I still always look to improve my coding and expand my knowledge. For my Adsense sites I use PHP and have used this website as a good refresher course:

[phptherightway.com...]

Then of course code itself is its own category such as auto loading, design patterns, reusing classes etc etc and databases such as mysql with proper indexing, normalization etc etc. With regards to using a sub domain, as you may already know there are so many "pipes" available by default to download a web page. Using a sub domain allows more pipes to be open, which means faster download of the webpage. But again, too many sub domains and you negate the extra pipes with having to wait for DNS requests.

Perhaps this is way off topic, but just wanted to share my thought process when it comes to trying to keep lean code/fast page load times.

[EditorialGuy]

Re keyword stuffing: There are times when the frequent use of a keyword is completely natural and organic--and when "elegant variation" is clumsy or impractical. For example, if you're writing about MGs, you're probably going to say "MG" quite a few number of times. Your writing would sound forced if you didn't use "MG."

Ditto for a page about "penicillin," "milk," "Rome," "cats," or the "iPhone 6." (If you were writing a 500-word review about the iPhone 6 for CNet, it might be unnatural not to use the term "iPhone 6" at least a dozen times.)

Of course, there are times when keyword stuffing is obvious. Let's say you're writing a page about the Widgetco WCR-1000 Wi-Fi Router. It would be natural to use the term "WCR-1000" repeatedly in your review, but even the most brain-dead reader would find it odd to see "Widgetco WCR-1000 Wi-Fi Router" used in its entirety more than once within the body text of the review.

[martinibuster]

There are times when the frequent use of a keyword is completely natural and organic...

Most any statistic in isolation will result in false positives. The accuracy of statistical signals improve when they're combined with other signals. Outbound links, prevalence of keywords, statistics related to words on the page- considered together they become more accurate.

This holds with other unrelated analyses. For example, there is a paper related to artificial intelligence that attempts to understand words and their different contexts. Turns out the accuracy of the system improves when the source documents that form the basis of the semantic analysis are considered together with the links that interlink those core pages.

[webcentric]

Yes, I'll grant that when you're discussing a product (for example), redundancy is somewhat of a necessity. When you're pushing a "thing," you're going to have to call a spade a spade most often. Still, sentence structure and article/review design is possible without constantly repeating the label for what you're talking about.

Example:
"When BlaBla2 first came out, I was amazed at how irrelevant it was compared to BlaBla1. The first iteration of this stupid concept was bad enough but the second has taken the meaning of babbling to a whole new level. Given the public's demand for silly blathering products, I have no doubt the latest version will be flying off the shelves in record numbers soon and you'll be seeing the results of this useless product's nonsensical algorithm popping up on your favorite local webmaster board in the very near future."

[lucy24]

using a wide variety of related terminology to avoid repetition (getting out the thesaurus if necessary to find a different way to illuminate something)

Isn't that what spammers do? A year or two back, someone hereabouts posted a very funny sample-- which I now can't find-- of an email gone awry, illustrating an excess of synonyms.

If I were making the rules, I would forbid anyone under the age of 40 from using a thesaurus. They should be reserved for what's-that-perfect-word-that-carries-exactly-the-nuance-I-want-but-I-can't-remember-it, not for picking from a vague list of synonyms that the writer isn't really familiar with but hey, this one sounds good, I'll use it. There's also the risk that if you use too many synonyms and variant forms, the text reads as if you're talking about several different things.

[seoskunk]

using a wide variety of related terminology to avoid repetition (getting out the thesaurus if necessary to find a different way to illuminate something)

Reminds me of a SEO telling a joke.....

"So this Guy walks into a pub,tavern,public house, drinking establishment, bar, inn, saloon........... "

[webcentric]

My first college English professor prescribed "daily dosages of dictionary and thesaurus" and it was the best thing that ever happened to me. Mark Twain is a superb example of someone who knew how to say things without repeating himself ad nauseam. Being faux-clever because you have a thesaurus is one thing. Having a decent grasp of language and knowing how to communicate effectively is another. Attempting to manipulate a search engine is still another. Using synonyms generally offers the opportunity to approach the subject from a different angle and, quite possibly, sink a point home that went unnoticed by the reader when said another way. My point isn't about stuffing an article, it's about comprehensive writing and, who knows, maybe it's about trying to be interesting at the same time. I agree agree with Lucy that exercises in variant expression can get tedious. Skunk's example demonstrates a list keywords that make a single sentence ridiculous. Use those terms over the course of an article though and they may help to bring focus to the subject. Of course, if you use "inn" and "tavern" in the same article, then you're muddying the issue because I don't know if you're talking about a place to rent a room or a place to get sloshed.

