Ktown4u Is Not a Reliable Predictor of Actual Sales, so Stop Placing Too Much Emphasis on It

    I often see users on this forum hyping up Ktown4u sales, a phenomenon that was made very apparent by this recent thread.


    This made me wonder: how well do sales figures on Ktown4u predict actual Gaon sales?


    To find out, I took the sales figures for the top 100 best selling albums on Ktown4u, and then located the corresponding Gaon sales for each one (combining versions if necessary).


    Following, I simply generated a graph that plots Gaon sales for each album against its Ktown4u sales, as shown below:


    Gaon_vs_Ktown.png


    As we can see, the results are all over the place. For example, just look at albums that have sold around 100,000 copies on Ktown4u.


    On one hand, we have Winner's Remember, which sold 97,072 copies on Ktown4u, and then went on to sell only 113,346 copies according to Gaon.


    Compare that to BTS' Love Yourself: Her, which sold only 92,308 on Ktown4u, but has Gaon sales of 2,514,204.


    That is a huge range of Gaon sales for almost the same Ktown4u sales, and there are similar examples of such disparities throughout the data.


    Also, when I tried to fit a trend line to the data points, I got an R2 value of only 0.2, which is quite low. If you're not familiar with R2, it's a statistical value that describes how much of the variation in the dependent variable (Gaon sales) can be predicted by the independent variable (Ktown4u sales).


    In the end, remember that Ktown4u is just one retailer out of thousands where people can purchase their K-pop albums.

    No surprise with your conclusions. In the end, ktown4u is just one kpop shop out of many.

    It entirely depends on where the fandom prefers to buy. It's even less reliable for any groups/artists whose albums are sold on a company endorsed shop like weverse shop.

    I think, like you said, it's always good to keep in mind that in the end it's all just speculation and rough predicting, even the best analysts aren't right 100% of the times (let alone those who aren't).

    There's simply too many variables involved that can influence the outcome, so focusing on 1-2 metrics for a model is shaky at best.


    That being said, for general, rough speculation and guesswork, there's no problem to use the available data to have some indication.


    For Ktown4u for example, you could use datasets that are close to what you're trying to extrapolate, like former and recent sales of the same group or sibling groups of the same company.


    For example, Wendy had 51k preorders in Ktown4u. Wendy's album 'Like Water' had something like 68k first day sales in hanteo.

    Red Velvet had 155k preorders in Ktown4u for Queendom. Their album had 150k first day sales in hanteo.


    While still a rough guess, you can see that both albums' first day sales are roughly around the Ktown4u preorders.


    This is where it gets interesting:

    Wendy's Like Water had:

    - 1st day hanteo sales: 51k

    - first month Gaon sales: 146k (now: 181k)


    RV's ReVe finale had:

    - 1st day hanteo sales: 30k

    - 1st month Gaon sales: 104k (now: 194k)


    RV's Queendom:

    - 1st day hanteo sales: 150k

    - 1st month Gaon sales: ???


    Will Queendom follow the same pattern and have about 400-450k for 1st month Gaon sales?


    Like said, it's all guesswork and so many variables playing a role.

    For example, if SM wouldn't bother to restock or decided to only restock to up to an additional 100k and no more, then that would be it, that'd be the limit.


    For general indications and rough guesses, it's interesting/entertaining to use available stats like Ktown4u.

    After all, based on the sales increase that the majority of groups experienced since 2019 with a factor of 1.5-3x, it wouldn't be a dangerous bet to say that a group who had their last album released in 2019, would see a significant increase in album sales.


    So if preorders show significant increase compared with a group's last comeback and album, chances are high that this will result in significant increase in album sales too compared with their last album release. Especially if multiple samelike indicators and metrics show a significant increase.

    Knew this since a lot of STAYs boycotted K-town4u this comeback - and yet they have 830,000 pre-orders whilst having merely 55K sales on Ktown.

