Tag Archives: numbers

Game of Thones : Facebook vs Youtube for video

Game-of-Thrones

This is a cross post from my work site.

Tuesday night niche parts of the internet went a little bonkers when the season 6 trailer of Game of Thrones was released, a week earlier than any fan anticipated.

HBO released the footage simultaneously on Facebook and Youtube. Both platforms quickly racked up millions of views, but I was really interested to see the what the difference was between the two and what that could mean for video sharing….(I promise there are no Season 6 spoilers in this post!)

Actually my thinking started last month when Ok Go – an American band, in part famous for their innovative music videos – released their latest video and chose to do so on Facebook only.

If a band that is famous for its music videos chooses Facebook over a dedicated video sharing platform, what does that mean for online video sharing? Has facebook overtaken Youtube as a video distribution platform?

Game of Numbers

Lets have a look at the figures that are publicly available for the Game of Thrones trailer, which used both platforms for a comparison.

14 hours after the release of the trailer the Youtube upload had gained over 6 millions views:

Game of Thrones Youtube

But the footage shared on Facebook,  well, that had over 19 million views:

Game of Thrones Facebook

 

So on the face of it, Facebook appeared to be performing over 300% better than Youtube.

But is it?

I suppose that depends on how each platform counts its views – how long does a video play for before it’s considered a view?

I’ve done some googling and Youtube, it seems, just don’t tell you what their time limit is – they don’t want people gaming the system, especially when you-tubers can earn income from advertising on their videos.  This from Atlanta Analytics seems to be the most plain English explanation on HOW Youtube counts it’s views:

“YouTube video count WILL increment when:

You watch a video on youtube.com, as long as you don’t reload the video a bazillion times….You watch an embedded video (using YouTube’s own HTML5 or Flash player) on another domain that requires you to hit play.

YouTube will NOT increment video count when:

You watch an embedded video in a player that has autoplay enabled (video begins playing immediately on page load).You watch a video that is loaded through a proprietary player via the YouTube API.”

But Facebook’s own insights shows me that public view count is:

“…videos on your Page watched for 3 seconds or more.”

From what I can gather from my reading it counts everything on it’s site or embedded elsewhere with or without autoplay.

So if videos on Facebook auto-play while you are scrolling through your feed, and if you are pausing for just a few seconds to read friends updates above or below the post it registers as a view, How accurate an indication of view counts are these figures? Did the Game of Thrones trailer really rack up that many views?

A look at Facebook Insights

Now I don’t have access to Game of Thrones video insights, but I do have access to other pages we’ve shared videos to and I can take a closer look at the figures there.

This is a video we shared onto the Stirchley Baths facebook page

On the public side of the site it says it has had 431 views, which for a page with 975 “likes” is just under half the audience, but when we look at the overview insights they tell a different story.

Ron-Colley-Stirchley- Baths

Of the 431 views, 348 were unique and on average only 28% watched to completion.

And when we really dig down and export the data to a CSV it tells another story again.

Lifetime Total Video Views 431
Lifetime Unique Video Views 348
Lifetime Total 30-Second Views 98
Lifetime Unique 30-Second Views 85
Lifetime Total Views to 95% 56
Lifetime Unique Views to 95% 54

So according to the insights of the 431 views, only 56 watched to almost completion, that’s 12% of the total number displayed by Facebook as a “view” And when we throw in another set of stats. Facebook’s Autoplay vs Click to Play figures then it tells you something else again:

Lifetime Total Video Views 431
Lifetime Auto-Played Video Views 402
Lifetime Clicked-to-Play Video Views 29
Lifetime Total 30-Second Views 98
Lifetime Auto-Played 30-Second Views 76
Lifetime Clicked-to-Play 30-Second Views 22
Lifetime Total Views to 95% 56
Lifetime Auto-Played views to 95% 40 
Lifetime Clicked-to-Play views to 95% 16

The number of people who actively chose to click to watch the video was far far lower than those that watched it through auto play, but the retention rate of those that chose to watch to almost completion was much higher when someone had chosen to click on the link (10% on the AP compared to 52% CTP).

You can also break this down further in the insights if you want to, to people who watched with and without sound, but you don’t need to to see that Facebook’s Autoplay in news feed has a positive impact on viewer numbers on its platform, but nowhere near to the degree that the public facebook figures would have you believe.

The same video on Youtube had much lower viewing figures (30 overall from 26 unique users) but had a 74% view to completion rate. A true like for like comparison with Youtube is not possible as Youtube don’t give as detailed analytics as Facebook, but on the face of it people who watched via Youtube, watched for longer.

Maybe this is because they are on dedicated video sharing platform, or viewing an embed on a site where they’ve intentionally gone to find news on a project.

Who’s the winner?

So which is better for video sharing? I think it depends. Looking at the Game of Thrones trailer was a folly. It is a massively popular television series with an audience of millions and fan base that has eagerly awaits any tidbit of information and will watch, re watch and share any news they can get on any platform it’s on.

But for community use, for local news and for niche topics both is best. Youtube for it’s search and the ability to share , tag and target niche audiences and Facebook for the sheer numbers, the way it will appear and re appear in peoples timelines and for accessibility.

But which ever is best I think we can see that when looking at popular content we can’t take the viewing figures at face value and if you want to embed a video using Youtube, don’t have the autoplay enabled if you want the view to count.

I suppose I should also finish this with a disclaimer. I am a Game of Thrones fan and this all started with me blatantly getting my Game of Thrones fix while I impatiently wait for the the sixth season to start in April, or George RR Martin to (finally) finish next book installment of the series The Winds of Winter, but I had some useful musings from it.

