I think that's a great answer from Chris - invest in some analysis. Figure out what you need rather than go with X, Y or Z. Base the spend on your requirements and defining those requirements if needs be. Hope that helps!

Yep - it's back. Looks like resolving the canonical issue fixed it. Seems it was a usual suspect after all.

Yep - bit of a weird one but in the end looks like the canonicals were the issue. Thanks for taking a look though man - super appreciated.

Hey Bernadette - thanks for the feedback. Site is back in the index now, looks like the canonicals were the culprit but the owners are keen for no future issues so I will dig in and take a look at these points. Cheers!

Hey folks

24 hours after we identified and fixed the canonical issue the site is now indexed again so it does look like it was indeed a canonical conundrum. Both the HTTP and HTTPS sites were claiming to be the canonical version so in some respects creating a conflict. We removed this conflict and it is now indexed.

Thanks for the extra eyes folks - appreciated and if anyone ever needs another pair of eyes to look a problem give me a shout.

Cheers
Marcus

Hey Krzysztof

Yeah, the page has little textual content but... neither does the amazon homepage. Ultimately the page is a jump in point for all the products and the content suits that. Certainly, I could understand Google not liking the page but would that not result in a reduced rank rather than a complete removal like this?

On the dodgy links front they have never done anything on that front - so anything there would be surprising (or just incidental cruft that is out there on scraper sites and the like).

Super odd.

Yep - super odd. 15 years or so in this game and never seen anything quite like this. Transient drops but usually it boiled down to some simple technical error or more often user error cough no index / robots.txt cough

Hey - the real issue here is the page is just not indexed. It's not there. Not that another page is a more suitable or preferential result. Ultimately that was the best page for a user to jump in at... The page is not even returned in a brand search so... can't see how any other page could be more suitable for that kind of search.

Hey folks

Hoping to get some more eyes on a specific problem I am seeing with a clients site.

Site: http:www.ukjuicers.com 

We have checked everything we can think of and the usual suspects here are not present:

• Canonical URL is in place
• Site is shown as indexed in search console
• No Crawl, DNS, Connectivity or server errors
• No robots.txt blocking - verified in search console
• No robots meta tags or directives
• Fetch as Google works
• Fetch & render works
• site command returns all other pages
• info command does not return the homepage
• homepage is cached and cache has been updated since this issue started: http://webcache.googleusercontent.com/search?q=cache:www.ukjuicers.com
• homepage is indexed in yahoo and Bing
• all variations redirect to the www.ukjuicers.com domain (.co.uk, .com, www, sans www etc)

The only issue I found after some extensive digging was some issues with the HTTP and HTTPS versions of the site both being available and both specifying the canonical version as themselves. So, http site used canonicals with http and https site used canonicals with https. So, a conflict there with the canonical exacerbating the problem it is there to solve.

The HTTPS site is not indexed though and we have set this up in webmaster tools and now the web developer has set redirects to ensure all versions even the https now 301 redirect to the http://www.ukjuicers.com page so these canonical issues have been ironed out.

But... it's still not indexing the homepage.

The practical implications of this are quite scary - the site used to be somewhere between 1st and 4th for keywords like 'juicers', 'juicer' etc. Now they are bottom of page 1 or top of page 2 with an internal page. They were jostling with the big boys (amazon, argos, john lewis etc) but now they are right at the bottom of the second page.

It's a strange one - i have seen all manor of technical problems over the years but this one seems to defy sensible explanation. The next step is to do a full technical SEO audit of the site but I am always of the opinion that with many eyes all bugs are shallow so if anyone has any input or experience with odd indexation problems like this would love to get your input.

Cheers
Marcus

Nope. No need for images. They just know about the content and link to it. The cached HTML shows they store a copy (or cache) of the HTML though. I could be wrong about the images but that would exponentially increase their storage needs so it seems unlikely.

If you have to ask that  question, I would say you are using too many.

The on page tool looks for one instance of a keyword in either a strong, bold or em tag so if you are doing this for SEO, the prevailing common sense  would be that any more than that one is too many.

If the highlighting provides some kind of benefit for your users then you can do more but if you are looking at bolding lots of text for SEO - then forget it.

The best  answer is to just use common sense and create something that is readable with the important keyword highlighted if you can do so in a natural way.

Hope it helps. 
Marcus

If the client is okay with it, then a footer link to credit the designer is okay, but it should always be nofollowed else you risk problems. You could do it branded but if you are not looking to manipulate the results, then nofollow it. If you want some SEO benefit from this then you are trying to manipulate the results.

Give this a read:

http://wpmu.org/wordpress-penguin-google-matt-cutts/

Then this:

http://www.seomoz.org/blog/how-wpmuorg-recovered-from-the-penguin-update

Then nofollow your credit links.

