Bitcoin: U.S.A. vs. The World

[BitPappa]

I started experimenting with ads for my Bitcoin Keychains on Facebook. I have two identical campaigns: one is just shown to people registered as being in the U.S., and the other ad is served to people in the top 15 Bitcoin countries (other than the U.S.). I was surprised to see literally 10x the clicks and response from from "the world" vs. the U.S.

Anyone want to try to interpret these results?

[darkmule]

I started experimenting with ads for my Bitcoin Keychains on Facebook. I have two identical campaigns: one is just shown to people registered as being in the U.S., and the other ad is served to people in the top 15 Bitcoin countries (other than the U.S.). I was surprised to see literally 10x the clicks and response from from "the world" vs. the U.S.

Anyone want to try to interpret these results?

I wouldn't take that too seriously.  There are a number of possible confounding factors.

1)  People in the U.S. interested in Bitcoin prefer to obscure their actual origin and don't register as being in the U.S.
2)  There are more unrecognized spiders in other countries (maybe).

Number two is just wild speculation, but number one is pretty likely to be true.

[pedrog]

• You may have more ad competition in the US.
  
• US users are more ad blind.
  

Second hypothesis doesn't seem very good, but it may be due to ad competition.

Are ads similar? Same images?

[BitPappa]

• You may have more ad competition in the US.
  
• US users are more ad blind.
  

Second hypothesis doesn't seem very good, but it may be due to ad competition.

Are ads similar? Same images?

Ads are identical, just the countries the ads are served to are different. And the response I'm noting is based on percentage, not raw numbers: 4% of non-U.S. ad viewers click the link, whereas .4% U.S. ad viewers click the link.

[pedrog]

• You may have more ad competition in the US.
  
• US users are more ad blind.
  

Second hypothesis doesn't seem very good, but it may be due to ad competition.

Are ads similar? Same images?

Ads are identical, just the countries the ads are served to are different. And the response I'm noting is based on percentage, not raw numbers: 4% of non-U.S. ad viewers click the link, whereas .4% U.S. ad viewers click the link.

Never used Facebook, but can you see ad competition for your keywords based on targeted countries?

That's a huge CTR difference...

[inform]

Wow this sounds good your experiment

[neurobox]

Good points above on spiders.. Can you extend the analysis to include some measure of engagement? Are you actually selling more outside of US?

[BitPappa]

I'm not aware of a way to see anything about the competition's ads. Also, I don't sell the keychains directly: I have a couple of online stores that sell them, and I'm not able to keep track of sales precise sales metrics.

If this kind of stuff interests you, create an ad on Facebook (you don't ever have to pay anything if you don't run it), and you can choose different likes for your target audience ("Bitcoin," for example), and choose different countries, and see how large Facebook says the audience is. You can play with different socio-economic parameters (ages, sex, education, etc.) and see how that affects the audience size. I've always thought it to be very interesting.

Anyway, thanks for your responses. I was very surprised by the difference in response percentage. My hunch (no particular reason) is that Bitcoin is much more trendy or "cool" in other countries, so that might be a reason people in countries outside the U.S. took the public action of liking a Bitcoin subject, as well as the private action of clicking the link.

I've advertised an iDevice accessory on Facebook before, and never noticed a huge difference in response between the U.S. and other countries (though it's likely iPads and iPhones were much less prevalent in other countries, for various reasons).

When I have more time I'll dive deeper, and try to look at the particular countries people who respond are from.

[tickets]

What was the total number of clicks?  If we're talking about just a handful, perhaps the result isn't statistically significant to begin with...

Also, which country accounted for the most clicks?

[inform]

Some news your experiment brother we very interesting this details 

[BitPappa]

Here are some figures based on just a couple of days running the Facebook ads, with spending limits set to $2 each day for each ad (one ad for the USA, and an identical ad for other countries).

The first number is the number of ad impressions. The second is the number of clicks. And the third is the percentage of clicks. The difference is 1-2 orders of magnitude, even larger than I first reported.

USA                                  
Right Column Ad
  6,049       2       0.027%   

News Feed Ad   
16,690      19       0.079%   


"The World"                    
Right Column Ad
  3,680      29      0.643%   

News Feed Ad
  3,397     153      4.349%   


Note the "the world" consists of these countries (supposedly the 15 biggest Bitcoin countries, by some metric):
Argentina - Austria - Australia - Brazil - Canada - China - Germany - Spain - Finland - United Kingdom - India - Italy - Netherlands - Russia - Sweden

As a control, I could try to run a non-Bitcoin ad, and see what happens. Maybe some of these 15 countries are click-happy? I don't know.

And that's all the info I have at this time regarding the ads. Note that back in July, when I was selling the keychains myself for BTC (now I use BTCtrinkets.com and BitcoinAnarchy.com for online retails sales), the most non-USA orders came from China.