Welcome to the Q&A Forum
Browse the forum for helpful insights and fresh discussions about all things SEO.
Hi,
Cyrus Shepard did a really good post on this not too long ago. There's a lot more to it than simply redirecting the old URLs. Check out the full list here: https://moz.com/blog/seo-tips-https-ssl
Post-launch, you'll want to closely monitor the index. Check back daily (it only takes a few minutes) to ensure that your HTTP URL count is decreasing and your HTTPS count is increasing.
It's not a matter of keeping them happy, it's simply helping our client accomplish what is best for their SEO during this migration. We have a really good relationship with them, so it's in everyone's best interest if we help find a viable solution.
Unfortunately, launching all at once isn't an option. This migration has been planned for months, there's no going back to the drawing board at this point. There won't be any brand confusion because visitors don't have much reason to be bouncing around to the different sections, they're very localized.
Thank you for your input, but it looks like I'm still on the hunt for a solution.
Thanks for your response Ruth.
Luckily, the URLs won't be changing....at least that's the plan anyway! 
I'm looking more for info on how to have one domain with different sections spread across 2 different servers. The new development company has a proprietary CMS and are redesigning portions of the site incrementally. They will probably be reluctant to work off of the old development company's server, which is why we're trying to provide a solution that makes both dev companies happy.
Thanks for your response!
The new dev company proposed the subdomain idea, which I'm adamantly against. So this is why I'm on the search for an alternative- surely, someone else has had a similar situation.
There is a proprietary CMS involved, I should have mentioned that initially, which is where the real challenge comes in. I don't think either of the developers would be thrilled about a recommendation where they put there info on someone else's box, unless there is some way to lock down access that I'm not familiar with.
Hi Mozzers!
I've got a client that is in the early stages of moving the development of their site to another company and therefore, a new server. The site is very large and the migration will take place over 18 months. In the beginning, smaller chunks of the site will be moved, and as that process gets dialed in, larger portions will migrate.
It was brought to our attention today that they (on either side of development) have not yet worked out the logistics of keeping the domain and URL structure consistent throughout the migration. The initial proposal was that they publish newly migrated pages to a subdomain, which we obviously want to steer away from. I'm now on a mission to find a solution that will make everyone happy; client, old dev, new dev, and us (as the SEO partner).
Does anyone have experience in managing SEO through a migration such as this?
You're off to a great start. Also be sure to check out Moz's new Keyword Explorer: https://moz.com/explorer. This will be really helpful in expanding your keyword list to be inclusive of more variations and other related topics to cover.
As Yossi said, configuring parameters in Search Console should help - _but, _that's only going to help you out in Google.
Adding a disallow for those parameters in the robots file will help solve the problem in other search engines.
The thin content is definitely contributing as well. Moz identifies dupes based on a source code match between any two pages of 90% or higher. When you consider all your template code is the same across every page, thin content isn't enough to differentiate the source code.
I also noticed on one of those screenshots that you got a one dupe of /shop/necklaces/ and /shop/necklaces/necklace/. If you can, I recommend removing that second one with doubled up 'necklace' folders, that's going to cause a lot of dupes as well.
Hi Jacob,
I took a look at your site, and the canonicals appear to be configured correctly. When you look at your duplicates in the Site Crawl report in Moz, and you click the + next to where it says "1 duplicate", what are you seeing? Is it a URL set like the example you've used above, or something else?
This type of duplicate applies to far more than just the homepage...so by implementing a single redirect rule, you handle this type of duplicate for every single URL on your site, and neither people nor search engines will ever see the undesired version.
If you do this via canonical tags, you have to include the canonical tag on every page, which could be very time consuming depending on how large your site is.
Hi,
Ideally, you would handle www and nonwww duplication by way of a redirect rule. Both versions of that URL should not render, i.e. when you go to www.example.com it should take you straight over to example.com.
You can handle this with canonical tags the way you've proposed, the URL in the href element should be the one you prefer. However, this should be reserved only as a work-around in the event that you can't get implement a server-side redirect rule to handle ALL www-to-nonwww issues.
That's the only method I know of deter unqualified clicks, interested to see if anyone else chimes in with some useful nuggets of info on this topic!
Hi,
Have you tried qualifying clicks through meta description and title tag copy? I've had pretty good success using this tactic for local businesses to deter clicks outside of their service area - because a similar problems exists where you can't add negative geographies for where organic results will show.
Hi,
Google Analytics is pretty good about assigning proper attribution to channels. If you had been sending out emails prior to tagging the links, that traffic was falling under either Direct or (Other).
