John Mueller: Title tags "not the most critical part of a page"

Forum Moderators: Robert Charlton & goodroi

Message Too Old, No Replies

John Mueller: Title tags "not the most critical part of a page"

martinibuster

2:14 pm on Jan 20, 2016 (gmt 0)

When asked what the most critical part of a web page is, he answered:

More like the actual content on the page.

Read it on SERoundtable, Title Tags Not Critical [seroundtable.com]

File this with the hashtag, #CrapJohnMuellerSays

[edited by: Robert_Charlton at 8:41 pm (utc) on Jan 20, 2016]

Robert Charlton

8:27 pm on Jan 20, 2016 (gmt 0)

martinibuster, thanks for the reference. I feel that this is very definitely worth noting, and the section of the John Mueller video that Barry in SERoundtable references is well worth watching. It's not clear from the hashtag precisely how you feel about it. ;)

We've been discussing in this forum a steady progression of Google title rewriting for several years now, and John's comment about keeping titles "short and to the point" is completely consistent with the trends that some of us have been observing.

That said, I've recently been seeing enough pages ranking with bad or missing title tags that I've wondered whether the title element might have been somewhat deprecated. I brought it up about 10 days ago in this thread (one of many we have on title rewriting)...

Title Elements Replaced With External Links
Jan 2, 2016
https://www.webmasterworld.com/google/4784588.htm [webmasterworld.com]

This is from my comments in the thread, which describe a situation where all of the title elements on a site had been inadventently de-optimized, and the deep pages are ranking anyway....

I can't say from this whether Google has started rewriting titles more often, as our two observations might suggest, or whether we're coincidentally seeing something that Google is now doing all the time.

Since then, I've also observed a notable example where a query for "brady" briefly but prominently returned a page about Tom Brady showing "untitled" as the title. The query came up as I was checking an analysis I describe in a current discussion about the January core algo update...

Google confirms Jan 2016 core algo update was not Panda
https://www.webmasterworld.com/google/4786324.htm [webmasterworld.com]

It should be said that the page was an otherwise prominent article on a high-authority sports site, and that QDF was a big factor. Overall, I don't recommend ignoring the title tags. I try very hard to keep them descriptive, as JM suggests, not stuffed with keywords. I prefer stay within the Google display limitation of 55 characters for essential content when I can (and yes, I know Google does parse longer titles). I think that Google would just as soon follow page title suggestions and to use them if appropriate and helpful... but there are deeper levels of page analysis at work.

Like everything else in Google now, though, Google is weighing multiple signals. IMO, it still makes sense to provide signals that are helpful.

Edited to fix typo

[edited by: Robert_Charlton at 11:12 pm (utc) on Jan 20, 2016]

lucy24

9:54 pm on Jan 20, 2016 (gmt 0)

I don't recommend ignoring the title tags

I should hope not; I think it's the only thing the validator insists on within the <head> element.

To me it seems like another of those things, like "alt" tags, where you should start by looking at how a given feature affects the human user. If a given feature is clearly beneficial then you wouldn't ignore it unless the feature's SEO disadvantages are, beyond all doubt or question, massive enough to outweigh the human-user benefits. (Or, mutatis mutandis: If a given feature is clearly detrimental then you wouldn't use it unless the feature's SEO advantages et cetera.) And how often does that happen? I mean in real life, not in the realm of rumor.

aristotle

10:04 pm on Jan 20, 2016 (gmt 0)

Re-writing page titles is one of the worst things that google does. The author of the article chooses the title that best conveys what the article is about, then google's algorithm changes it to something that's less accurate, and sometimes even totally misleading.

This is a terrible disservice to searchers, because it makes it much harder for them to find the information they're looking for. It's also a big disservice to site owners, since they miss out on well-targeted traffic and get a lot of mismatched traffic instead.

Everything about this is bad.

tangor

10:27 pm on Jan 20, 2016 (gmt 0)

Rarely have G rewrite mine as they are short and sweet and specific. I have had clients that don't heed that advice and continually moan over the alternations. Because Title has been abused in the past (or worse, MISused, ie, same for every page) I don't fault g for trying to be meaningful. At the same time, I truly do believe any weight Title once held has been diminished significantly.

EditorialGuy

10:28 pm on Jan 20, 2016 (gmt 0)

Re-writing page titles is one of the worst things that google does. The author of the article chooses the title that best conveys what the article is about, then google's algorithm changes it to something that's less accurate, and sometimes even totally misleading.

1) We're talking about "pages," not "articles."