    Yup. Especially for many hybe acts and/or acts that are on the weverse shop. you can see for example how SVTs ktown4u numbers decreased despite their overall sales increasing once more carats started using the weverse shop and/or when they had better international distribution. E.g. Heng:garae was their first album i think when they were on the weverse shop but carats didn't use it too much immediately. As the fandom got more comfortable with it, you can see the decrease in sales from them on ktown4u. Eg. Heng:garae sold over 500k on ktown4u while Your choice sold around 200k on ktwon4u... despite that... Your Choice has sold more than Heng:garae on Gaon.

    Another example is MOTS:7 and MOTS: Persona not being the best selling albums on ktown4u but has around 2-4 times the amount on Gaon as the best selling albums on Ktown4u.

    last time I shopped at Ktown4u my TXT album was charged customs, when other sites like weverse didn't, so stop shopping on that site, I only use it as a last resort. Regarding giving figures, I think they should not do it because they are not exact at the time of sending because there are always returns, cancellations, etc.

    Quote from Anpanromane

    As a fan of bts and txt I'm aware of that, Weverse and other shops are much much more used to buy the albums.

    Quote from nahadoth

    It entirely depends on where the fandom prefers to buy. It's even less reliable for any groups/artists whose albums are sold on a company endorsed shop like weverse shop.

    Quote from kidsgonestray

    Knew this since a lot of STAYs boycotted K-town4u this comeback - and yet they have 830,000 pre-orders whilst having merely 55K sales on Ktown.

    Quote from worldlove27

    Yup. Especially for many hybe acts and/or acts that are on the weverse shop. you can see for example how SVTs ktown4u numbers decreased despite their overall sales increasing once more carats started using the weverse shop and/or when they had better international distribution.

    Quote from moonlightelle

    there's plenty people who use other sources such as weverse etc.

    Yes, I'm glad that a lot of people mentioned this. Different fandoms have different buying patterns, and I think that's a major reason why we should be careful about using Ktown4u sales as a proxy for overall sales when we're comparing two different artists.

    Quote from Millano

    So there’s no reliable numbers for total sales incl. Weverse?


    Trying to find out sales for several acts - B.I Waterfall total to name one

    I would say that Gaon sales are the most reliable, and that is generally the consensus on this forum from what I've read.


    As for B.I's Waterfall, it had sales of 99,678 according to Gaon (source).

    As an addendum to my initial post, I decided to determine a "Gaon Sales Multiplier" for each album in my data set simply by dividing each release's Gaon sales by its Ktown4u sales. In other words, I determined how many times more an album's Gaon sales were than its Ktown4u sales.


    I then plotted these "Gaon Sales Multiplier" values against each release's Ktown4u sales, and the resulting graph is shown below:


    GaonMultiplier_vs_Ktown.png


    First of all, we see that "Gaon Sales Multiplier" values do not appear to vary depending on the amount of Ktown4u sales.


    More importantly, we see a huge range of values for these multipliers. At the low end, Winner's Remember had a Gaon multiplier of 1.17, while at the high end, BTS' Love Yourself: Her had a Gaon multiplier of a whopping 27.24.


    The mean multiplier is 5.10, while the median is 3.38. The standard deviation is 4.63, while the relative standard deviation is 4.63 ÷ 5.10 = 90.91%, which is quite high, implying that the multiplier values are very spread out instead of being clustered around the mean value.

    Quote from Zenning

    For Ktown4u for example, you could use datasets that are close to what you're trying to extrapolate, like former and recent sales of the same group or sibling groups of the same company.

    I appreciate the detailed response. And yes, I do agree that it's probably safer to use Ktown4u sales for prediction purposes when you're dealing with sales from the same artist.


    Now your post did inspire me to take a closer took at the relationship between Ktown4u and Gaon sales for Red Velvet.


    I found the Ktown4u sales figures for all albums from Red Velvet, Irene & Seulgi, Wendy, and Joy, then found corresponding Gaon sales numbers (using data compiled by @halfdavid).


    I then determined the relationship between Ktown4u sales and Gaon sales and charted the results over time:


    RV_Ktown4u_Gaon.png


    As we can see, Ktown4u has become a larger and larger portion of Red Velvet's sales over the past couple years, and now represents around a third of the group's Gaon sales.


