Long titles performing better in Google, even with lower keyword density

[silverbytes]

I've seen some long titles performing better than short ones in Google lately. If user search a 3 words keyprase, in most cases you find

1- A long title in first position using those 3 words, some other word/s and again the 3 words or at least 2

And a site not performing first result even with 3 exact words searched.

If we talk about titles only my feeling is Google considers better the long title instead the short one (even when short title hits 100% of keywords 3/3 and long one, not just 4/6) having a lower keyword density.

Ie: search = "blue used widgets"

#1 result "blue used widget bla bla widget" better than
#5 "blue used widget"

Forget what site is better or relevant. I'm talking about title and how Google considers that.

[tedster]

Even as a content analysis factor, I think density is a webmaster's tool but not a direct part of the algorithm. So it wouldn't surprise me that it's not a factor in title tag relevance either.

Do you think there might be some downside to having a title tag that exactly matches a popular query? Seems to me that might be an SEO ploy more than a true content provider's approach.

[silverbytes]

I don't think matching exactly query have a downside itself, but I think Google scores long titles better than short ones even in cases like described: 100% acurracy on searched term, but it's just my guessing.

And agree with seo ploy. But to be honest in competitive search terms even important sites try to get #1 position tweaking title no matter if title does really represent the content fairly.

[CainIV]

I would agree. I am finding that prominence in the title, as well as a combination of semantics is playing a big role now. Longer titles with closely related themed works are working very well.

[potentialgeek]

The rationale for rewarding longer titles could be a better user experience. Instant understanding of a web page's content is more easily possible through a longer title (usually).

I don't know if I believe in keyword density for improved ranking but I suspect spam keyword density could be counted to kick a site down. I suspect Google checks spam anchor text density for its 950 penalty.

p/g

[silverbytes]

Perhaps patent rewards longer titles with a 60-70% density better than less than 4 words 100% density, considering fact keyphrases do not go over 3 words in most cases, to disencourage the seo practice to target exactly what users find, making titles "spammy" instead informational.

I'd love to hear something official about that.

[sandboxsam]

I have seen this used since Sept 2006 [webmasterworld.com...]
Example: 12 keyword phases separated by commas or *. 36 words long many repeat words.

<title>City1 phase tail1 *City1 Phase tail2 *City1 phases tail3 *City1 phrase tail4, City1 phrase tail5 *City1 phrase tail6 *phrase2 *City Phrase tail7, phase3, phrase4, phrase5</title>

I always thought Google would eventually view this as spam or else I would have used this technique.

There is one webmaster in my area that has setup her clients’ titles using this technique and it works!

So, is this spam or good technique?

[silverbytes]

I think that is seen as spam by google and if somebody mannual check on that, will penalize the site. My understanding is title with more than 10 words is not realistic and google knows that.

Also agree with user that said "I would have expected that words would be weighted, descending, from left to right. "

In my opinion beyond 12 words there is no lif talking about how that counts for ranking. But my feeling is google started to reward longer titles since those are more descriptive in attempt to minimize the spam (not to encourage it with extremely long titles). I'm seein my solid 3 words titles close related to page but hitting the "master keywords" go down in list lately, and larger get better.

But why don't you try the following and post again:

1) 2 pages with same page rank with large 12 workd and 2 pages with same pr with 6-9 words using keywords at beginning but also descriptive words later.

And check if serp changes in positive or negative way in each case.

[Miamacs]

I don't think keyword density would play a role anymore...

Prominence, sure ( is the document ABOUT the keyword or just mentions it somewhere in line 283? ), but not density.

Also on longer titles... do you think that the possibility of the title being more unique ( as longer, more descriptive titles tend to be ) would play a role in this?

I mean there might be much more pages titled exactly the same when it comes to short, straight-out-of-adwords choices, right? there's not much uniqueness in that...

But if you chose something fitting for your pages, and if those pages were unique themselves, it's likely that your titles will be as well.

...

[jimbeetle]

The rationale for rewarding longer titles could be a better user experience. Instant understanding of a web page's content is more easily possible through a longer title (usually).

Yep, and if -- big IF, cap I, cap F -- Google uses SERP CTR in any way, a page with a title that entices the user's click would better perform.

[silverbytes]

Also on longer titles... do you think that the possibility of the title being more unique ( as longer, more descriptive titles tend to be ) would play a role in this?

Indeed NO.

I see a bunch of keywords comma separated, succeding over descriptive terms.

Should I report that as stuffing spam?

And I'm afraid thart jibeetle be right. If users prefers to click on ridiculous long spam titles, they will be rewarded.

[edited by: Robert_Charlton at 8:40 pm (utc) on Aug. 12, 2008]

[jimbeetle]

I really wasn't speaking about spammy titles. I had in mind well-crafted titles in response to potentialgeek's comment.

[Miamacs]

I recently spoke with someone who said that she'd automatically click the first result without even looking at the title ( but she wouldn't know what that 'other' button was next to search ).