Facebook Reach – How to reach your non profit/hyperlocal audience for free

Something has been really frustrating me recently about Facebook – and it’s been covered lots on other blogs – the reach that pages are getting with their posts has seemingly plummeted  since Facebook has moved towards a model of trying to get you to “boost” your content – or rather pay to get a wider audience to see it but when you’re a not for profit, or a community website paying to promote content just isn’t an option….and probably isn’t the best option anyway….

Insights and more importantly reach – specifically how to reach more people is something we get asked about a fair bit at social media surgeries so I’ve been doing some digging to try and work out what on is going on and can we improve page reach without paying?

The other day I posted a story about young girl Freya Powers, to the WV11 website and facebook page – Freya has not long been diagnosed with non-hodkins lymphoma and her parents are fund raising for  for the cancer ward she is being treated on. We basically put a call out asking for help with the fundraising.

This is what the reach looks like on the front page of Facebook:

Freya post 1

22 people liked this content, 38 people shared it and 4,168 people reached – just under 60% of our pages fans

BUT 38 people have shared this and our reach is less that our total page likes? I know from past experience that when you see shares like that the reach is normally way way over the total numbers of likes on a our page, so I looked deeper.

Here is the expanded insights  for that post:

Freya post 1 expanded

It’s still showing a reach of 4,168 put but it’s now showing 42 likes  – 22 on our page  – and a further 20 on shares…. And then I realised something  – on previous versions of Facebook insights you could view a separate figure for viral reach and this wasn’t showing, so I went searching.  I download the full  insights file….

They say this:

  • Lifetime The number of people who saw your Page post in News Feed or ticker, or on your Page’s Timeline. (Unique Users) 4186
  • Lifetime The number of impressions of your post in News Feed or ticker or on your Page’s Wall. (Total count) 12033

But then the figures from “fans” only say this:

  • Lifetime The number of people who saw your Page post because they’ve liked your Page (Unique Users) 2018
  • Lifetime The number of impressions of your Page post to people who have liked your page. (Total count) 4731

So it appears facebook have started to combine  pages organic reach with it’s viral reach in its overview and when looked at like that it’s worse than I thought.

Facebook have delivered our content to 32% of the people who liked our page –  the rest of the figures have come from people sharing the post which is a huge drop in reach to what we were seeing just 12 months ago.

Reach in decline

Edgerank Checker  have done a much wider piece of research into this and they’ve noted a year on year decline too

For the typical Page on Facebook in March 2014:

  • Organic Reach per Fan = 6.51%
  • Fan Reach per Fan = 6.46%
  • Viral Reach per Fan = 0.99%

Where were we before this?

Organic Reach per Fan (Median):

  • Feb 2012 = 16%

  • Sep 2013 = 12.60%

  • Nov 2013 = 10.15%

  • Dec 2013 = 7.83%

  • Mar 2014 = 6.51%”

Filtering

I Googled to see if I could find an answer for this steep decline and the best explanation I could find was on this Tech Crunch article – The filtered feed problem – Why Is Facebook Page Reach Decreasing? More Competition And Limited Attention

It is well worth a read but essentially it  says – while we add more people and more things to our facebook feed by making friends and liking “stuff” the amount of  time we spend looking at the feed stays the same so facebook are working to try and make that time relevant – which means filtering irrelevant content.

Which means fans wont always see the stuff we want them to see.

It also has a really interesting simplification on the algorithm facebook uses to filter posts.

BUT importantly it also states that all pages are not treated equally

Facebook is penalizing pages for link bait tactics and spammy posts.  Post memes or banal “share this if you love your mom” or “1 like = 1 respect”  text and image based content or over share repetitive content you are going to see a sharp reduction in the amount of places your content is seen.

So how do we stay on the right side of facebook’s filtering and increase your reach without paying to boost your content?

Based on the post from Wv11 I shared above we have an average post to fans reach of 32% –  which by looking at all the sites I’ve read through today puts us at the higher end of the scale for organic (non paid for) reach .

Why do I think that is? Well, I think on whole it’s by following  these set of tips

  • Be useful and relevant – is the content something your page fans would want to see – is it the type of posts they are used to seeing? Write about the things you know about and your audience will care about.
  • Accessible content  – Think mobile as well as desktop. Lots of people use their mobile to browse the internet and specifically Facebook , is your content mobile friendly  -When we post links to stories  try and summarise that content for mobile only readers – some people still wont or cant click on on external links from the page, especially those with contracts or PAYG mobiles that include free facebook browsing but not a lot of other data.
  • Photos, photos and more photos.  Anything we post with an image is guaranteed to the get more love than either text or video.  It takes the lowest amount of effort from the reader to take in the information so they engage with it, which improves your reach. You can also invite readers to tag themselves and friends in photos from events. Videos are also popular but get less views from mobile than desktop. I think this maybe down to the data usage again
  • Timing. Facebook insights are a wonderful thing. They are broken down to such an extent that we can even know when our fans are online – We get a graph that averages hourly visitor numbers over days  – but we can also see data for each day of the week which allows us to post when potentially the most people are online to reach the widest audience possible.
  • Don’t post repetitive content – if we want to re-share something – we reword it or post content with a different photograph so that facebook can see your making an effort to share something new.
  • Don’t spam.

I don’t deny it’s frustrating to have put the effort in to build and audience of over 6000 to only reach a third of them. But in the short term a third of 6000 is still 2000 more people to connect with than if Facebook didn’t exist and in the long term hopefully as Facebook filtering gets better and spammy pages are penalized by following these rules the relevant, informative pages will see their reach stabilise, or increase.