    Thus, Queendom's current Ktown4u sales of 162,872 could translate into nearly 490,000 copies sold on Gaon (assuming a 3x multiplier) if recent trends continue.

    Ah, that's nice!


    Well done :-):claps:

    Quote from atropos

    Depends on the group. For SM groups and Blackpink ktown4u is a good indicator of sales. Now for jyp or bight its not as the fans get the albums from various other sites.

    It depends.


    If the multiplier is fairly constant across album releases, like for example 10% of ITZY's or TWICE's sales are bought at ktown4u as seen with multiple albums, then you could use the ktown4u data too, but with a different multiplier.

    However, if it fluctuates too erratically even within a year, then it's no use.


    wsoet, is that data or charts you can easily put together, based on the data you already have? Would be interesting to see, if the multiplier is different yet still constant when checking up on other, JYP or BH affiliated, groups.


    If the move towards weverse distribution and buying is a fairly new trend for all fans of HYBE and JYP groups, then probably not, it'll probably fluctuate too much.

    Quote from Zenning

    is that data or charts you can easily put together, based on the data you already have? Would be interesting to see, if the multiplier is different yet still constant when checking up on other, JYP or BH affiliated, groups.

    It's not easy to put together, since I have to look up all the sales figures manually. However, I did gather enough data to put together this chart for the current best selling girl groups (and associated sub-units/solos), since I was curious as well:


    Ktown4uPercentGG.png


    Some observations:


    • In the past, Ktown4u consistently represented about 6% of Red Velvet's Gaon sales, but this percentage shot up dramatically starting in late 2019. What was the reason for this? And do we expect this trend to continue?


    • Twice's sales on Ktown4u have always been low, but did rise somewhat, peaking with More & More (10.87%) before declining noticeably with their latest releases. But what was the reason for this decline? And why were their sales so low to begin with?


    • Blackpink's Ktown4u percentage varies wildly, with the lows being Square Up (5.67%), How You Like That (8.31%), and the regular version of -R- (5.44%). ForgottenSoul  SAINTROSE  atropos, since you all stated that Ktown4u is a good predictor of overall sales for BP, can you explain this?


    • Historically, Itzy has done better on Ktown4u compared to Twice, but why? Also, why has Guess Who (3.09%) performed poorly on Ktown4u compared to the group's other albums? Are Itzy's recent Ktown4u sales low for the same reason that Twice's are?


    Overall, I'm not seeing all that much consistency in the percentage of Gaon sales represented by Ktown4u sales, even when we're talking about the same artist, so I still think we should be careful with our predictions.

    Quote from SAINTROSE

    I guess u are right. Can u explain the low percentages of R because i don't understand how it's that low?

    Yes, I was also surprised at how low the Ktown4u sales of -R- (regular version) were. Here are the actual sales numbers I used:


    Date Album Artist Ktown4u Sales
    		 Gaon Sales Gaon Multiplier
    		 Ktown4u %
    2018-06-15 Square Up Blackpink 22,619 399,185 17.65 5.67%
    2018-11-12 Solo Jennie 35,333 107,198 3.03 32.96%
    2019-04-05 Kill This Love Blackpink 123,215 492,805 4.00 25.00%
    2020-07-17 How You Like That Blackpink 25,512 306,900 12.03 8.31%
    2020-10-02 The Album Blackpink 525,504 1,384,313 2.63 37.96%
    2021-03-16 -R- Rosé 25,995 477,859 18.38 5.44%
    2021-03-16 -R- (Kihno) Rosé 28,791 57,441 2.00 50.12%
    2021-04-19 -R- (Vinyl) Rosé 26,899 52,518 1.95 51.22%


    Can you think of an explanation for why certain releases have such low Ktown4u sales?

  • This thread contains 7 more posts that have been hidden for guests, please register yourself or login to continue reading.

Participate now!

Don’t have an account yet? Register yourself now and be a part of our community!

Register Yourself Login