Welcome to the Q&A Forum

Hi! I wrote that article.

I asked Gary Ilyes about this in a middle of my talk at Big Digital in Adelaide last year as he was in the audience. He said yes, PDF links do pass PageRank. That said, I still use header canonicalisation to consolidate PDF into HTML files. Just to be sure.

Hrefang tag must point to translations of current page and itself. For example on https://www.memoq.com/de/ubersetzungsprodukte and other translations of that page you should add:

Use the same logic for other subpages and you'll be ok.

Cheers!

Both tools are right. You have been testing home page in flang.dejanseo.com.au where you have correct hreflang implementation, and that is why you pass the test.
But if you for example test https://www.memoq.com/de/ubersetzungsprodukte you will get errors, just like you are getting it in google search console. The reason is that hreflang tag is not properly implemented on that page.

1. If the SEO company has the ability to easily remove those links then you don't want them anyway.
2. Sadly, you may experience a slight drop, depending on your overall link profile.
3. On a positive note, you likely have a much cleaner link profile now.

If it makes sense to have them there then leave it, however if you feel there has been a traffic drop then it may be worth investigating if it was these 1700 indexed articles that have caused it.

Next step: Check your backlinks for unnatural anchor text distribution.

Panda still runs in installments, not continually. Rewriting content sounds like a massive task, hope it's worth it (e.g. is it better to write new stuff instead?). Have you got any pagination present on the site or indexable search results? We are under assumption here that you are certain that the problem is caused by Panda filter and not other factor and that your page layout and ads are not the cause of the drop or that the problem is not link related. I see no problem with 301 of pages with duplicate content to a new better content page. Sounds like something users might appreciate as well.

Submit a reconsideration request explaining the situation in as much detail as you can. Worst case scenario you will find out if it's a manual or algorithmic penalty. If no manual action has been taken then you will have no other option but to try harder to remove the links from dodgy sites.

My choice with be a rel canonical, however have you considered using rel prev and next?

Here are a few footprints which Google can use when evaluating your links (some are algorithmic and some manual if you are under audit/review and some are cross-referenced):

• Time of the link placement and any related groups of links generated within the same timeframe
• Link location (blog roll, footer vs content and navigation)
• Pattern of buying from known link sellers
• Strict use of exact match commercial terms in anchor text
• IP address / C-class
• Author name in article submission
• Open calls and invitations for link exchanges, selling, buying
• Widget anchor text manipulation

Or anything else that can be seen as a link scheme.

I would focus on local citations as much as possible and until all relevant places have been covered. While you're at it you will score links as well so probably the best way to start as it's easy and ROI is fairly good. Next step is identifying best branding channels - a very broad and difficult question - I guess your client's budget will be the helping factor and you'll have to decide what to attack first in line with budget size.

Prioritisation is not easy and you will no doubt hit the wall many times before you find the right solution.

You're about to target a multitude of potential search terms. What needs to happen is series of prioritisation activities. Your work starts of course with on-site review and understanding whether site architecture and the way home page cascades down to other sub-units passes link equity to other pages and if that arrangement is optimal and in line with your priority research.

Tools like GWT can be used to easily extract the highest ranking page for any number of terms. Pages with multiple phrases in serps or in reverse terms and which pages that rank highest for should be the starting point in understanding how to model the site structure - naturally with common sense and user in mind.

Once on-site is sweet, based on whether the content is linkworthy enough you may choose to work on link building towards key pages. Again, you almost never have time and resources to do all at once so prioritisation is the key thing here.

I hope this makes sense.

Looks spot on, and how do you extract value figures for these terms (e.g. search volume, CTRs, conversion rates and values). Do you rely on Google keyword tool only or use any other tools?

What process did you use in phrase research? (tools, steps, sorting, selection criteria...etc)

If you are in their circles they are more likely to +1 more than one thing you post in the future. That's a good investment to me.

If they have problems ranking for phrase with their own brand name, even knowing it's a destination tells me that the domain is probably not as authoritative as it could be. Knowing this I would work on content and link building. I am assuming the KW is in the domain? Your question is quite broad so it's hard to give more specific suggestions. Is destination actually related to their offer, are they product or service based?

Sorry this gets me upset. What a load of bull*#& about .edu links. The only thing that's resulted from this myth is increased rate of spam on educational institutions around the world. I wish people would stop doing this.

Yes .edu domains tend to have good trust from search engines, but that doesn't mean that if you create an empty profile and whack in your link Google will factor this into their link graph, especially if you use duplicate or no content at all. Same goes for forum/comment spam.

To get great value from .edu domains (like with anything else) you would have to have a nice editorial link connected to another page or home page, pagerank and not 1000 outgoing links to viagra and payday loans. To get these types of links you would need to earn them through contribution, engagement or become some type of sponsor. Either way it isn't simple, quick or cheap.

If you base your link building on tactics like the one you describe above you're giving Google search quality team something to penalise you for if they get to review your website at some point in the future.

There are better ways to make money from your site. Worst case outcome? Manual action by Google. Best case scenario? Algorithmic PageRank reduction where your toolbar PageRank goes down by 1/3. e.g. PR6 -> PR4 making it less appealing for link buyers.

Indirectly by linking out to random sites can put you in a bad neighborhood and reduce your websites reputation and trust by search engines.

I would like to add one point many people forget - disclosure. Google is pretty forgiving in terms of linking of own websites if the disclosure is clearly stated and it doesn't look like an SEOs attempt to manipulate link signals.

My suggestion is to link to your other site via a specific page on your first site and explain their relationship. This is a good practice in general, not just search engines.

1. Log into Google Webmaster Tools
2. Add and verify the directory as a sub site
3. Set the site migration in GWT towards new URL (Site Configuration > Change of Address)

You may also want to go to: Your site on the web > Link to your site and find the content which was linked most and best linking sites and ask people to link to a new URL. This will give you all of the value as 301 redirect does not necessarily pass all the link signals across (hinted by Matt Cutts).