Forum Moderators: open

Message Too Old, No Replies

Web design firms & the Florida update..

Looking for input from others..

         

SlyGuy

5:11 pm on Dec 2, 2003 (gmt 0)

10+ Year Member



From another thread:
Posted by sparticus

As a web design company we've always put a link to ourselves on every page of every site we build (right next the the copyright info at the bottom). Until now we've always formatted it this way:
Copyright © 2003 Client Name ¦ Created by Widget Web Design

..With the words 'Widget Web Design' as the link back to our site.

'Web Design' happens to be part of the company name, but are we supposed to change our link strategy now because web design is a commercial term?

Any ideas?

Posted by Marcia

sparticus same thing here, I've been tempted to start a thread to see what web designers were experiencing.

Well, I was thinking about it too, Marcia ;)

Our Particular Situation

Our company, for various reasons, uses the "Website Design - Widget Media" at the bottom of the webpages that we create for clients. Plain text, nothing special.

During Florida, we dropped from page 1 of the SERPs to page 17 or so, for several keywords.

Referrals from Google are down 50%. We, along with about 40 other design firms in the city, dropped like a bag of ball bearings.

What I've Noticed

We're not the only ones.

Take the keywords "web design" and add any city/state/province to the beginning or the end (example - "web design calgary"). Take a good look at the results for your area.

I'm not referring to the "relevence" of these results, I'm just curious to know if other web design companies have noticed a complete overhaul of the SERPs in their markets?

Is there a corrilation between the anchor text a design company uses and the current Google results?

I'd love to hear from the other designers out there.

Thanks..

Please note: this thread is not to intended as a complaint form for the current Google results.

steve55

1:47 pm on Dec 10, 2003 (gmt 0)

10+ Year Member



Interesting reply..

Just wondered because I have just appeared it the top ten on www2.google.com for one of the most commonly searched for phrases relating to web design (according to wordtracker)...after making such a change.

Bobby

6:24 pm on Dec 10, 2003 (gmt 0)

10+ Year Member



Steve55,

I use mostly image based backward links and have been very successful with them.

My 3 word keyphrase which was previously at the top spot for singular version is now at the top spot for plural version.

ex: blue widget firms

It is no longer anywhere in the top 100 for the singular now.

Nearly ALL my image-based backlinks have "firm" in the alt tag and I am trying to assess the reason it has lost ground.

Coincidentally the word "firm" appears 2 times and the word "firms" appears 2 times, but the title has "firm" as well as the H1 tag.

I wonder if stemming plays a role and "firms" is now viewed by the spider as appearing 2 times while "firm" is viewed as appearing 4 times (inside the word "firms"). Maybe "firm" has been overoptimized.

Ideas?

glengara

1:33 pm on Dec 11, 2003 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Bobby, be aware there's speculation G may now be ignoring Alts from linked images.
[webmasterworld.com...]

steve55

3:00 pm on Dec 11, 2003 (gmt 0)

10+ Year Member



For info - I didn't use any alt tags for my backward links..

Bobby

3:11 pm on Dec 11, 2003 (gmt 0)

10+ Year Member



glengara,

Interesting little test you did with "tintle bintle".

I haven't seen the famous "these terms appear only in links pointing to this page" for quite a while come to think of it.

steve55,

I think the benefit of the link regardless of whether or not the alt tags are factored into the ranking is that any link pointing somewhere must have some value, and if the sites share a common theme I would guess it helps the rankings.

I have a series of alt tags in links that all appear on the various clients I have, but it's too complicated to assess the value of the tag with respect to ranking, yet they all appear.

TheWhippinpost

1:16 am on Dec 12, 2003 (gmt 0)

10+ Year Member



Good post Marcia and others, and one that "lends" itself (amongst other theories here) to sommat I've increasingly been thinkin about recently.

