Welcome to the Q&A Forum

It's hard to offer really concrete advice without a bit more information on the specific context in question, but here are some general pointers:

1. You probably shouldn't duplicate the content across YouTube channels. You can't Geotarget YouTube channels and in practice, this is not really any different from having duplicate content across multiple websites. If you do duplicate the content, you'll cannibalise the effectiveness of the videos ranking on both YouTube and Google organic searches, plus you'll split the view count which will harm your YouTube optimisation.
2. Absolutely 100% do not put the sales focused product videos on YouTube - as in doing so you'll drive traffic to YouTube instead of your own site, which means you're adding an unnecessary step in the conversion funnel which will probably cost you money. You cannot sell from a YouTube channel and external links on YouTube get extremely low click through rates, with very very few people going to YouTube to search for products to buy (you only go to YouTube to watch a video). There is a bit more information about this for you in these links: http://moz.com/blog/the-marketing-value-of-youtube , http://moz.com/blog/building-a-video-seo-strategy, http://moz.com/videos/view/2013/building-a-winning-video-marketing-strategy
3. Annotated external links do not get clicked very often, I wouldn't rely on this as a strategy at all. It basically won't work.
4. You can't geotarget annotations, no.

So, in short - I probably recommend just having a single channel and only putting the entertaining videos on YouTube. bear in mind that YouTube doesn't drive much traffic, so your goal here should be brand awareness, not traffic acquisition.

I hope that's useful!

Thanks,

Phil

That sounds like a nice idea and will work nicely for users - you'll just need to consider whether you would rather the YouTube videos or the pages on your site rank for the targeted keywords and optimise everything accordingly. There aren't any duplicate content risks there.

Thanks Robert, Glad you found it useful!

If you want a bit more detail on some video stuff, I have written a couple of longer posts around the topic:

http://www.distilled.net/blog/social-media/youtube/creating-awesome-videos-for-seo/

http://www.seomoz.org/blog/getting-stuff-done-by-video

and Dave Sottimano put together a good guide for YouTube optimisation earlier in the year:

http://www.distilled.net/blog/social-media/youtube/youtube-seo/

Hopefully they might help your endeavours too.

Hey Andrew,

So, as far as i am aware you will need unique thumbs and video locations. You might be able to duplicate thumb information, but any duplicate locations will flag as a GMWT error.

There's nothing technically stopping you uploading the same video to your hosting package time and time again and then just placing the duplicate videos on different URLs and submitting a sitemap with unique   video:content_locelements despite having duplicate content.</video:content_loc>

However, this isn't a brilliant thing to be doing and sooner or later, Google will work out that you're using duplicated videos and you could then get devalued for it, which will prevent you from getting nice blended video results

There are a few things which will indicate to G that you're duplicating:

• Audio Wave forms.

Google can crawl elements of audio files.- It's how they ascertain a lot of copyright infringements on YouTube - by matching wave forms.

• Video Lengths

If you have tons of videos, all of which are the identical length and have identical thumbnails - it's going to be obvious.

• YouTube Hosting

If you're using YouTube to host your videos - it will abundantly clear to YouTube that you've duplicated the content, by the inevitably poor hotspots metrics you'll receive and the lack of views. For product videos, as a rule, i would recommend avoiding YouTube for hosting anyway - Wistia is a great option for that sort of stuff.

Duplicate videos are ultimately the same as duplicate body text and are more or less frowned upon by the engines if exercised at too great a scale - especially following Panda. Ideally, you want to be thinking big and aiming for the unique videos - not only from a Search perspective, but also from a CRO and user perspective. No user is going to enjoy browsing the site if they end up watching the same content multiple times.

I contest the idea that you can't scale unique video content easily. You just need to work out which elements can be duplicated and which need to be unique. Always get unique audio - and wrap some stock elements around a small portion of unique footage. I produced the following YouTube series in 1 day. Filmed, cut, edited and uploaded 64 unique videos.

http://www.youtube.com/playlist?list=PLDCBEEAFF5571810B 

They're not complicated, flashy or impressive - but they are all unique, full HD and with appropriate metadata, which is the minimum you need to aim for.

If you have thousands of products for an ecommerce site - sure, it might take a few weeks to produce everything. But you'll be much better off for it.

anyway, hope that's helpful!