2) Page titles should "best convey what the [page] is about," but all too often, they're written with blunt-force SEO in mind. Let's not pretend that the average page title is a work of literature or even a work of journalism.

3) The title element isn't part of the page's content, any more than a meta description or a list of meta keywords is. (The page's actual content is within the body tags.)

4) Google wants to present searchers with useful search results that fit the SERP format. If the author of a page writes a clumsy, unwieldy, keyword-stuffed title, it makes perfect sense for a search engine to use its own description of the page in the search results.

Storiale

11:47 pm on Jan 20, 2016 (gmt 0)

aristotle + : You are totally correct. Click Through Rate suffers! Our competitors can not optimize their Title tags like we do, our 52-58 character Titles (many of them) are optimized like you would for PPC, adding benefit, incentive, keyword phrase - gets rewritten exactly like our competitor's site [Keyword Phrase] Website Name.

I mentioned this in another thread... it is a horrible decision! They wouldn't do this to PPC because they want as many clickers as possible. The only logic I can think of is to test usefulness of a page's content by normalizing Title Tags to be indistinguishable from each other. I hate this change. Our competitors are lazy, so it makes them equal to us.

aristotle

11:55 pm on Jan 20, 2016 (gmt 0)

Another bad effect of this is that it most likely makes google's search results worse. This is because many sites will lose some well-targeted traffic that they would otherwise have gotten if the title hadn't been changed. This means that over time the site will attract fewer backlinks, since fewer well-targeted visitors will see it and possibly link to it from somewhere. This will adversely affect the site's rankings.

On the other hand, some bad sites might see an undeserved boost in rankings if title changes are accidentally in their favor.

So some sites are unjustly demoted, whereas others are unjustly rewarded. And the overall final effect is that google's search results become worse, thereby just adding to searchers' difficulties in finding the information they want.

buckworks

12:08 am on Jan 21, 2016 (gmt 0)

If the author of a page writes a clumsy, unwieldy, keyword-stuffed title, it makes perfect sense for a search engine to use its own description

That much is true, but the trouble is that's not the only kind of title that gets rewritten.

Some of the changes I've seen seem have no discernible rhyme or reason.

fathom

12:13 am on Jan 21, 2016 (gmt 0)

So a page about dog farts won't rank well for viagra when the title is viagra but a page about viagra will not rank better with dog farts as the title. Meaning the page content is indeed the most important part but I don't believe his quote suggested the title element is any less important than it was in 2003.

According to Google's official guide [static.googleusercontent.com...] that assumption would be wrong... But I never watched the video.

<added>You can spin anything to be "for" or "against" but the original was spun as "critical to rank"... And NO with the 199 other factors leaving the title element as a brand will NOT GUARANTEE you won't rank well.</added>

ken_b

12:48 am on Jan 21, 2016 (gmt 0)

I thought that G rewrote titles to more closely reflect the search query and thus the displayed title would be more likely to attract a click.

fathom

12:49 am on Jan 21, 2016 (gmt 0)

So some sites are unjustly demoted, whereas others are unjustly rewarded. And the overall final effect is that google's search results become worse, thereby just adding to searchers' difficulties in finding the information they want.

The gauge for knowing what is just or unjust requires a special crystal ball. Since it is highly unlikely that this crystal ball exists, pages tend to rank where they should be in a fair system. Until the fairness changes with new approaches to combat what is newly discovered as being unfair.

Just another one of my inane observations, to enlighten the fog of the unknown or to simply brighten the thread.

lucy24

1:00 am on Jan 21, 2016 (gmt 0)

Titles aren't rewritten to conform more closely to what the page is about. They're rewritten to conform more closely to the search that brought up the page. Different search, different rewritten title.

Maybe, just maybe, the author's deliberately chosen "Riding Widgets in Guam" is meant to give a clue that the page is not actually about mountains in Finland, even though the distribution of (key)words on the page might lead a machine to think so. Or maybe, just maybe, the author's deliberately chosen title is intended to set off the page from all the other pages on the same subject, in which case this information might itself be useful to the searcher.

Either way, isn't it the author's responsibility? Sure, publishers change book titles-- sometimes over the author's express wishes. But booksellers and libraries certainly don't. It's not their job.

EditorialGuy

1:53 am on Jan 21, 2016 (gmt 0)

Titles aren't rewritten to conform more closely to what the page is about. They're rewritten to conform more closely to the search that brought up the page. Different search, different rewritten title.