PPC is the only medium that would cannibalize organic since you can have impressions in both spaces at the same time. The drop in organic traffic could just be the ebb and flow of your business. Since most businesses have seasonality, year over year comparison is often a better yard stick for performance.
Is there a particular search engine that you can attribute the traffic loss to? Or perhaps a piece of content received more entrances last week simply due to the timeliness of that content?
If you're able to predict your links that easily, they're probably not very good links and therefore won't contribute much to the DA of your site.
Since your blog is on a subdomain, yes, you will need to set up a separate WMT profile for it. The blog will also need its own robots.txt and XML sitemap files, since technically speaking, subdomains are regarded as a different site.
Hi Joe,
Specifically addressing your first question, 301 redirects that take URLs to HTTPS won't cause any loss in link juice as they would with a typical 301 redirect. This information came out of a Q&A with Mueller, which you can read more of here.
For your second question, that largely depends on your server, but you should be able to do this with a single rule, as opposed to redirects on a one-to-one basis.
HI,
You might want to try adding the NOODP tag across all pages on your site.
This tag will allow you to 'opt-out' of Google filling in this data for you. The ODP in that stands for Open Directory Project, aka Dmoz, which is where Google grabs this information from if it feels your titles and descriptions aren't the best.
I think the solution you're looking for is pagination markup. Use the rel=prev and rel=next tag to point search engines towards the proper paginated URLs. This effectively eliminates duplicate content issues where http://3mil.co.uk/insights-web-design-blog/page/3/ is flagged as a duplicate of http://3mil.co.uk/insights-web-design-blog/page/2/.
You can read more on this topic in this article on the Search Console help site.
I'd recommend keeping it on the site. This type of content has a good chance at garnering some quality links, so you don't want to dilute their value by redirecting.
Since your content is very time-relevant, it would be very beneficial to your users to include this information. It could also help with organic long-tail queries, in the case where someone searchers 'fashion trends for summer 2016'. You're currently not providing the date information, so your chances of appearing for that query are much lower than if you had date-published info directly on the page.
You're welcome, have an awesome day!
Those blogs posts are going to be your long-tail keywords, so the more specific you get with the keywords that you target, the better. Put yourself in the mindset of someone who would benefit from that content, then try to think about what you would search for to get to that content. For example, for your post "Steps to follow in buying a CRM", some qualified queries might be "things to consider when choosing a CRM", "how to choose a CRM", "what do I need to know before buying a CRM".
Hi Luke,
I've had many clients in the same situation. The approach I typically take is to split up the street address between a Suite A and Suite B (or some similar variation of that). Google is still able to pinpoint the geographic location of the business, but also recognizes that they're not identical.
Hi,
Your statement "I know that targeting more than one keyword per page in a website leads to keyword cannibalization" is close, but not quite true. Keyword cannibalization happen you talk about the same topic or keyword set across multiple pages, search engines aren't sure which to rank higher, so they usually opt for none of them.
Content should be centered around topics these days, not keywords. Based on the sample blog topics you shared, it seems like you're doing that quite well. You're still talking about CRMs but in different ways, so I think your strategy is good.
You're in the driver's seat here. _You _have the content _they _want. If you lay down your requirements and they don't want to play, then don't give them permission to use your content. It's really that simple. You're gaining nothing here with their rules, and they gain a lot. You should both be winning in this situation.
Adding nofollow to links that point to dictionary pages will prevent search engines from getting there, but since the pages are in the index (and you don't want to change that) you're still facing the duplicate content issue.
I know it's a huge project to take on to add content to these pages, but it seems as though it's your only option. Perhaps you could split the project up between a few people and each update one page per day. That way it doesn't turn into a major time-suck.
No problem! I meant to mention this in my first comment, but I also noticed that there's no robots.txt file in place. That's obviously not going to help your indexation problem too much, but nonetheless something you should know about.
You can do that, but you should check Google Analytics to see how many organic entrances you get to these dictionary pages first. If a lot of people enter your site that way, rel=canonical is going to hurt your traffic numbers significantly. For example, when you add a canonical tag to this page (http://anglu24.lt/zodynas/a-suitcase-lagaminas) that points elsewhere, the suitcase page is going to get dropped from the index.
The ideal situation would be to create more unique content on these pages. You're getting duplicate errors because more than 90% of the source code on the dictionary pages is a match. When you consider the header and footer, and the other code for the template, it's the same everywhere. The dictionary pages are very thin on content, so it's not enough to differentiate. If you can, build out the content more.
Here's a few ways you might add more content to each dictionary page:
Looks to me like Google can't properly access your XML sitemap. I tried to put it into 2 different validator tools and URI Valet and none of those tools were able to access it. It could be something with HTTPS. Did you recently switch the site over to secure?
