Working with the numbers I have


After writing my last post bemoaning the lack of good data to judge the impact of various marketing activities, I decided that I should start tracking the data that I do have access to, namely YouTube hits.

The upside of YouTube is that it’s an international measure–while different retailers dominate in different countries, pretty much everyone goes to YouTube. In addition, I’ve been checking it at around 9 pm Pacific Standard Time, which is about 1 pm the next day in Seoul. Usually television or radio appearances happen in the evening, so that’s actually a good time to check the counters–the previous evening’s appearance has presumably had its effect, and the group hasn’t made its new appearance yet. (The big exception to this was October 22, when I checked it at 3 pm and 10 pm PST for reasons I will explain below.) At first I was annoyed at myself for not thinking of doing this sooner, since the album has been out for almost three weeks now. But then I realized that if the existing fanbase is basically satiated by this point, then an uptick in views would presumably be coming from new people–and it’s good to know what reaches them.

This is by no means a perfect measure–“Very Good” is on more than one YouTube channel, although the Seven Seasons channel I followed does get the most hits, and it appears first when you search the site. Sometimes fans have campaigns to drive up the number of views (which helps with winning the countdown TV shows), and if the Korean fans organized something like that, I probably wouldn’t know about it. Presumably those campaigns would precede a television appearance, however, and I noticed that the big upticks tend to happen after I went to bed (i.e. in the later afternoon/evening in Korea), which suggest that they are the result of the TV appearance itself, not of a campaign that preceded it. But YouTube’s counters also stall, sometimes for hours, so I can’t be sure about the timing, or even that a flat day-to-day reading is actually real (since it could be caused by a really long stall).

After a couple of days of this I realized that I could also track Very Good‘s listing on the Amazon K-Pop bestseller list. The issue is that the album has been bouncing around wildly (It’s #21! It’s #6! It’s #38!) with no apparent pattern. That suggests to me that Amazon (which serves a primarily American audience) doesn’t sell that many K-Pop albums, and that small shifts in the number of people who buy Very Good (or any of the other K-Pop albums they sell) have an outsize impact on the ranking. So I gave that up for the most part (but I will note that Blockbuster has been selling almost as well most days on Amazon as the newer album).

Although I didn’t start tracking views until this week, if memory serves, it took “Very Good” about 11 or 12 days to reach a million views, suggesting that the video was getting about 100,000 hits a day when it first came out.

Date                                  Views (in thousands)           Views added (in thousands)
10/19 (~9 pm PST)            1,181
10/20 (~9 pm PST)            1,219                                    38
10/21 (~9 pm PST)            1,235                                    16
10/22 (~3 pm PST)            1,266                                    31
10/22 (~10 pm PST)          1,284                                    18
10/23 (~9 pm PST)            1,320                                    36
10/24 (~9 pm PST)            1,357                                    37

You’ll notice that a lot more views were added on October 20 and October 22 than were added on October 21. Block B appeared on the music television show Inkigayo the 20th, while on the 22nd they appeared on Simply KPop and The Show. (These were all musical performances.) They didn’t have any television appearances on the 21st. So if Block B appears on a music show, then their views get a noticeable boost.

I checked their counts twice on October 22 because did a review of “Very Good” that was released at 2 pm PST. (I didn’t see it until an hour later; YouTube had a lengthy counter freeze that afternoon, so that delay may not have affected accuracy any.) Given EYK’s subscription base (they have more than 425,000 YouTube subscribers), I was curious to see what the impact was. By 10 pm PST EYK’s review video had 27,000 views, and it looks like a good chunk of that audience went on to watch “Very Good” itself.

The numbers for October 23 are complicated because in addition to the EYK video, Block B appeared as non-musical guests on the comedy program Weekly Idol. But the EYK review video had 60,000 views by 9 pm PST, the number of “Very Good” views increased by 54,000 in that same time period, and I think it’s fair to say that being on EYK helped. (I don’t know this for a fact, but I’m somewhat doubtful that lots of people tune into Weekly Idol because they hope to find new music–it seems like the kind of thing that would appeal more to existing fans, although I’m sure the show has its own fan base as well.)

On October 24 Block B did nothing promotional, but “Very Good” added a healthy 37,000 views nonetheless. On that same day, the EYK review added only 10,000 views. That could mean that there’s some other unknown factor driving “Very Good” views, but the optimist in me is hoping that this is the result of EYK reaching an audience that knows less about K-Pop, so there’s a surge in interest by those people (who hopefully tell their friends). Interestingly, the Very Good album has been in Amazon’s K-Pop top 10 consistently all day, and while in the past Blockbuster tended to follow Very Good‘s ups and downs (consistently trailing it slightly), on October 24 it decoupled and spent the day much further down the list.

Anyway, what I’m taking from this is that, yes, it’s certainly worth it for fans to work to boost song scores on the televisions shows and to vote up videos on Eat Your Kimchi (the former mostly reaches an Asian audience, while the latter reaches an English-speaking one, so both are good). Getting the most votes is the only way to get a video reviewed on EYK, but with the television shows voting actually counts for a relatively small fraction of a song’s score–sales matter more (details here). Since I wasn’t keeping track of YouTube views when Block B won Inkigayo, I have no idea how much or even if a win is better than just an appearance–but an appearance is pretty damned good, so keeping the group’s score up high enough for it to be invited back is important. The nice thing is, YouTube views add to the show scores, so there’s a virtuous cycle there of higher show scores => more appearances => more YouTube views => higher show scores.

(ETA: I did this for another week, just to make sure October 21 wasn’t some kind of weird fluke, and it wasn’t–on days they didn’t appear on television, they got about 10,000 fewer new views than days when they did.)


7 responses »

  1. Pingback: Will no one think of science! | My Other Blog

  2. Pingback: And I am torn | My Other Blog

  3. Pingback: Interesting….. | My Other Blog

  4. Pingback: How we doin’? | My Other Blog

  5. Pingback: Why certain things are not worth paying attention to | My Other Blog

  6. Pingback: Data and other random things | My Other Blog

  7. Pingback: On being a more-effective fandom | My Other Blog

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.