In that case, shouldn't the rewritten title increase the likelihood of a clickthrough?

glakes

2:14 am on Jan 21, 2016 (gmt 0)

Google rewriting titles is an example of how content creators have limited control over how their content is used and displayed to others. Imagine the first Star Wars being referenced by a variety of rewritten titles to suite someone elses benefit. George Lucas would have a fit. Have we as webmasters lost our way and find how Google changes our text for their benefit? In fact, how is it different when a spammer takes one of our in-depth pages and spins the text compared to Google spinning our titles? Each is doing this for their benefit, without our permission.

EditorialGuy

2:58 am on Jan 21, 2016 (gmt 0)

Glakes, the text between an HTML page's <head> and </head> tags isn't "content." The content is the stuff between the <body> and </body> tags.

lucy24

6:37 am on Jan 21, 2016 (gmt 0)

shouldn't the rewritten title increase the likelihood of a clickthrough?

But at what cost? Do you really want people to land on your site if they're going to turn right around when they realize it wasn't what they were looking for? It's not the same situation as using a meta description in place of a visible-text snippet*; there at least you did write the description yourself.

The content is the stuff between the <body> and </body> tags.

This strikes me as a needlessly hair-splitting distinction. If the human user sees it, it's content.

* I don't know how common this is overall. In the specific case of ebooks, I'm all for it.

scottb

11:00 pm on Jan 21, 2016 (gmt 0)

Maybe I'm in a bad mood, but sometimes I find it hard to believe that the content is all that important either.

I can produce unique and original content that gets ranked #40 while irrelevant pages that barely match the keywords in my search get ranked much higher.

Yes, there are plenty of examples that do rank well based on high-quality content. There are plenty of others with low competition that don't.

tangor

12:24 am on Jan 22, 2016 (gmt 0)

g is a black box. Some think it has magical powers and should be able to see that THEIR site is better than any other simply because it IS better (in their eyes).

Got news for all and sundry: The web is a huge mind boggling bit of number crunching, automatically done or not, that beggars the imagination. The sheer number of webSITES much less the quantum leaps above that number in actual (and dynamic) webPAGES being sifted through tens (hundreds?) of thousands processors in a few dozen data centers wading through several petabytes of data being shifted globally in x number of languages, using algorithms written by faulty human brains is the full answer on why the serps look screwy time after time. If g's algo ever became AI it just might ask "How about a game of tic tac toe?"

Major changes have happened at g over the last 10 years. Now we are entering into the "fine tuning" of all those changes and it's not going as well as we'd like. Worse, economic pressures in the real world and corporate expectations also take their toll.

So, to keep the pot simmering (instead of boiling over) things like title edits were introduced... and we still, as webmasters, have no clue what triggers that rewrite. We do, however, know that title no longer has the early (pre-Florida in fact) weighting and is even less valued as merely 1 of 200+ signals involved.

Content remains king and might influence any rewrite of title that g does. What I've seen over the last three years is smaller moves by g to refine this process ... and no matter what they do there will be hordes of webmasters unhappy with anything they do. Nobody likes to hear "grin and bear it" (or is that "grind and tear it?") but what truly can be expected from a "free" service where no contracts for performance are in force?

There's a myth out there that g works with webmasters to make them money. Like all myths, it is hard to kill that "tradition" with reality.

Years back I dropped "keywords" from any title UNLESS they were part of the actual title. I have relied on PAGE CONTENT (as in content is king) for serp insert and ranking, as that has always been what the USER seeks: Content. The title that fits is the title to use, and I keep those short and sweet.

glakes

3:52 am on Jan 22, 2016 (gmt 0)

Glakes, the text between an HTML page's <head> and </head> tags isn't "content." The content is the stuff between the <body> and </body> tags.

Technically, yes it is meta information. However, this meta information is provided by the author of the page with the intention of displaying it to their users. Google modifying the title of a page is akin to the same spinning spammers do. Scrape, steal, spin, etc. Google respects authors rights no more than the same low lives that steal and mutilate our content for their own gain.

EditorialGuy

4:06 am on Jan 22, 2016 (gmt 0)

Google modifying the title of a page is akin to the same spinning spammers do. Scrape, steal, spin, etc. Google respects authors rights no more than the same low lives that steal and mutilate our content for their own gain.

"Authors [sic] rights" don't come into it. Google's organic search results are annotated links, and there's no rule of copyright (or even of polite behavior) that requires a linking site, such as Google or your site or my site, to let the link recipient dictate anchor text.

If you really think Google is akin to a "spinning scraper," you have a choice of easy-to-implement defenses: Robots.txt or <meta name="robots" content="noindex">.