Cheers,

Phil

Hey Ross,

I will try and offer some advice point by point:

1. If it's a good blog and the content gets regularly indexed, then it can absolutely be worth doing another guest post to get more links. In these instances, often it's a good idea to link to other pages on your site and to vary the anchor text across the board. While you will always want to ensure part of your link building efforts go towards getting a variety of links from different domains, having 4-5 nice links across the domain will invariably be better than getting just one. Definitely work to maintain any guest blogging relationships you have - those are hard to come by! I'm afraid it doesn't work link a "one site, one vote" system.

2. It depends on the blog, the content and the context - but it ultimately wont make a huge difference either way. Rather than focussing on the position of the link in terms of placement within the code - think about choosing the placement of the link from an audience perspective. Where would the user want to see a link to my website? If your guest post happens to focus around the specific service your company offers, by all means put in some relevant links within the flow of the page copy - ensuring the words around the link are somewhat keyword focussed. If your guest post content is less related to your business, then just put a couple of links in the author box profile, within a sentence explaining what your site does. This sentence will provide some relevancy. I normally aim to get one branded link to the home page  and one anchor text link to a deep page for any guest post content.

3. No, not necessarily. Back a few years ago, this was a common assumption, but the value of a link is not worked out on a higher/lower scale. Footer links and sidebar links are generally thought to be less valuable than good contextual links, but the exact positioning of the links makes little difference. The exception to this, however, is when a page is massive and contains dozens of links within a great deal of text. In these instances, its generally preferable to be first rather than last on the list - but those sorts of links are typically not the kind you have control over anyway.

Hope that's helpful!

Thanks,

Phil.

+1 RB.

RB is a boss.

To me, this is a bit like finding a successful infographic and then asking "Isn't image content killer link bait?".

Video, like image, is a medium for content delivery - rather than a type of content and this is how it should be considered in the context of link bait. Video content can be excellent link bait, but the form itself is not by default any more "linkable"  than images, words or applications.

"Linkbait video" it requires a specific approach which often doesn't tie in very well with commercial interests. For someone to link to a video, there has to be specific value in the experience of watching it through -- this can be an aesthetic experience, humour, instruction or exciting delivery of information. for more information and advice on what makes a "linkable" video - check out this blog post on the topic - http://www.distilled.net/blog/social-media/youtube/creating-awesome-videos-for-seo/

There are also some specific problems with using video for Link Bait

1. If you don't secure your video properly, the content can be ripped and uploaded to YouTube. As you rightly pointed out, Youtube is the worlds second biggest search engine, which means that individuals will use it to find your video content, rather than your own site. This means that often a lot of your earned links will go to an erroneous YouTube channel, rather than the targeted domain.
2. Video is invasive when embedded on a website. Unlike reading or looking at an image, watching a video is an active and engaged process. You cannot skim read a video and thus, if someone embeds a video on their site - it has to be because they want their readers to temporarily disengage from their own content and spend the time watching the thing through.
3. People don't link to mediocre videos. The internet is awash with average video content, the vast majority of which is on YouTube and never gets watched or linked to be anyone. While you can always pick-up a few links with an average blog post, PDF document, infographic or plugin, run of the mill video will get you nowhere.

So, yes, Video can be great link bait - but there's a lot of risks involved and you have to do an exceptional job really get the links coming in. I would encourage you to have a play and put something out there, but don't expect to get it right first time. Videos can be valuable even if they don't get links as they offer the opportunity for blended organic results, video results, exposure on YT and can help improve the overall richness and quality of your site.

Hey,

As you correctly indentify - since a standard YouTube embed is in an iFrame, the content loads separately and wont effect site speed. If you choose to use an older style flash based YouTube embed, then this will require a bit of loading, but it wont be significant or anything to be concerned with. However, if you were to have a single page with tons of flash embeds, then this could be slightly problematic - so avoid that and use iFrames where possible.

Phil

Hi David,

I have been on the hunt for something like you suggest for a while, but i'm sad to say it doesn't seem to exist outside of the current YouTube Keyword tool.

Even though the YouTube Keyword tools doesn't give you the search volume data in the CSV export, you can still simply copy-paste the information off the page to get the relatively scant data Google provide. Additionally, using the general Google Adwords keyword data is probably as good an estimate of search volume as any. The numbers here and in the YT keyword tool are invariably inaccurate and should be taken with an extremely large pinch of salt.

