Frantic Fish

Hey Josh, hope you had a good Xmas / New Year. Design - like it, bright and contemporary looking. Not sure that the wallpaper and poppies gel, perhaps flat colour instead of a motif would work better. Less is usually more. SEO - 1) you're using an <h1> for the banner text on all pages. Personally I'd not use an <h1> for a strapline. I might keep it on the home page, but on the others I'd replace it with just a plain <div> and use <h1> where you're using <h2> 2) description tag identical on all pages, this is a quality factor with Google (Webmaster Tools warns you of dupe description tags). Make them unique. Keywords too for the sake of completeness, don't just use a default one for the whole site. 3) Echo what has been said about the text links at the bottom. If they were going to different pages then they could serve a purpose, but multiple anchors pointing to the same page (especially on the home page pointing to itself) is a technique used by spammers and you don't want that association. Footer links likely count for less than links elsewhere so consider adding links in the body text / main part of the page to important pages instead. 4) Best practice is to not link back to /index.php, instead link back to /. I think Google may have figured out that these are the same page nowadays but this way you guarantee that internal and external PR flow to the same place. Peter

Yep, Google favours keyword domains - absurdly so. If you have the natural keyword domain for your key phrase you have a MAJOR advantage over everyone else. Recently re-launched a keyword domain. It has two links - one from the business' old site (business name in url), and one from it's sister site (another keyword domain, the other side of the business). One page site and it was 1st in Google within a week for a useful term - 'widget service town' let's say. Outranks bigger, older sites and the manufacturer (household name) site too.

Definitely don't use Blogger - it's awful for SEO. WordPress can be customised to be very SEO-friendly. I'd have the blog on my site because a. increase pages on my site, gives visitors more reasons to stick around b. links to the blog benefit the site as a whole

It's not the .ae because that doesn't redirect or duplicate any of the content on the .co.uk - it's just parked. Do 'site:monocouche.co.uk' in Google and note that your home page is not the first result. This is a sure fire sign of a serious problem (or penalty). In fact the first three results on the site are 404s; every page I try to get from your site from Google's cache is a 404. If you've changed the urls then you need to redirect the old ones to the new ones using a 301 header response. Not saying that's all you need to do (because the home page should still be first when you search for 'site:') but it's the first thing I'd do.

I'm with thefandango; posting lists like this isn't that helpful for new users because a lot of the stuff in there isn't that black and white, even if it is right - and some of it is just plain wrong. What would be helpful would be to provide a list of 20 great articles on SEO; how to do keyword research, how to write content, how to structure your site, how to optimise your pages, how to look for quality links, links to avoid, techniques to avoid. These regurgitated lists that circulate only help to start myths. Here's my top 5 SEO tips: 1) Read the Experts Type 'SEO' into Google and pay attention to the people at the top for that term. Read their tutorials and lists. 2) Analyse the Top Rankings It's not that easy to reverse engineer Google, but they are showing you the sites they like best for any term. Crunch through citation profiles (not just links) until your eyes hurt, look for common factors. Look at the content on the top sites. What are they doing? 3) Test Put what you're learning into practice. 4) Record Everything. If you don't know exactly what you've done, how can you begin to guess what was effective? 5) Observe & Analyse What results have you achieved? What are your visitors doing? What pages do they like? What links do they click?

The SEOBook.com one is pretty good too.

This is a post I'm making on a number of sites, so apologies about the duplication, but it's important for me to gather as many opinions as I can. My question is this: is there any benefit to using WordPress or Blogger over and above creating my own blogging software for my site and my clients? I'm launching a new section on my new site. Part of it will be a blog with the standard topical industry observations. I'll also have articles - some just simple pieces, a few hundred words. Others will be multi-page researched pieces with graphs, comparison charts, data etc. Weighty stuff (at least I hope). I can get unlimited free programming from a knowledgeable PHP Linux guy (me pops) so I'm leaning towards getting a custom system developed that can be used as a template for client work. Clients I have right now will just be posting simple articles with comments - the classic 'blog'. Am I right to think that a custom solution would be better? Advantages I can see are 1) Security No-one can see what my back-end runs on, hackers will have to work far harder. 2) Cost & Time after rollout I don't have to get clients to host with WordPress to be fairly secure. If they don't host with WordPress, someone has to check for all patches, updates etc. Billable time (pain for the client) and not something I want to be doing anyway. 3) Lean mean install - not all the extra crap I don't want that comes bundled out of the box, I don't want to have to think about 'hacking the core' as it were. I really just want - text / images / links / CSS / doc upload - add / order categories for nav - comments (captcha, pre-approved) - pagination / archive - extensionless urls - search (via tags and detailed for all text) This is all stuff we've done together before and won't stretch my programmer at all. So my question is, is there any reason that I should consider pre-built CMS systems - and specifically WordPress or Blogger? The only one I can think of is the initial time (i.e. cost). Once you've practised a bit can you install and config a system that does what I mention above plus get it styled to match the look of the existing site in a few hours? Or am I missing something? Is there something else I'm missing about these systems that means I should consider them above a custom one? Very grateful for comments, observations, opinions, stuff I'm overlooking etc.

