Bryan Caplan  

"Likes" as Tipping

The Ideological Turing Test... Tacitus, Peace, and Desolation...
If your work has a tip jar, conventional wisdom tells you to "seed" it.  Before your first customer shows up, put some of your own money in the jar.  When the marginal person sees money in the jar, he'll feel more obligated to add some.  Your "seed" might even provoke a chain reaction - imagine a world where no one adds money to an empty jar.

Question: Do Facebook "Likes" work the same way?  Does the fact that one person likes something prompt more people to like it?  Is there a chain reaction?  Can you increase your "Likes" (by more than one) by [shudder] simply self-Liking everything you write? 

COMMENTS (15 to date)
Jeff writes:

I wonder if there is different behavior when the "likes" are anonymous vs when it is possible to see who "liked". Imagine if you knew for sure that _all_ of the money in the tip jar was from the proprietor.

NZ writes:

Maybe. The analogy would work better if Facebook "likes" were anonymous, the way cash tips are. But as far as I know, clicking on the "X number of people like this" link reveals exactly who liked it. So even with a second fake account people would start to notice the pattern and, like, you risk losing credibility.

Ryan Murphy writes:

The technical definition of Adam Smith's "sympathy" would certainly suggest so, wouldn't it?

I don't think many economists would be very good at answering this question because we aren't built to answer it.

Scott Wentland writes:

I can understand why you would want to increase tips, because it's money. But why deliberately increase "Likes"?

Lori writes:

1. Of a certainty Facebook definitively knows the answer to that question.

2. They ain't gonna tell u. It's their trade secret.

Scott writes:

Seeding a tip jar can be done secretly, and it only works if customers assume that you don't seed. There's no pressure to tip you if I know that you're the one who put the seed money in the jar. Self-liking cannot be done secretly, so it won't work. There's no social pressure to like you if I know that you're the only other person who has liked you.

Brandon Berg writes:

Might work, until someone figures out what you're doing. The Internet does not look kindly on such transgressions.

Shane writes:

What a fascinating idea. I've wondered about the dynamic on the Freakonomics blog, where readers can rate one another's comments. Very low-rated comments are hidden (unless you specifically select them) while high comments are highlighted.

So I wondered if this could lead to a kind of accumulative advantage scenario, with high-rated comments attracting more views and more rates and low-rated comments simply disappearing.

I know a member of an online discussion forum who always starts a new thread with TWO posts instead of one. When other members see two posts, they seem to think that a discussion is already taking place, so they click on the link and - he believes - are more likely to get involved.

I work with a lot of different blogs, and in my experience, seeding Likes or other kinds of social media acknowledgements with your own 1 Like or with a few Likes by you and your close buddies is not very effective. It quickly becomes obvious to readers that there is always the same small number of Likes or other popularity votes.

However, once a certain critical mass of Likes or such is exceeded, there is a small positive effect in bringing in new readers and starting a chain reaction. The minimum size of the critical mass depends, I think, on the alternative blogs the readers frequent. For economics blogs, 20 Likes might make a post look like hot stuff; for readers following Kim Kardashian's wedding plans, the threshold is likely in the thousands.

Facebook and Twitter Shares are more effective than Facebook Likes (in the sense that they burgeon more quickly into all kinds of social popularity hits). The reason, I believe, is that the person actually writes something or at least takes more than the one moment it takes to click a Like button. That you enjoyed something enough to spend a bit of time promoting it probably signals others that the entry was possibly worth a read--and maybe further promotion.

Seeding comments with some of your own comments and those of your friends does, in my experience, have a small positive effect which can burgeon for a blog. Again, I think the reason is that the time expenditure involved is a stronger signal than merely a positive vote.

Who does the Liking or Sharing is of course the biggest signal. When someone with a big name posts about, Likes, or Shares a blog post, that generally brings new readers--and even more Likes, Shares, comments, etc.--in in droves. So, Bryan: If you are going to seed, you should do it by making a side-deal with, say, Tyler Cowen to always Like your posts in exchange for your always Liking his. Of course, the amount you might have to pay him to compensate him for the erosion of his reputation might be deterring! A side-deal to do it only occasionally might result in a lower price--and more effectiveness because it's less likely to be obvious to readers.

[P.S. I've never seeded the Like or Sharing buttons on EconLog, but I did test them initially to make sure they worked. Also note, as a technical aside, the number of FB Likes visible next to a FB Like button on this and other blogs is usually the same number as that visible next to a FB Share button for a page--reflecting the total of both kinds of FB activity--even though the buttons function slightly differently when users click them.]

Joe Cushing writes:

The problem is, likes aren't anonymous.

John Hall writes:

Facebook also bumps up things that you or your friends liked.

A. writes:

I think it's probably true, but isn't trying to maximize "likes" a little, uh, pathetic?

Jack Davis writes:

I absolutely believe likes increase the chance of other likes following. Any number of psychological experiments prove that people follow the herd. While it's true that the likes are not completely anonymous, if there a number of likes few people click to see the names. I certainly don't.

And, yes, I agree with A. that it is pretty pathetic.

Tom writes:

I'll second pretty much all of what Lauren says. Commenting is worthwhile if (a) there's a vibrant user community and/or (b) post author response. I'll also add that some high status people are good recommenders and others are not, and I filter accordingly.

Peter St Onge writes:

Self-liking, and buying likes, are common ways to seed social proof in social networking.

People buy twitter followers, facebook likes, RSS subscribers, Youtube views.

They also form reciprocal networks (cf "like you back" Facebook pages).

So, yes, your intuition is correct :)

Comments for this entry have been closed
Return to top