| 10:59 am on Jun 23, 2006 (gmt 0)|
I would add the meta descriptions and check the page titles. The more information the better.
Google will crawl when it wants to. The best way to encourage it is to get new inbound links.
| 12:38 pm on Jun 23, 2006 (gmt 0)|
If you decide to use meta description tags, make sure you use a unique description on every page and keep it 55 char or longer. Otherwise Google will still tag on your nav links at the end of your meta descriptions.
| 1:15 pm on Jun 23, 2006 (gmt 0)|
Don't want to discourage you, but as I mentioned in this thread [webmasterworld.com] Google doesn't seem to be taking any notice of the meta tags at the moment.
In fact you described to a tee what shows up on a site: search for one of my sites. Instead of the unique meta descriptions, I'm seeing the navigation links and stuff like "page 1 of 2" showing up for every URL of the site, with the result that Googlebot seems to think every page is the same and the site's gone supplemental.
This is a four-year old site which used to rank well and show unique meta descriptions for each page. Not amused Google. :(
| 3:32 pm on Jun 23, 2006 (gmt 0)|
I respectfully disagree with malachite.
Recently, meta descriptions have become very important to Google; pages with none are often dropped; pages with non-unique descriptions are made supplementary. I believe malachite's problem is a different one.
For example, no-one (except a webmaster) searches for 'domain.com', so looking at that serp will not give you a clue as to what the 'typical searcher' is seeing.
Descriptions should reflect the page content, include rlevant key words, for two reasons:
1. Google likes it that way, and a happy Google smiles on you - like it or lump it :)
2. The text that appears in the serps depends 80%* on the search term - if the meta description it, then the meta description is likely to be used. If not, then it is most unlikely to be used.
20%* is down to which datacenter the search utilized - some may prefer the ODP description (where there is one), one or two may ignore both.
*Percentages are speculative; contents may settle during transit.
| 4:20 pm on Jun 23, 2006 (gmt 0)|
So how do I solve the problem of the navigation links (E.g home, login, products, etc) appearing at the top of the page?
If I put the page content the first thing in the source code, but use CSS to display the menu on the top (e.g a human will see the menu at the top but a bot will see the content first), will it have any effect on boosting my rankings?
| 4:34 pm on Jun 23, 2006 (gmt 0)|
Some folk believe that's so, but I've seen none come back and say "Wow that really worked".
We know that Google looks at the whole code, for all sorts of reasons (eg checking that TITLE reflects content), and I suspect it matters very little what order things come in.
| 4:44 pm on Jun 23, 2006 (gmt 0)|
Quadrille, feel free to disagree! And if you've any tips on why the situation I outlined has occurred, do share :) I'm groping in the dark as much as everyone else on here at the moment, wondering what's up with Google.
All I'm saying is that in common with several other posters throughout this forum, I'm seeing odd stuff in the results.
I'd just like to know why Google, which used to like and display the unique meta-tags each page has, has suddenly started ignoring these and replacing them with navigation links? If meta-descriptions are *so* important to Google, why has it stopped showing them? It's been using them for 4 years :o
| 5:17 pm on Jun 23, 2006 (gmt 0)|
I wish I had the answers! :)
I suspect that they are experimenting on several fronts, which is worrying, but we'll see.
But I also strongly suspect that Google is deliberately working against 'webmaster searches'; I'd certainly only worry about 'bad serps' after a keyword search, not after a domain search.
| 5:21 pm on Jun 23, 2006 (gmt 0)|
<<For example, no-one (except a webmaster) searches for 'domain.com',>>
that's not necessarily true. when i worked for a mid sized newspaper, month after month, the top searches were for the domain name or variations of it. people type into search engine boxes like it was the address bar.
| 6:14 am on Jun 24, 2006 (gmt 0)|
I just tried searching for one of my pages using two search phrases. The first one contained words that are all in the page's meta description tag. The resulting SERP used the description tag. The second search phrase contained only words that are not in the page's meta deacription tag, but are nevertheless prominent on the page. The resulting SERP used snippets from the page text (containing the search words) as the description.
This is exactly the way I remember it used to be, quite some time ago (a year or two?). If descriptions have been behaving differently lately, perhaps Google simply changed them back. Under these "rules," I wouldn't expect a "site:" search to show meta descriptions, because it doesn't search on specific words, so there's nothing to match on.
| 8:47 am on Jun 24, 2006 (gmt 0)|
Well, you learn something new every day! Thanks!
|when i worked for a mid sized newspaper, month after month, the top searches were for the domain name or variations of it. people type into search engine boxes like it was the address bar. |
But that's still not an issue here - such 'searchers' know what they want, and so will click on the site they chose, regardless of the detail.
Someone who is genuinely 'searching' - ie has not already made up their mind, is pretty unlikely to type a domain name; a key word or phrase is the norm, and so i still believe those seacrhes are the ones the matter.
The reason I make this distinction is simply that many folk do not have their domain name either in their description or in the ODP description, so searching for it often gets a very different result compared to a 'genuine' search.
| 9:59 am on Jun 24, 2006 (gmt 0)|
Moving content above navigation menus using CSS will generally help. Google parses HTML to build description snippets (as opposed to just stripping away HTML and looking at what's left) when it doesn't find a META description tag and it does a multi-phased scan from top to bottom, using HTML elements like H, NOSCRIPT, HR, DIV, TABLE, P, ALT and other stuff as markers. Little things like H1 text wrapped in HREF can make Google pick up your navigation text instead of your page copy.
| 6:21 am on Jun 25, 2006 (gmt 0)|
"Moving content above navigation menus using CSS will generally help."
Could you be so kind as to point me to a post here that gives a "how to" on this... I'm familiar with css, just need to see a code example to do this to my own site.