From what I've seen there isn't really a 'duplicate content penalty', more of a duplicate content filter for content used on multiple sites. Google's guidelines specifically advise against creating duplicate or similar pages or sites. Google will usually pick what it feels is the best (i.e. oldest or best linked) version of a page and display that. If you put yourself in Google's shoes, they're effectively saying "I know about this content already. Exactly why should I be interested in it again?" Sites that have their content duplicated on other sites rank, but the duplicators usually won't. I have seen dupe content rank if well-linked, or if only part of the page is dupe then original content on the rest of the page can help it rank.

It's also very risky. First, once you stop paying the monthly fee do all your links disappear? I've spoken to people who've had this happen to them. Second, if Google spots a network of sites selling links they will bring it down. This happens very often. The SEO company have made their money usually - the client on the other hand suddenly has a load of worthless links pointing at their site. Either way they've only rented rankings and traffic. There is a much better way to do SEO for the client. It's not quick, it's not easy, and the client has to get involved in providing content (whether writing themselves or paying others to do it). But to a great extent it is cumulative and the client keeps their traffic and rankings for months (even years sometimes) after they stop paying the SEO.

Google's load balancing means that you could hit different datacenters from the same search done from the same machine. Even querying the datacenter by its individual IP no longer guarantees that you are actually getting results from that DC.

As Google and other engines use geo-targeting more there are three things that help: 1) country-specific domain Google UK was hit a few years ago by a bug that killed non .co.uk sites in the 'UK only' results. Also this will help you host in another country and still get featured in 'UK only' results. 2) country-specific links From what I'm hearing about Google Caffeine, this is getting dialled up in the new algo. 3) country-specific hosting This has always been absolutely essential if you don't have a country-specific domain. In Google UK run the 'site:mydomain.xtn' search for 'UK results only'. If there are no results then you really need to host here. Related to (1) I just tried a .co.uk that I know is hosted in Germany and found it. I'd still host in the UK myself though - who knows if it is an extra factor?

Dangerous strategy. Link networks are usually doomed by definition. You need Google to notice the links in the network. Too many sites and Google will notice. A lesser risk is a peeved competitor for any one of the sites benefiting from the network shopping you, but this risk grows with the network.

Not a direct effect but there is a related effect. There are directories and organisations that will link to your site if your code (and that in your portfolio of sites) validates.

Googlebot will crawl a site as often as it wants. Big sites like the BBC get crawled multiple times every day; smaller sites less often.

They are both the same facility, just accessed by different routes.

I like extensionless urls too. Apart from the geek factor, they look better when used in adverts (on or offline) and are easier to give out over the phone.

The Local Results have seemed very vulnerable to multiple listing attempts by the same company for the same term. For locksmith and plumber searches I've seen half the results for the same business on occasion. Google obviously had to act on that, so it's likely that what you're both seeing is related to anti-spam measures they're taking. odmassive, I would recommend removing the listing for the new address and trying to claim the existing listing at the old Birmingham address, then updating the address. If you can't do this then you need to get the address changed in the sites that Google Local is using for information. If these sites are getting their data from a core provider (MarketLocation, the Local Data Company, InfoServe etc) then you update the address with them first and see what effect this has. Sites that use their data will hopefully update within a few months. Then it's a question of going through the list of those that have their own data / don't update and getting the address corrected. Could be a long hard slog.

That is not correct. A 301 says to Google (and any other bot) "This content has moved permanently to <new location>" If you change your urls your toolbar PR for these new urls will be non-existent until the next PR 'update' (when Google's internal PR is exported to the toolbar to be seen by the public). 301 redirects from one domain to another domain can cause trust issues with Google and that can affect ranking. And making changes to your urls on an old established site with lots of pages can cause a temporary blip in rankings while Google gets its head round the change. Spammers buy old domains and 301 then to new ones, so this is why Google can wait a while to transfer trust that an old url had to a new url. I've used 301 redirects many times when redesigning or optimising sites and have never had any permanent issues. In most cases I have never had any temporary issues. Here's Matt Cutts from Google on 301 redirects -

heh heh heh

Like others here I still include them just as good practice. Just because they aren't counted by Google for rankings, doesn't mean that they aren't used as a quality indicator though. Also, I find them handy as part of search facilities on larger sites. You can add mis-spellings that you wouldn't want in the page text and still return that page as a result for site search.

Click Through Rate (CTR) does also play a part in the ad placement. Google say that this is about showing the best ads to people, but it's a lot easier to understand like this : - ad 1 is bidding £1.00 a click, gets shown 10 times and clicked once. Google makes £1 - ad 2 is bidding £0.25 a click, gets shown 10 times and clicked five times. Google makes £1.25 Ad 2 is obviously what the searchers want, but it's what Google wants too. The example is a pretty extreme one but just to demonstrate the principle. Try different ad texts - Google will rotate them and choose the most popular ones automatically. Also make sure that you are targeting focused keywords rather than bidding on broad match terms. Bids for targeted keywords are sometimes less. If you're sure your ads are as targeted and clickable as they can be then your only option after that is upping the bid.