| 11:01 pm on Dec 10, 2008 (gmt 0)|
Good spot, pageoneresults. We've seen some other "granularisation" of Google SERPs this year too, from the tagging with dates, to snippets about forum posts.
It's not exactly "building pages for users", but I find it hard to argue when Google want assistance with their results pages ;)
| 11:02 pm on Dec 10, 2008 (gmt 0)|
If I search for a book do I want to see 500 descriptions like that or do I want to see an exciting description of the book in someones own words?
BTW that's been in the guidelines for at least a few months.
| 11:07 pm on Dec 10, 2008 (gmt 0)|
|do I want to see 500 descriptions like that |
I'm thinking that Google want to process the information or at the very least reformat it before returning it in results. A bit like the idea behind microformats [google.com], but those never really got enough traction.
| 12:10 am on Dec 11, 2008 (gmt 0)|
|If I search for a book do I want to see 500 descriptions like that or do I want to see an exciting description of the book in someones own words? |
That's where your knowledge of Tagged Facts along with a well written IPW opening sentence may give you a slight advantage over the others. ;)
|BTW that's been in the guidelines for at least a few months. |
A search within WebmasterWorld makes no mention of it. I will typically search first before posting to make sure it wasn't discussed recently.
From my perspective, these are very solid suggestions and for Google to specify that they are now looking for and utilizing Tagged Facts is news for us who like to focus on that particular element. I know, there is an entire group of those who feel that the meta description has little to no value in the overall scheme of things. I tend to think otherwise.
<meta name="Description" content="News and Discussion for the Web Professional. Company: WebmasterWorld.com, Address: 3801 N Capital of TX Hwy E240-181, City: Austin, State: Texas, Zip Code: 78746, Telephone: 512-231-8107, Fax: 512-231-1653">
Oh the possibilities...
In searching further on this, I see related discussions back in 2007 October. But none on WebmasterWorld, that is odd.
Ah, found the original Google Blog reference...
Improve snippets with a meta description makeover
2007-09-27 - [googlewebmastercentral.blogspot.com...]
^ One year later? Man, I need to get out more often.
| 2:44 pm on Dec 12, 2008 (gmt 0)|
In doing some back tracking of meta description development amongst various sites that I manage, I've used this Tagged Facts method by default for different applications not knowing what to name it.
For example, I may have a listing of professionals who are in the process of branding their name and credentials. They may do this through a profile page of some sort. While reviewing some of these profile pages, I'm noticing Google showing "Show map of" listings for those where Tagged Facts are either included in the meta description or are part of the first block of content seen after the <body> element.
I'm setting up pages now for experimentation using the Tagged Fact approach with a little more definition. In some cases, I might do something like this...
<meta name="Description" content="Company Name: ABC Company, Contact Name: John Foo, Address: 123 Anywhere Street, City: Foo, State: California, Zip: 12345, Tel: 714-555-1212, Fax: 714-555-1212">
What is a Tagged Fact?
According to Google's unwritten definition, it is a Tag followed by a colon, space, Fact, comma, space, Tag, colon, space, Fact, comma, etc.
Tag: Fact, Tag: Fact, Tag: Fact
|In this example, information is clearly tagged and separated. |
<meta name="Description" content="Author: A.N. Author, Illustrator: P. Picture, Category: Books, Price: $17.99, Length: 784 Pages">
I've done some searching around a bit and cannot find a clear definition of a Tagged Fact. I'm wondering if there's a default listing of tags that should be used to clearly define the facts. For example...
Is it just the colon that defines the Tag?
[edited by: tedster at 8:36 pm (utc) on Dec. 12, 2008]
| 10:04 pm on Dec 12, 2008 (gmt 0)|
I think the tagged facts may be googles way of eventually moving toward more of a semantic web, by publishing the change and support of, in anticipation of the next developement phase which seems to be the semantic web. As soon as i saw this, I thought of yahoo's searchmonkey and the party i went to at their campus..
| 10:05 pm on Dec 12, 2008 (gmt 0)|
To come at this from another angle, perhaps see it as good copywriting advice.
There's nothing lengthy to read on a Google SERP, right? Users' eyes are flicking down the listings for a match, and reading a few words at a time. Scanning for matches - words that suggest they want a particular link, because it has the information needed to satisfy their query.
If that's true, then packing as much information into a description as possible has to be a good thing. I'll certainly experiment with this as a focus when writing them. So, for the contact example, a potential SERP might be:
|Contact Example Inc. |
Tel: 1234567890, Email: email@example.com, Write: Example house EX4 MP13, Shout: very loudly
I'm not sure the style or punctuation needs to be that rigid, but I freely confess to getting much to lazy with copywriting descriptions. I blame Adwords for that ;)
| 1:42 am on Dec 13, 2008 (gmt 0)|
|I'm not sure the style or punctuation needs to be that rigid. |
In finding all the discussion on this that I can, Google does make reference to the formatting of Tagged Facts indirectly here and there. For example, they have an example where there is no space between the comma and the next tag and they make a statement to the effect of that it is not good practice and provides a less than satisfactory visual.