Even if you're not linking back and forth, you're still diluting your SEO value over many domains, as opposed to providing one place to demonstrate expertise in your niche. However many domains your spreading your content over, is how many times you'll be duplicating some of your work. Say you have 5 domains, one for each appliance type, then you're reporting and conducting analysis for 5 sites, you're updating 5 XML sitemaps & 5 robots files, 5 different sites to get/monitor reviews for, 5 sites to monitor rank for....you get the point. Additionally, if you get a really good link for _one _of those sites, it only benefits that domain, whereas if you're operating all under one site, that link helps all service lines, not just the one whose domain got that link.
If they can determine that they're connected (which is highly likely since you'd be linking back and forth), all of them.
That's getting a bit into the black-hat realm. I would stay away from any strategy where you have microsites for each service offered. Linking back and forth between microsite domains like you mentioned is going to look very sketchy to search engines. These days, SEO is much more about the quality of your content and how much of an expert you are in your niche, and less about the keywords you can stuff in your site.
Hi,
I'd stick with the domain name you're currently using. There is no SEO value to what domain you have. This used to be true ~10 years ago, which is why you see a lot of domains out there like AirConditioningRepairHouston.com, etc.. You may not have much domain authority right now with the current domain, but if you switch, you'll have zero. Additionally, any links you already have will lose about 10% of their value when you redirect them to a new site.
Hi,
That sounds like a definite candidate for duplicate content issues. A true responsive design only has one set of page elements coded, which then rearrange based on screen size, that's what makes responsive the optimal solution for SEO. Search engines only have to read one code set per page and they know it'll render for most devices. In your case, I believe search engines will view that as a tactic to game the system; one version of the content is essentially cloaked when the other is displayed.
You can probably find a good WP freelancer that can do this for you. I don't know of a simpler way to implement such an attribution solution.
You can leverage that phone number by populating a field on the form based on what phone number displays.
For example, you may have the following dynamic phone configuration:
555-1234: Google Organic
555-3456: Google Paid Search
555-5678: Email Marketing
When someone submits a form, there is a hidden field that looks at what phone number is displayed and then populates the field with the coordinating marketing channel. As far as execution of this logic, I'm afraid I can't help, our Dev team handled all that. But it's pretty simple logic, so I think most developers would be able to write this relatively easily.
Are you, by chance, using any type of dynamic phone tracking on the site? We implemented an attribution model like this by leveraging the displayed phone number.
I just ran this query for bvstate URLs indexed for the H&R Block site. Mozbar shows canonical tags with bvstate in them, and Screaming Frog finds no canonical tags at all. There is a deeper issue that is not simply Google ignoring them.
Do you have different examples? The Home Depot link doesn't work when trying to view the actual page on the site. With the Microsoft link, the canonical is working, as the version with the parameter is not indexed in Google, but the canonical version is indexed, which is what I would expect for a canonical that is being obeyed.
No problem, glad to be of assistance!
HI Donald,
On the site that borrowed the post from the original publisher, add a canonical tag that points to the URL of the original blog post. This is the ideological way to handle this, and how news sites should be crediting the original publisher when syndicating articles.
In some cases, implementing a self-referring 301 redirect may cause an infinite loop in which your homepage would not be accessible at all, so I can understand your dev team's reluctance.
A canonical tag and a 301 redirect pass the same amount of link authority, so in this case, they serve the same purpose and provide the same benefit. I'd stick with the canonical tag and pick a different, more valuable battle to fight.
Hi Rachael,
Spending time pursuing links these days is a pretty futile task. You'll spend hours prospecting and doing outreach, and then most site owners won't respond to you.
The best thing you can do to generate links is create high-quality, unique, trustworthy, and linkable content. Get your content in front of the right people, and it will acquire links naturally. The best type of link to have is one that no one asked for, the content was just good enough for someone to say 'hey, this is useful stuff, I'm going to reference it on my site'. Of course now, the tricky part is getting your content in front of the right people. You can leverage most of the usual channels for content amplification; social, organic, email, etc. etc.
Hope that's helpful!
Hi ,
Did you follow all of Google's Change of Address instructions? If redirects are in place, chances are you missed something else minor that can have a big impact.
Hi Thomas,
That should work. You can confirm this by modifying your robots.txt file in Search Console and testing a handful of URLs to ensure they're blocked the way you want.
Hi Sigurd,
Your implementation of canonicals is almost there. You're currently using the relative URL, when you should be using absolute URLs. As per Google Search Console guidelines, relative URLs here can cause issues.