Or, if you want Google to send you traffic, you can simply accept the fact that search engines get to decide how they display search results.

Walt Hartwell

4:47 am on Jan 22, 2016 (gmt 0)

Am I the only person that interprets John's comments to mean that the title element should accurately reflect page content?

Years back I dropped "keywords" from any title UNLESS they were part of the actual title. I have relied on PAGE CONTENT (as in content is king) for serp insert and ranking, as that has always been what the USER seeks: Content. The title that fits is the title to use, and I keep those short and sweet.

No, with reflection, I suppose I'm not the only one.

tangor

5:02 am on Jan 22, 2016 (gmt 0)

@Walt Hartwell and everyone else:

There is no magic way. You is what you is (bad grammar intended). One either has content, thus inbound traffic, or one is gaming the system, which will be caught out and demoted,

Only way, these days, to get past that is to have a few million to throw into the game for paid advertising, ensuring an above the fold first page serp result. I don't have that, the millions of course (sigh), but showing up 5th, 6th or 7th place doesn't hurt one bit. No dollars expended, only commonsense.

Walt Hartwell

5:50 am on Jan 22, 2016 (gmt 0)

I'm of the opinion there are quite a few shades of gray, meaning it is possible to have quality content and also perhaps not quite fully conform with rules not quite cast in stone.

tangor

6:03 am on Jan 22, 2016 (gmt 0)

I'm of the opinion there are quite a few shades of gray, meaning it is possible to have quality content and also perhaps not quite fully conform with rules not quite cast in stone.

You intrigue me, sir! What is the "not quite cast in stone" thingie?

Walt Hartwell

7:07 am on Jan 22, 2016 (gmt 0)

You ask more brother than I normally care to address.

Quality content is simply providing users with information they provide valuable. It might just be the start time for a movie close to them or it could be an in-depth article about the progression of Parkinson's disease and the ways to slow that progression.

If you, as a content provider, nail what users need, search engines will analyze, dissect, rate your pages, and rank them against other pages providing similar information. Best user satisfaction gets higher placement.

That's what should/could happen in the best possible world of search engine positioning.
It doesn't always happen that way though.

Search engines typically provide guidelines for how to show up well in their search results. Those guidelines probably work pretty well if you want to rank for the best Sushi stand in Fargo, ND. More competitive terms require more effort to rank well. Guidelines are not cast in stone.
I'll find it interesting how mods treat the "sushi stand in Fargo, ND" string. I'd leave it just as the example it is, it can't be very competitive.

fathom

8:30 am on Jan 22, 2016 (gmt 0)

From an onpage vantagepoint only it is clear that the better potential will come from all elements supporting each other so no matter what order JohnMu might hint at, that doesn't include most signals or the most powerful ones.

Internal links to the page are worth substantially more.

fathom

8:58 am on Jan 22, 2016 (gmt 0)

That's what should/could happen in the best possible world of search engine positioning.

It doesn't always happen that way though.

So by default your claiming the core algorithm automatically changes at will, makes unilateral decisions like a brain... e.g. BrainRank is truly "the wizard of Oz".

glakes

11:12 am on Jan 22, 2016 (gmt 0)

If you really think Google is akin to a "spinning scraper," you have a choice of easy-to-implement defenses: Robots.txt or <meta name="robots" content="noindex">.

Or, if you want Google to send you traffic, you can simply accept the fact that search engines get to decide how they display search results.

So what you are saying is that Google has the right to modify our work without our consent in order for us to participate in the digital economy. Such a nonchalant attitude is part of the reason why Google could care less whether they rank the original works or stolen copies. As long as Google can surround it with ads, it's okay for them to do as they please because they send traffic - sometimes more to the stolen work.

Instead of accepting that some multinational for profit corporation can use our work as they please, we should instead be asking regulators why Google is able to alter our work with zero accountability.

fathom

12:35 pm on Jan 22, 2016 (gmt 0)

So what you are saying is that Google has the right to modify our work without our consent in order for us to participate in the digital economy. Such a nonchalant attitude is part of the reason why Google could care less whether they rank the original works or stolen copies. As long as Google can surround it with ads, it's okay for them to do as they please because they send traffic - sometimes more to the stolen work.

Instead of accepting that some multinational for profit corporation can use our work as they please, we should instead be asking regulators why Google is able to alter our work with zero accountability.

I won't make claims as to Google's rights but you did consent. Not consenting suggests not being indexed.

This 124 message thread spans 5 pages: 124