Tuesday, January 16, 2018

HK television industry news: TVB’s new “cross-platform” ratings system

At the end of the last year, I recall mentioning (though can't remember if it was on my blog or in one of the discussion forums) that TVB will be adopting a new ratings system starting in 2018 (there was a general mention at the time that other stations in HK would be adopting the same system as well by default).  Well, this went into effect already and the first reports are already out.    Before we look at the actual numbers though, let's first understand what goes into the calculation.  For TVB specifically (all the articles out so far are TVB-centric – I have yet to see anything about the other TV stations so not sure how specifically they will integrate this new system), the ratings number will now be a calculation of the following 3 things:  1) traditional TV ratings (those who watch via traditional TV set same time as broadcast– these are the same ratings we have now),  2) myTV Super "live" ratings (those who watch via myTV Super platform same time as broadcast), 3) myTV Super 7 day re-watch ratings (those who don't watch same day as broadcast but go back at a later time to watch).

In order for the above method to make sense, here's a little bit of background and explanation.  We all know that nowadays, in HK especially, very few audiences watch TV on a traditional television set (by "traditional", I'm talking about those "vanilla" TV sets that are only capable of receiving free channels and maybe cable subscription – doesn't include the ones with OTT/set-top boxes) .  With all the advances in technology (and the fact that Hong Kongers are known for leading busy lives that are always "on-the-go"), the trend in recent years has been for audiences (those who still bother with television content that is) to "watch TV" via the Internet (online on a computer or hooked up to a television set via OTT set-top box) or via mobile devices such as cell phone or tablets.  In addition to watching via multiple platforms, the time of day that audiences watch has changed as well – gone are the days when a typical HK family would sit around the dinner table with the TV on and watch programs together while they ate (those who grew up with TVB in the old days like I did would probably be familiar with the phrase "rice mixed with television", lol).  Instead of watching a TV program at the same time that it is broadcast, many audiences nowadays do "catch-up" watching – basically watching all of that week's episodes at one time, either at the end of the week or on the weekend.  So then supposedly, this means that the ratings going forward should, in theory, be more "accurate" in the sense that it takes into account modern audiences' viewing habits.  

According to a Mingpao article I read today, TVB released their first ratings report under the new system and this is how a few of their programs fared (this covers the week of 1/1/18 to 1/7/18): 

Golden timeslot – 1st, 2nd, and 3rd line series:
Come Home Love: Lo and Behold = average 23.5 points, 1.53 million audience share (households)
My Ages Apart = average 26.1 points, 1.7 million audience share (households)
Heart And Greed (Heart of Greed 3) = average 25 points, 1.62 audience share (households)

Midnight timeslot (this is the timeslot at midnightwhere TVB usually airs old series / classics)
Files of Justice V = average 7.4 points, 480,000 audience share (households)

Weekend timeslot 
Cantopop at 50 (Anita Mui Tribute special) = average 18 points, 1.17 million audience share (households)

Source: Mingpao 


Obviously, the above is not an all-inclusive list, plus there is no comparison point in terms of what the ratings would've been for that same week if the old method was used, so personally, I am hesitant to say whether this new system is truly "more accurate".  While it does reflect better the viewing habits of audiences in the current environment, I have some reservations as to accuracy due to a few reasons. 

-          First, it appears that the calculation method for the first number (traditional TV ratings) is the same as it was previously – this number has always been a bit convoluted, as it is based on a series of algorithms and formulas performed on data that is a bit suspect to begin with, since it looks at "average" households and takes a "sampling" based on population.  I'm not a math whiz and hated Statistics in school so maybe I'm just not understanding the calculation properly but I've always been suspicious of this ratings number and what it represents, which is why I rarely ever take ratings into consideration when it comes to TV series (I basically could care less what the ratings are, since it won't impact whether I like the series or not).

-          Second, I'm not totally clear how the second number (online platform "live" ratings) is calculated.  With traditional TV ratings, despite not understanding the formulas and algorithms specifically, I at least DO have a general idea overall how the ratings are calculated (which is basically installing some type of device into the TV sets that will monitor when it is tuned to a certain program, measuring down to the minute, then sending that data to the data processing centers to get put into their various formulas).  With online ratings, how does this work?  I'm assuming it's based on number of unique clicks or something of that nature but honestly, I'm not too sure.  I was able to Google some information in general about the different methods that Internet ratings are usually captured and calculated, but without knowing which method TVB (or whoever is putting together the reports over there) is using, I can't say with any amount of confidence that this number is capturing what it's supposed to capture.

-          The third number is the most "mysterious" one in my book because honestly, how do they even capture who is going back to watch the series and who isn't?  And does it depend on how many episodes are being watched?  So like what if I missed 2 weeks worth of episodes and I binge-watch the first 7 episodes on Saturday and the last 3 episodes on Sunday?  How does that affect the number?  And what if I watch outside of the 7 day window – does that get captured in the following week's data or it doesn't get counted at all?   Obviously there are a lot of "unknowns" still with this one so until I have time to do more research and truly look at what this number is capturing, I can't really say a whole lot about it.

So in conclusion – for me personally, nothing much really changes because as I said earlier, I'm not a ratings-watcher so how well a series does or doesn't do won't impact whether I choose to watch or not….it's the content and the quality of the production that matters in my book.  With that said though, it will be interesting to see how this new ratings system evolves in the coming months as more and more people catch on and start digging into things more.  I'm sure TVB will be asked to explain their calculations at some point and if they do, then I'll be sure to post an update at that time.


  1. Thank you for sharing your thoughts. I truly appreciate your efforts and I will be waiting for your next
    write ups thank you once again.

    1. @Anonymous: You're very welcome and thank you for the kind words! Sorry that the writing in the post was a bit choppy though, as I was in a bit of a rush trying to whip up something quick between projects. Hopefully the next one will be less rushed! :-)

  2. Just sounds like TVB wanting to manipulate the ratings to keep them from falling below 20 or 15 becuase at this rate it will be the norm by next year. It’s more like them trying to trick people into thinking their series are good.

    Ratings or not, if a series is good people will know and it’d be the talk of the town like OCTB and The Menu. Instead of worrying about ratings they should worry and overhaul their entire filming system and production crew while forcing everyone onto new training like Ricky Wong amd HKTV.

    1. @sport: I completely agree with you. Now that I had a little time to look at it, adding in the internet stuff definitely makes the ratings higher. I remember Heart and Greed was in the low 20s when looking at just traditional TV ratings but now it's in the mid-20s...all the other series too are higher than what they were previously, so it's obvious doing the cross-platform thing helps their numbers look better. But I agree that at the end of the day, it really doesn't matter because more and more, people in the industry have expressed that they care more about the word of mouth that a series generates and being in the audiences' favor than how high the ratings are.