You can't get a YouTube specific list of keywords through the general Adword keyword tool as far as i am aware, though i'd be happy to be corrected on this if anyone knows different.

Sorry I can't provide you with the solution!

Cheers,

Phil

Hey,

So - I actually don't think either of those two strategies are worth considering in order to get more nofollow links. Particularly with YouTube - if you connect your site (by linking) to a video that will inevitably perform poorly on YouTube/vimeo etc - you're just sending negative branding signals about your site back to Google and it wont help you in anyway.

So - my first questions should be - why do you need more nofollow links? A very small percentages (think it's 3% or something) of the webs links are actually nofollow,- and these are typically accrued naturally through profiles, comments, wikipedia articles etc. The only reason i can think of why you might want to be increasing them at a small scale like this would be to cover up a thin and questionable backlink profile. If this is the case, then i don't think this strategy will help to mask the unnatural links/suggest improved quality - especially if you've already been slapped by Panda. Multiple profiles will look spammy and i can't see them helping you in any tangible way.

Probably a better use for your Product videos will be on the site - hosted on a secure third party platform (not YouTube), to show Google that you're expanding your media types and creating richer content. Coming out of a Panda penalty - this seems like a much better thing to be doing and will send the right signals. Additionally - i would use the great images on your site, rather than using other services to post your content for the sake of nofollow links.

Like you suggest, you can increase nofollows from quality comments and profiles generally - but i come back to the question of the underlying purpose here.

So - My suggestion is: Put all the content on your own site only- and securely host the videos with a third party solution (vimeo pro, wistia etc) rather than using YouTube. You don't need to give away content to get nofollow links - but if you're trying to make a bad backlink profile look good - that isn't the way to do it.

Hope that's useful!

Phil

Hi Richard,

The long and short of it is that yes, this will eventually get picked up as duplicate content and devalued, but moreover, if you upload the same video multiple times, you will be diluting the amount of potential views each version will get and therefore hampering the rankings on YT, Google video and Google organic.

Therefore, I would strongly recommend against uploading duplicate versions with differing meta-elements. What is the ultimate goal of these videos? Are you trying to get blended results to YT through submitting video- site-maps? Or are you trying to rank on YouTube for a variety of long-tail terms?

Either way, duplicate content won't be the way to go. I would either recommend optimising a single video for a host of long tail terms by building lots of strong anchor text links to it and hosting the video on a page on your own site which has a list of relevant phone numbers/addresses in the source but only delivers a single one to a human visitor based on IP?

Or, if you want to make lots of videos - they will all need to be broadly different so there is value in watching the whole series for any given user. The best way to do that is to take a simple concept and then just scale it by having duplicated branded wrapping and differing the internal content, such as with the following example, which I built last week: http://www.youtube.com/playlist?list=PLDCBEEAFF5571810B&feature=view_all

Those were made entirely within one day, without any expensive software, so it is possible to scale unique videos. If you do something like this, then you could add different phone numbers and addresses in the meta-data and tagging. Your channel actually has lots of good unique content, so I think you'll be best off sticking down this route and just scaling your production further.

Thanks,

Phil

So - i don't think this site has a video sitemap - though that could be helping. A lot of videos are implemented in flash and Google have recently begun crawling more and more flash files and indexing them as videos. Google can't see past the flash code - which includes <param name="movie" indicating="" that="" the="" file="" is="" essentially="" a="" video.<="" span="">

Whether the content is a slideshow rather than a video is more a semantic than a technical issue - as it's technically identical to a normal flash video. The video rich snippet it receives are essentially relevant.

If you have lots of flash image sequences implemented across sites - then i would encourage you to treat them as videos and use the movie schema markup- as well as creating and submitting a video sitemap.

You won't always get the video rich snippet results for the content - but it can't hurt to provide the engines with more metadata and information.

I recommend having a read of the following posts i've written on this topic. The answer is always somewhat nuanced, but hopefully this will give you a good overview of the perspective you should be taking when trying to answer your question.

http://www.seomoz.org/blog/hosting-and-embedding-for-video-seo

http://www.seomoz.org/blog/building-a-video-seo-strategy

http://www.distilled.net/training/video-marketing-guide/

I generally concur with the comments so far - it's definitely a good idea to have the transcript within the HTML, marked up with Schema.org and available for users to see.