...

can't remember my answer though. I'm not sure if i was able to counter that
*grin*

Note that most people I know would actually read through at least halfway the SERP before clicking anything. And my experience as a user is that comma separated (wish)lists of keywords only seem to hold out on age old sites... which rank top 3 'in spite' and not 'because' of their spammy titles.

...

but then again, exactly because I don't see them all that often, I haven't tried monitoring this for a while.

[sandboxsam]

Using Google, I only see seven words in the title even through the title tag contains 36 words. So what is the point if the user can only see seven words? Sounds like hidden text to me, which is being rewarded.

Did I say this is the first result at the top of the page.

It doesn't which browser you use; you will never see all 36 words.

[youfoundjake]

I have found that a title accurately reflecting the page contents will always do quite well, based on if the search term is equal to the title.
I have 3 sites that the titles are all written to not only contain keyphrases such as "blue widgets" but also having the more popular search terms such as "where can i purchase blue widgets"
I would say that well over half of my visitors come from a result that directly correlate to the title of the page, almost word for word.

[silverbytes]

Google only shows x characters and beyond that you se a "...", but my guessing is those keywords beyond you can see in serp are being rewarded. If that is correct, that is bad.

[sandboxsam]

silverbytes
"If that is correct, that is bad."

This is correct and I agree; it seems bad to me too, seems to be hidden text.

But is it bad to Google? Is this an oversight by Google or is this how Google really wants us to construct title tags?

If I'm trying to use clean concise titles while Google wants strings of keywords; then I guess I have been mistaken?

[CainIV]

Google only shows x characters and beyond that you se a "...", but my guessing is those keywords beyond you can see in serp are being rewarded. If that is correct, that is bad.

Google views and registers the entire tag. They truncate the title for options reasons in the SERP's - so that one website does not get an advantage in terms of title length over others, and for uniformity for the browser.

In general, prominence plays a factor. Google expects the most important theme to be first in the Title. This is logical document layout 101.

[edited by: tedster at 7:44 pm (utc) on Aug. 14, 2008]
[edit reason] add quote box for clarity [/edit]

[silverbytes]

I'm seein again a 18 comma separated title peform #1. Again it is absolutely necessary to have Google's word on that. Guidelines about it. Is that a case to report or a tendence to follow all of us.

[edited by: tedster at 7:56 pm (utc) on Aug. 14, 2008]

[tedster]

In general, prominence plays a factor.

I used to think so, too. But in more recently times, and especially with titles, I've seen a lot that makes me question that traditional idea of prominence as an important relevnce factor. After all the concept of prominence was pretty much popularized by WebPosition Gold in the nineties - before Google was even giving "backrubs" ;)

[youfoundjake]

I believe that google will truncate the title after 64 characters...
Found this from a couple of months back...
[webmasterworld.com...]

Also going back to the WebmasterWorld thread of building the perfect page basics...
[webmasterworld.com...]

[tedster]

You're right, Google will truncate what they DISPLAY of the title element. But what they include in their ranking calculations is another story.

[youfoundjake]

Sorry, yeah, should have been more descriptive.
Google will display upto 64 characters, but my understanding, is that while Google may index the "whole" title, that more weight is applied to the first 64, possibly to prevent keyword spamming..which appears to go counter to what the OP brought up,...

[tedster]

Yes, and that's one reason why the OP's practical observation is worth a discussion. It does fly in the face of what many of us think!

[CainIV]

I used to think so, too. But in more recently times, and especially with titles, I've seen a lot that makes me question that traditional idea of prominence as an important relevnce factor

Hi tedster - what are you seeing, more specifically - that leads to question traditional prominence.

For the sectors I watch, the top 10 websites read like a duplicate title nightmare :)

[CainIV]

I used to think so, too. But in more recently times, and especially with titles, I've seen a lot that makes me question that traditional idea of prominence as an important relevnce factor

Hi tedster - what are you seeing, more specifically - that leads to question traditional prominence.

For the sectors I watch, the top 10 websites read like a duplicate title nightmare :)

[silverbytes]

Agree CainIV and tedster. No matter how many chars are displayed, Google is rewarding keywords beyond what you see in serp.

[sandboxsam]

I have also seen in conjunction with long title tags the use of an asterisk (*) as the first character in the title.

This has been going on since 2006.

I think it would appear, to most people searching, that this is added by Google as a means to highlight this website. I didn't think this was acceptable to Google?

There are several websites in my sector that do this and still rank very well. (yes, they have long title tags with repetitive words.)

[tedster]

People have been trying special characters in the title elements for a while - sometimes to attract the eye and sometimes just to separate keywords. I haven't seen any really odd characters (like a heart) in many months, so Google may have taken some action against those. I've heard nothing official about that, just a personal observation.

An asterisk at the start of a title seems like a minor sin - but also a minor advantage if any. It's just trickery, and for me, not a sign of trust or a solid business. It makes me less likely to click

[silverbytes]

I'm seeing a competitor using a period at first and ranking #1. I personally avoid that kind of tricks because are not good to users. But I see it works for them so ?...

Somebody quoted GoogleGuy "Just speaking for me personally, I wouldn't use them. Hope that helps.. "

I really hate when deceptive techniques perform good on Google.