I think the answers must conceptually lie within the bigger picture and strategy that G set for itself some time ago - I use the words "conceptually" and "picture" because G has striven through it's technology to gain a higher-level view of the web from which it can better see relationships and patterns (like the Kartoo site in a sense but computationally)... coupled with the underlying philosophy of how it thinks relevance is judged... but that leads to a question:

How does one discern an "authoritive" site from a "spammy/commercial" over-optimised site? In light of the persistent remarks made about anchor-text post Florida, a picture is emerging - in my mind at least - it would seem to me that one way to distinguish them today might run as follows - Note: this is an analogy not an algo!

If a page on a site (Site A) has the same 10 KW-loaded inbound links, let's say, for the sake of argument, that G now counts them only as one - hence affecting PR and all that entails.

If a similar page on a similar site (Site B) also has 10 inbound links but with only 2 KW's in the anchor text, and the rest comprising of anything from the kind often posted in forums such as ; http:/www.someplace, or "here", or perhaps just the name of the site etc... then the "penalty" would only be a dilution of the 2 KW links, thus, Site B will have an improving PR overall - This could explain why reports of a parent company for example, have recently been doing better than it's SEO'd franchises/branches etc... because the links they get could be more voluminous and varied.

We could extrapolate this mathematically so that if Site A has 30 KW-loaded links, ie... 3 KW's x 10, then we could say that G counts them as just 3. It's still awarded PR/importance but it takes much more to build it up... hence why dynamically-generated doorway/cloaked pages maybe (alledgedly) getting results.

Of course this model is penalising heavily, that's 'cos it's simple for illustration purposes; it could be that a certain density of identical anchor text needs to be hit before tripping the filter - and that could be made relative to perhaps other classic SEO incidences like title, <h1> tags for instance and also compared with both the "awarding" site and the "receiving" site... Don't forget that G has been experimenting with "word clustering" for some time and if it wants to eradicate, or reduce the influence of crosslinking between same-owner SEO'd sites it could compare, using clustering, the "sameness" of the KW's and weigh accordingly.

So how does that differentiate an authoritative site from a commercial/spammy/SEO'd one?

Well, truth is, a "genuine" content-authority-site - say medicine for example - will (generally) have numerous pages covering all manner of topics - as opposed to topic-specific Viagra sites - and for the most part, the general lay-person doesn't link to them with targetted keyword/phrases but there will be a few that "accidently" do, and they'll be others that use perhaps a secondary KW relating to the linked page. So while they may take a "hit" if too many incidences of KW-links trip the filter, overall, the other inward links will counter it and that would explain, IMO, why shopping sites a'la Kelkoo etal are reportedly doing (seemingly) well in the SERP's.

If one couples that with the ability of G's algo to process the surrounding words of a link also, it's possible that G can say, '...hey, we don't REALLY need to prescribe so much importance to anchor text - In the REAL WORLD, people don't link using KW anchor text extensively, only SEO'rs".

I believe one has to think of the larger picture coupled with history to give us an insight into G's strategy - Pagerank, algo "upgrades", clustering, Froogle, AdSense, seemingly niche-specific SERP's "weirdness" at various intervals suggesting a filter test, the Toolbar etc...

An operation with the vision that G has will no doubt have 1, 3 & 5 yr plans each pointing in one direction with multiple projects. One by one they eventually become live and bolt-on synergistically with others. I believe we're seeing G approaching the final phase of the bringing together of the pre-IPO business plan.

When Froogle has rolled out and is fully operational, it may become all apparent - G seperates commercial from content-driven sites and "educates" webmasters in the process - through Pavlovian-based rewards & punishments - to optimise accordingly - so better "cleaning" up and delineating the SERP's and making life easier for the user.

This is already getting lengthy and I really need pictures to help me explain fully what I'm really trying to say but I will say this; if G pulls it off, t'is very clever ;¬)

PS... I don't know whether "penalise" is the right word, I just think perhaps that the leverage that anchor text once attracted is now maybe geared lower, which is a penalty for those that "torqued it up" and now feel deflated but in the scheme of things still has weight.

This 36 message thread spans 2 pages: 36