Added: Avoiding creative grammar and composition because someone used it as a spam technique once (or a million times) seems a bit extreme. Again, we're back on "what not to do." It would be a shame if the result of all the spamming we've seen means that we can't write an engaging and informative article because it's become Black Hat to be creative with our thoughts.

[seoskunk]

It would be a shame if the result of all the spamming we've seen means that we can't write an engaging and informative article because it's become Black Hat to be creative with our thoughts.

How would you define black hat? I think in Google's eye's anyone who tries to manipulate the SERPS is a blackhat, so that's all seo's.

[webcentric]

I think in Google's eye's anyone who tries to manipulate the SERPS is a blackhat

Can't ague with that. I do think there's a difference between producing content to manipulate the SERPs and producing content that's worthy of a decent mention in the SERPs. In some cases they could be exactly the same (by design or not). I'm simply saying it would be a shame if great writing is being penalized because Google thinks it's designed to manipulate the SERPs.

[lucy24]

My concern is variation for the sake of variation. The more distinctive a word is, the more readers notice if you use it too often. But if your impartial critic points out that you said "distinctive" in two consecutive sentences, then going for the thesaurus to replace one of the two with "unusual, striking, singular, characteristic, extraordinary" is probably not your best fix. Go for a complete rewrite instead.

That's if you're writing for humans. If you're writing for a search engine that can't be trusted to understand that your entire page is about widget repair even though you only used the word "widget" once ... then there's a problem.

[netmeg]

Meh. I have one criteria - and so far, it's served me pretty well both for myself and my clients. I read it out loud. If it sounds awkward or weird or unwieldy or silly or keyword-heavy out loud, it's going to scan that way on a page as well. It's probably not foolproof, but it mostly works and I don't have time to obsess over it.

[Dymero]

"Focus on longtail phrases"

If mean trying to rank for many longtail phrases on one page, then I agree. But if you think there are "long tail" topics that are worthy of a page on your site, then I'd definitely try to go for that. Though, and this should be SEO 101 by now, like any other thing when targeting keywords, it has to make sense. Using "Widgets San Diego" as the title of your new page is just asking for a smack down.

[RP_Joe]

In my research, longtail keywords, and keyword phrases can still be a ranking factor, depending on the competition.
The theory of longtail keywords is that there are fewer people who have an exact match for that longtail. However, in 2015, with so much content out there, very difficult to be the only website with an exact match on a longtail keyword. But if you are, and you have the other factors covered, responsive, good user experience, fast website, decent design, then you can rank very highly for that unique keyword.

However, the value of such rankings is certainly questionable. You're so far outside the mainstream, that type of traffic does not monetized very well.

I see people over and over again trying to figure out how Google analyzes the website. Google Chrome, Google toolbar, and android all report back to Google what people are doing. Isn't it easier just to monitor would people are doing, and reward based upon what people like?

[webaddict]

Meta Description? Really? This can't be real in 2015. Google dynamically generated at least 50% of all meta descriptions that are listed in it's search results. It's only increasing every year, how can this put it in a small list of factors. This has to be the most useless thing one could focus on if they're trying to build a website that ranks and generates traffic.

[lucy24]

This has to be the most useless thing one could focus on

Now, wait a minute. Even if a meta description isn't in-and-of-itself any use as an SEO factor, that doesn't mean you should ignore or even omit it. I don't think most websites partition their design-and-development into separate and unrelated tracks: one strictly for UX, one strictly for SEO, one strictly for RAM efficiency, and so on.

:: vague mental association with "this carpet may be good at hiding dirt, but that doesn't mean you should buy it for the express purpose of hiding dirt" ::

[webaddict]

Fair enough Lucy, I guess someone could find some use for the meta description. But I really don't see a point for it when associating success with Google for UX, SEO, Design, Development, Call To Action etc. Google rarely uses the meta description that one creates and if they're not using it in the listing description, it really doesn't seem likely they're using it for ranking the site either.

With all of that said, Bing is still using the manually created meta descriptions so I can say that might provide some control over user behavior when they're in the results set of Bing.