Additionally, footer links are now listed under Link Schemes in the Google Webmaster Guidelines as follows: "Widely distributed links in the footers of various sites"

http://support.google.com/webmasters/bin/answer.py?hl=en&answer=66356

As an alternative, if your client is happy with it, you could always have a credit page on the site or a blog post where you get a link from the site owner and the nofollowed footer link could point to this blog post or page but I would not assume this was always okay with your clients.

Hope this helps you resist those dark link building urges that will only hurt you downstream!

Marcus

When I set up my initial site, maybe 10 years ago (shudder) I bought BowlerHatSolutions.co.uk - the solutions means absolutely nothing, nothing, it's just a tired, horrid cliché in the same vein as every design shop using SomethingCreative.com so when we got the chance to go after www.BowlerHat.co.uk we wanted it, we wanted it bad.

So, I set a price in my head, and haggled with the domain owner for nearly a year and eventually, he came around to my way of thinking. It was a good deal more than £200 but was very valuable to me and to my brand - for a UK based SEO and Web Design the Bowler Hat icon is a really strong identity so... Well, I wanted it and eventually got it.

What are your other options?

www.I-wanted-whitby-jet-but-this-was-all-that-was-left.com?

Marcus

Hey

I agree with Michael in that largely, directories are dead. However, the evolution of directories and sites that add value to the listings are alive and kicking. Yelp (not so much in the UK) is one but TripAdvisor is really a a directory of sorts and that provides boatloads of additional value.

In the UK we have several directory types that are alive and well:

• TripAdvisor
• Trade Directories - plumbers, electricians etc
• Care Homes

My father is actually working on a directory at the moment - it is a list of all pubs in the UK that are dog friendly. It will detail everything for dogs, walks in the area etc etc. It will allow for

So in some respects it is the wrong question - Google does not punish wholes categories of sites. Rather they demote sites that have no real value and don't add anything (or that are set up to help other sites rank and for no other reason).

To make this work you have to ideally have a USP and provide a level of value that makes it desirable for folks to look at your site rather than the sites of the businesses in this category and to build up enough traffic so the business owners want to pay to be listed and get a preferential listing etc.

Hope that helps
Marcus

Hey,

There is a general consensus that 300 words or 500 characters is enough but I tend to think that is borderline useless advice here. How long should a blog post be? My answer would be long enough to fulfil the promise of the title and short enough to not be boring and overstay it's welcome.

Some great blog posts may be 100 words, some fantastic posts may be 10,000 words. It really depends on what you are talking about and how long it needs to  be to accurately do what it says on the tin.

There is a tendency in search to try and do SEO by numbers. We want 300 words. Well, 300 words does not denote quality. Quality denotes quality. If your main criteria when writing blogs or hiring writers is to hit 300 words then you have already failed. If your main criteria is to provide high quality answers that don't already exist or to do things way better than they are already being done then you are on the right track.

That's a somewhat philosophical answer but a better one than any general advice you may get re word or character count. If you want a more specific answer tailored to your situation then look at the blog posts that do well in the niche you are targeting. How long are they? Could some of them benefit from more detail? Could some of them be trimmed a bit to provide a more succinct answer?

The only caveat to this advice would be to avoid going too long and if you have something that is thousands of words long consider making it part of a series and having several posts that make a greater whole. Think of them as chapters of a book or report and that these parts can then more accurately fulfil the promise made by the whole. A great strategy can be to write up a series like this and then to create an overview or summary that links out to the more detailed sub posts and you then target the high level and more detailed keywords within your topic.

Hope that helps!
Marcus

Competitions are always good but really, you want to have a rich social presence that is driven by your content strategy.

So, you post great content on your blog - people read you and follow. They may even then tweet it through your social plugins and you may pick up more readers / followers.

You have a strong presence on twitter, post interesting things, interact with other interesting people to build relationships and ultimately pick up more followers like that as well.

Then, you have strategies for getting followers like competitions where  people have to follow you.

Really though, get some good content on the site and tie in twitter very closely with your content strategy and be active on twitter and you will start to build followers.

No silver bullet really, like everything else, the real answer after all the bluster is simply - hard work.

Marcus

The simple answer is... forever, unless you have some pressing reason to remove them. Certainly, I would keep them there till I knew there was no indexation of these other links and no requests for them for a say, 12 month period.

If you have external links pointing to any of these old pages, then again, that is another argument for keeping them in place.

Just leave them there, comment them as legacy in redirection script and forget about them and focus on something else.

Hope this helps.  
Marcus

Hey Philipp

Okay, SEO aside here and assuming folks are landing on your site and not on the section of the site that is targeted to their location / language (which is a whole other discussion but not your question as far as I can tell) then you want to redirect these users to the correct language pages.