Set your <div>  to style="display:none" and then include a button which will trigger the text display in a drop down, tab or Lightbox overlay.

Alternatively, the Jquery route that Irving suggested is just as good.

Whatever the solution you go with, you just need to stick to the principle of allowing the text to be visible and easily accessible to users if it's also going to be visible to the search enginges.

However, if you're already including transcripts flat on the page, then aside from aesthetics, there's no reason to bother hiding the text .

The issue is, this isn't what the map location feature was designed for. If you take this approach, you're ultimately attempting to trick YT into thinking that your content is unique when it isn't and this will not be looked at kindly by Google. While your approach may work for now, the YouTube videos wont rank very well and you are putting yourself highly at risk of a spamming penalty in the future.

I think the single video option with dynamic page could definitely work, but it might be tricky to get the one page to rank 100+ geographical terms

An alternative might be not using YT to host, as Google can see the analytics info for YT and give negative quality metrics based on the fall out rate of the YouTube vids; which will inevitably be extremely high if it's an advert.

I would recommend using Wistia - http://wistia.com/ which Moz use, as they have an automatic sitemap submission service which is very nicely put together and easy to use if you're scaling the content over a vast franchise. You could upload multiple versions of the same video with Wistia and create unique pages for each franchise location and then you should be able to get blended results for multiple geographical terms - especially if you hook the pages with the videos on to the Google Local listings.

All that said, Google will be able to crack down on duplicate video in the future. Audio wave matching isn't that hard - they already do it for YouTube and will do it for other services in the future. If you can get away from totally duplicated videos, especially on the audio front, it would future proof you from Panda and other dupe content smack-downs. Maybe get a different actor (with appropriate regional accents) to do the VO for each franchise?

Hey Shailendra,

Many hosting solutions allow you to do this and you can also put links in at the end of an HTML5 video with Javascript.

For the latter option - I recommend reading through this page http://bubbles.childnodes.com/ which gives you some usable code.

If you're looking to host your videos with a third party provider - then I recommend either Wistia or Vimeo Pro which both allow you to very simply add clickable links as a "call to action" at the end of a video.

+ Add Image or Video URLs

Hope that's useful.

Phil

This http://www.seomoz.org/blog/hosting-and-embedding-for-video-seo  should answer many of your questions fairly succinctly.

Hi Andrea,

Unfortunately this is actually a bit of a front end development question that it's very difficult to answer without first having a look into your site and a specific example of what's going on.

Any chance you could share a URL with us? I would imagine that the problem is either with the styling of the table, or some default settings in your CSS file related to the video embed.

To fix it, you may just have to manually adjust the size of the video frame to match the table boxes, which you can do in the Wistia superembed interface.

Thanks,

Phil

Sure,

Here is a list of some content Moz and Distilled have done on Video.

http://www.distilled.net/blog/social-media/youtube/creating-awesome-videos-for-seo/

http://www.distilled.net/blog/social-media/youtube/youtube-seo/

http://www.distilled.net/blog/seo/getting-links-and-seo-value-from-your-youtube-videos/

http://www.seomoz.org/q/best-way-to-host-video

http://www.seomoz.org/blog/getting-stuff-done-by-video

http://www.seomoz.org/blog/video-sitemap-guide-for-vimeo-and-youtube

http://support.google.com/webmasters/bin/answer.py?hl=en&answer=80472

Without having explored the page and the sitemap, I would say that what your developer says is entirely possible and represents a good way forward from your situation.

Google have made a shift in the past few months with the way they index sitemaps - and if everything more or less matches up, even if they can't see the video on the page, they will often ascribe the page with a rich snippet and index the page as if there is a video present.

It's not surefire and having the video hidden beneath JS and delivered dynamically can prevent content being indexed - but it's certainly worth giving it a go (especially if you're using Wistia, as it's a simple process) and seeing if the video content gets indexed anyway. If it doesn't work, then I'd recommend going back and seeing if you can find a way to make the embed visible outside of the JS (perhaps just hide it with CSS and make it visible in a separate tab?). You could then have two methods of triggering the video playback for users if you wish.

Cheers,

Phil.