I've already got my reformatting done on me test sites. I went through over the past day or so and refined some already in place Tagged Facts and then added them where they were not present. I'm interested to see the outcome. I can see some of it now from previous use but not to this micro level where I'm specifying each piece of factual information with a tag.
| 2:33 am on Dec 13, 2008 (gmt 0)|
What is ...
| 3:02 am on Dec 13, 2008 (gmt 0)|
Inverted Pyramid Writing
|Inverted pyramid writing uses the first paragraph of a page to serve as a concise summary or abstract of the page content. |
| 6:42 am on Dec 13, 2008 (gmt 0)|
|A good meta description can bring all this data together. For example, the following meta description provides detailed information about a book. |
<meta name="Description" content="Author: A.N. Author, Illustrator: P. Picture, Category: Books, Price: $17.99, Length: 784 pages">
Why no book title in this description? Is Google saving bandwidth?
As I understand it, Google selects snippet text by finding best matches for the words in the query. A page for book, I'd think, would probably come up in searches most often when the query is for the book's title. If this is the case, then the above is probably not a good meta description for a book... or not a sufficient meta description, at any rate.
If present snippet rules are followed, text on the page that includes the book title is more likely to be returned in the snippet than the above description would be... unless the use of tagged facts somehow changes how Google selects the snippet. This might be the case, but I didn't see it suggested.
And another consideration... if you're an author, bookstore, publisher, reviewer, or curious reader, you might prefer to have a more colorful description... perhaps one that at least touches on what the book is about (in less than 150 characters). Most librarians I know would prefer that too, though I've known some who would prefer the bibliography style listing.
[edited by: Robert_Charlton at 6:49 am (utc) on Dec. 13, 2008]
| 10:31 am on Dec 13, 2008 (gmt 0)|
I think in the book example, the title is likely to be the name of the book, and so is not really needed in the description itself. I've not seen any problems with Google choosing a description over a snippet even if the description does not exactly match the query - although there is the bolded-text CTR thing to consider, of course.
| 9:16 pm on Dec 13, 2008 (gmt 0)|
|I think in the book example, the title is likely to be the name of the book, and so is not really needed in the description itself. |
This is what I meant by "saving bandwidth". Yes, the title is redundant if it's in the page title, and it would be bolded in the serp. Google might not want it to appear twice.
The question is whether the use of tagged facts in the description affects the mechanism for selecting the snippet.
|I've not seen any problems with Google choosing a description over a snippet even if the description does not exactly match the query... |
I've generally seen that matches predominate. Perhaps there are other snippet selection factors in use.
My experience is that there is almost always some common vocabulary, and that other factors (like sentence breaks) enter into what Google actually displays within 150 characters. I've used these guidelines with some success to shape snippets returned for most likely searches a page would rank on.
If a tagged description without the title would be returned for book title searches, though, that's an indication that Google is considering other factors before it jumps to page text or other description sources.
The searches you're seeing without query matches might also have other such factors involved. And perhaps there are different "matching" criteria for different kinds of content and queries.
| 11:11 pm on Dec 14, 2008 (gmt 0)|
Very intereating. I have searched but cannot find any definative guidlines for "Tagged facts" other than the book example.
I can see how:
<meta name="Description" content="Author: A.N. Author,
Illustrator: P. Picture, Category: Books, Price: $17.99,
Length: 784 pages">
is potentially more useful to the user and could generate a higher CTR than a one sentence intro for a book, but how would this apply to other other industries?
<meta name="Description" content="Hotel Name, Example Hotel
Adress: 123 street adress, Town: Some Town, City: Major City name, Category: Hotels, Price: $99 per night,
A correct format for a hotel? - Are there any tagging guidlenes by industry? Also is it viable to use taggs and a hand written description too?
| 11:49 pm on Dec 14, 2008 (gmt 0)|
|Are there any tagging guidlenes by industry? Also is it viable to use taggs and a hand written description too? |
I don't think this type of meta description gets any special treatment, so at this stage I believe it is a copywriting task. Maybe a short line of text followed by some "facts" would be suited in your case?
Bob: I guess I've come to expect broad coverage with a meta description. But I'll definitely be doing some checking into how frequently they're actually displayed. I know for advanced searches (e.g. broadly, those contain any punctuation characters) you're much less likely to see a defined description, but for keyword searches, I'm close to relying on them to get a better CTR.
| 12:53 am on Dec 15, 2008 (gmt 0)|
This looks very Dublin Core-ish to me.
"The Dublin Core Metadata Initiative is an open organization engaged in the development of interoperable online metadata standards that support a broad range of purposes and business models."
Without a shared vocabulary (structure) these efforts have limited benefit.
"DCMI traces its roots to Chicago at the 2nd International World Wide Web Conference, October 1994."
14 years to develop a shared metadata standard?
| 10:38 am on Dec 15, 2008 (gmt 0)|
I agree its on the Dublin Core side. I assume that Google is "clever" to derive a structure for such taggs ( as the do with site links ).