I guess I'm just worried that if someone is listing a meta description as one of the major factors of 2015 in SEO, they might not know what they're talking about.

[feiman]

This has to be the most useless thing one could focus on if they're trying to build a website that ranks and generates traffic.

While meta doesn't have any ranking impact on its own, a well written meta description (that Google doesn't ignore and overwrite) can have a significant impact on CTR. If you are in the camp that believes click through, bounce rate, time spent on page, depth of visit, etc are ranking factors, it is perfectly reasonable to look at meta descriptions from a strategic perspective.

[martinibuster]

Feiman, good observation. I agree, I too view it as a strategic action for on-page SEO. Part of the struggle is getting searchers to visit your site. The other part of the struggle is to make sure they stay on the site. So even if you get the meta description right, it's possible that failure to cause visitors to hang around, may give cause to a demotion in the SERPs.

To be honest, I don't believe in ranking factors and have stopped thinking in terms of ranking factors for a few years now. All those lists that have factors lined up in order of importance, I truly consider them obsolete, because I believe that not only have Google and Microsoft killed ranking factors, they have turned them against SEOs. Despite the title of this discussion and the opening post, I believe that ranking factor optimization is dead.

For the past few years I've been finding that a strategy based approach works well for obtaining my search marketing goals. But how can I communicate the strategies involved when the search community still thinks in terms of ranking factors?

[Dymero]

martinibuster, be that voice for change in mindset. You're a very active contributor here, so expound on it whenever possible and appropriate. I happen to agree with you. For example, when I mentioned long tails above, I meant in terms of finding ways to attract more traffic to a site by writing about site-appropriate topics that people are searching. I'd consider that a tactic as part of a strategy, not a ranking factor.

Going forward (and really for a couple years now), any approach should be about what can a site owner or marketer or writer do to attract more traffic/readers/customers to the site, rather than chasing specific factors that might or might not inch a site forward in the SERPs. It's just going to be a more rewarding use of our time to do that than face the constant roller coaster of fear and relief (or just fear) that comes with the Google algorithm update/refresh schedule.

[netmeg]

(I knew it was a trap)

[elguiri]

To borrow a phrase from Matt Cutts, Google is focusing more on "things, not strings."

So instead keyword optimization, or long tail optimization, we should bundle those things together and think "entity optimization."

This concept groups together a bunch of things: keywords, yes, but also synonyms, niche vocabulary, niche style, niche references (in and out), good user experience (for people associated with or an interest in that niche). That to me looks like an SEO strategy for 2015 (but also 2012 and probably 2016).

[martinibuster]

An entity is simply a person place or thing and neither of those need optimization. What I'm proposing is unbundling keyword or longtail optimization, all of it. Activities related to that have been turned into classifiers used to root out spam. What percentage of total pages feature search engine queries as a major feature of the page? That is a signal of a site created to rank well in the search engines (not an authority site) and sites created to rank well are precisely what search engines do not want. Sites created to rank well are not authoritative. So as I stated previously, the OLD WAY of SEO, where you begin with "ranking factors," where you begin with a list of search queries, where you begin with so-called Ranking Factors, and build pages around them is obsolete and quite likely harmful. It's deprecated. Organic SEO as recommended and practiced by the industry is more and more being turned into the signals of poor quality, the hallmark of sites search engines do not want.

Just as eight years ago we reached a crossroads where the industry woke up and understood that reciprocal linking was not recommended by Google, was not "white hat" and was actually being demoted, I believe for the past few years, since Panda, the search engines have been gradually tightening up on Best Practices Organic SEO as practiced by the SEO Community. Which is why advice such as, "Build pages with the word HOW DO I" is misguided. Why lists of Ranking Factors are misguided. These are the things they are using against you.

Do a search for how to tie a prince nymph fly and there aren't any of those sites created to benefit from search queries. Just authoritative sites like Orvis and the site created by noted fly fishing author, Charlie Craven. Yes, it's about quality but to my mind it is as much about whacking sites created with By the Book SEO.

This most recent quality update is proof of the idea that Ranking Factors are misguided. It's Google going further down the road of putting into practice what's been described in various research studies.

So if we throw away ranking factors as we have known them for the past ten years, what do we have left?

•User experience metrics (all of them)
•Shorter title tags
•Original content
•Engaging content that provides an answer, teaches, informs, is useful, delights
•Original images
•Quality site design
•Descriptive meta description