So, we have two options here:

1. HTTP 301 - Moved Permanently: Now I don't feel this is correct. The resource has not moved, it is just not correct for this user based on their language preference.
2. HTTP 302 - Found: This is used to indicate the resource has temporarily moved to another location so is maybe more suitable as results from page A will not be completely ignored

In fact, if you dig into the HTTP status codes documentation a better option here would seemingly be a 303 which is classified as 'The response to the request can be found under another URI' and for my money that is more suitable. But, problem is, no one seems to use the 303 redirect and everyone seems to use the 302 in it's place.

So, I ask myself, what does Google do? When I visit www.google.com from the UK I am always redirected to the www.google.co.uk site. Is this a 301, 302, 303 or something else entirely? So, I checked quickly in webbug (or you can do it Chrome by looking at the Network tab in Tools > Developer Tools > Network Tab) and it redirects with a 302 status code.

Request: HEAD / HTTP/1.1
Host: www.google.com
Connection: close
Accept: /
User-Agent: WebBug/5.0

Response: HTTP/1.1 302 Found
Location: http://www.google.co.uk/

So, I am not sure there is a definitive answer as from a search engine perspective we would want to folks landing on the right page due to our geo location and language targeting but that does not escape the need to show people the content in the correct language.

If this was me and I was dead set on a redirection I would go with a 302. I can't claim that is an authoritative answer but it is certainly my opinion based on my research here.

I guess the alternative would be to maybe detect the users language settings and load a pop up that then allows them to select and redirect so it is not done at the request / response level but rather a choice the user makes themselves (then maybe cookies or other options could be used to deal with language for those users on subsequent visits). In the UK http://www.babycenter.com/ does this and it pops up asking me which version of the site I would like to visit (choice is always a good thing).

Possibly a good question to ask in the Google Webmaster Help Forum: http://productforums.google.com/forum/#!forum/webmasters as you will often get a Googler pop up to answer specific questions or this question may well have been asked before (albeit in a different way). If this is not search focused then really it comes down to what you think works best for your users.

Hope that helps! 
Marcus

Some further reading if it helps:

• http://support.google.com/webmasters/bin/answer.py?hl=en&answer=182192
• http://support.google.com/webmasters/bin/answer.py?hl=en&answer=62399
• http://www.mattcutts.com/blog/seo-advice-discussing-302-redirects/
• http://en.wikipedia.org/wiki/List_of_HTTP_status_codes
• http://en.wikipedia.org/wiki/HTTP_302

Hey

As Peter mentioned, in principle, this is no different to any other kind of office address. In practice though, you may find some issues as there may be several other tenants who have had this address previously and subsequently there will be a whole list of different names, phone numbers (2/3 of the Nap), website addresses and other info associated with this address.

Also, the Regus building will likely have lots of other businesses in there at the same actual street address which may create further noise.

The general advice is simple.

• get registered with Google+ Local
• build citations at the important
• optimise the website with local in mind
• generally try to ensure the website is high quality
• Try to be active socially, publish content, build local links, encourage honest reviews etc

But, my approach here would also include a detailed audit of all business listings that exist for that address. Get everything in a spreadsheet and contact the sites one by one to either update or remove the listings. Also, to identify if there are any Google local listings that still exist for the old businesses so these can be updated and removed.

Google is probably your best tool here and a search using the elements you know such as office number (Post Code / Zip Code), address will be the first starting point.

"[Office Number]" "[Street Address]" "[Post Code]"

My business is in an office complex called the Custard Factory in Birmingham and the search above for us would look something like:

"112 Zellig Building" "Gibb Street" "B9 4AA"

This should give us a good starting point. From here we will likely find several other interesting pieces of information we can use to refine these searches:

• Business names
• Phones Number

It's not search based buy you may be able to ask the Regus folks for a list of all previous tenant names of this office and that could give you a good running start here.

Then, it's a process of more traditional searches for those businesses and any information relating to them.

"business name" AND "post code" -www.businesswebsite.co.uk

So, for our business address that would be something like

"Bowler Hat" AND "B9 4AA" -www.bowlerhat.co.uk

This will give you a list of results with the business name, post code and remove any listings for the clients website. You may have to play with this a little as the business may have name variations but this will be a good starting point for further investigation.

Citation based factors (Quality, Consistency & Volume) pay a large part in Local SEO and simply building new citations if there is lots of noise out there and some active local listings may not do what you want it to (Or I could just be really OCD about all this stuff).

I suggest a read through at least the foundational ranking factors here but this should give you plenty of scope to get started: 
http://moz.com/local-search-ranking-factors

Hope that helps!

Marcus

I agree, get a great domain name that is short & user friendly and concentrate on building the brand. You can still SEO the site and as Elias mentioned, any benefit of exact match domains is being slowly chipped away so concentrate on your users.

I wish I had known what I know now when I registered my company and domain name!