It will be interesting to see if they add some sort of tagging defination/structure to Webmaster tools.
| 1:49 pm on Dec 16, 2008 (gmt 0)|
|Are there any tagging guidlenes by industry? |
Maybe they are going to come when Google launches an auto-generated web directory with categories like
Books -> Author -> Title -> Price
Hotels -> Country -> City -> Name
| 12:00 pm on Jan 12, 2009 (gmt 0)|
As a footnote to this discussion, I've noticed Google automatically including what could be described as "Tagged facts". The pages in question have a bullet-point list of benefits at the top of the page - Google is reformatting this by exchanging bullet points for semi-colons, and using this in the snippet.
E.g. bullet points:
- Free stuff
- Available in green
- Very tasty
Free stuff; Available in green;Very tasty
| 3:59 pm on Jan 13, 2009 (gmt 0)|
I've spent hours trying to find definitive information on "Tagged Facts" and cannot find anything other than what is listed at Google.
Improve snippets with a meta description makeover
2007-09-27 - [googlewebmastercentral.blogspot.com...]
|What's changed? No duplication, more information, and everything is clearly tagged and separated. No real additional work is required to generate something of this quality: the price and length are the only new data, and they are already displayed on the site. |
There's a bit more information at the above Google Blog post from Raj Krishnan, Snippets Team.
I think it all comes down to interpretation at the moment. There are surely quite a few default tags I would think based on the commonalities. You know, stuff like Author: and Price: which are used as examples in Google's documentation.
My guess? You create your own Tagged Facts by formatting in a particular way. It is the colon, the comma, and proper spacing that are key. If you have a colon that is naturally occurring in the wording, that may cause a break in the tag, I'm just guessing now based on all the examples I've seen.
It is interesting to see the find that RA posted above. Google is taking list elements and breaking them down into Tagged Facts. I wonder if the list were formatted more semantically if the tagging would have been different. Hmmm, let me go set up a test page...
| 4:11 pm on Jan 13, 2009 (gmt 0)|
Google should create new tags instead of trying to stuff the Description! (Which may still have a character limit.)
I remember old html editors used to have so many different tags for Author, Email, etc., a whole list of different tags.
Incidentally, today for the first time while conducting medical research on Google I saw authors listed in a line at the bottom of regular Google search engine result info.
Unless you're a librarian, people might prefer a sentence description which is much easier to read than all this jerky colons and commas stuff. New Google Tags solves that problem. It also gives a better user experience.
| 10:18 pm on Jan 13, 2009 (gmt 0)|
The description needs to be succinct, accurate and create a sense of "ohhh, that's exactly what I was looking for... click!".
I think "tagged facts", as described in that guideline, fail on the third aspect of a good description tag.
example, which would you click.
Widget: 9 inches tall, 3 inches in diameter, slightly obtuse w/500 word description on a standard template with two sidebars.
Widget: The atomic model depicted can make your life simpler by taking care of your pets every need.
I think humans still need a little encouragement to click because on a page with 9 bland descriptions the lively description is king.
| 11:01 pm on Jan 13, 2009 (gmt 0)|
And I thought the keyword meta tag was dead! Seems like it has been resurrected in the form of Tagged Facts.
| 1:35 am on Jan 14, 2009 (gmt 0)|
Ditto. You beat me to it. And of course, this is all Google specific. What would yahoo abs msn make of this type of description tag?
Why not simply add a new tag, or use the keyword tag?
| 1:53 am on Jan 14, 2009 (gmt 0)|
|Are there any tagging guidlenes by industry? |
Google base tags
Is it likely they will try to marry this data ?
| 8:13 am on Jan 14, 2009 (gmt 0)|
Potential Geek, I could not agree more, Google should just specify a special tag to put in webpages.
As for implementation, I can't wait to see this being rolled out to all those website's that don't follow such discussions. In fact if I were Google I'd look at websites that implement a "tagged fact" approach to their meta-description and, where found, immediately mark them down in the SERPS as the chances are that they are going to be using methods to manipulate the Algo!
| 10:01 am on Jan 14, 2009 (gmt 0)|
I may have jumped to conclusions slightly with the example above - I never realised that Google replaced bullet points with semi-colons when bullets appear in snippets.
That combined with the snippet-generation technique of picking whole phrases to show (rather than broken sentences) can result in a snippet that appears very "tagged fact"-like. But I don't know that it was a deliberate thing on the part of Google's snippet generation team.
But regardless of intention, the bullets did result in an attractive snippet.
| 10:31 am on Jan 14, 2009 (gmt 0)|
|<meta name="Description" content="News and Discussion for the Web Professional. Company: WebmasterWorld.com, Address: 3801 N Capital of TX Hwy E240-181, City: Austin, State: Texas, Zip Code: 78746, Telephone: 512-231-8107, Fax: 512-231-1653"> |
Given that you've got a sentence description and some tagged facts should the first sentence be separated from the tag fact list by a full-stop or a comma?
| 11:14 am on Jan 14, 2009 (gmt 0)|
|Is it likely they will try to marry this data? |
daveVK, that is the closest I've seen to a list of what appears to be the proper nomenclature for tagging. I'm going through that entire section now but I think I'm going to use that as a starting point for a Tag Library. Thank you for the reference!
| This 37 message thread spans 2 pages: 37 (  2 ) > > |