Live data from Hacker News

Hot Dogs & Caviar

daltoncaldwell.com

41–47 of 47 posts

Re: Hot Dogs & Caviar

#41
post #17

To make matters worse, social media ad inventory is rapidly moving to phones, where advertising is even more challenging.

Mobile advertising is only in its infancy. If you think ads are invasive now, just imagine what life will be like when they move to a screen 1/10th of the size!

Re: Hot Dogs & Caviar

#43
post #26

A new “social ad unit” will be created at Facebook or Twitter in the next 12 months that will manage to be far more profitable than current ad units, not piss off users, and immediately be embraced by advertisers. I call them "tweets". Seriously, most tweets are ads, so just charge people to send them. If you have over ~1,000 followers then you are a de facto marketer and you could pay per tweet per active follower.

Won't that just incentive quantity of tweets rather than quality of tweets? Where is the value in that service?

Re: Hot Dogs & Caviar

#44

We get it... the ad model is broken, a subscription model is, in theory, always better. It's simpler. Generally, simpler is better. However, I don't think the app.net experiment will succeed. I'll be brief: - Outside of phones (& WhatsApp), I can't think of a single paid non-business communication platform. - App.net will not have a good one-player or even two-player mode (good 1-player: DropBox, Github, Evernote). -…

If we agree that the ad-supported model is broken, why not try to create an alternative?

You absolutely should. But it doesn't look like this IS the alternative (at least the viable one).

That said, no one appears to have any other firm ideas about what the viable alternative should look like (only what it shouldn't). And as it's your money and not mine, I say go for it, good luck!

Re: Hot Dogs & Caviar

#45
post #2

We are building the new social ad unit at Referly - and every word is written and targeted by humans, not demographic algorithms

The trouble with that approach is that while algoritms aren't great in this area (at the moment), they are at least trustworthy (in that they'll do what you tell them). People on the otherhand are often untrustworthy, and their definition of targeting is often likely to be quite different to yours and mine. It will be interesting to see how you solve the problem of untrusted users, my feeling is the ideal solution will end up being a combination of humans (users and staff) and algorithms. How to make that scale of Twitter/Facebook will be challenging.

Re: Hot Dogs & Caviar

#46

I can accept bad metaphors and analogies in HN comments; this a technically focused community and the comments are so far above above most online forums that nitpicking about literary devices would be ridiculous. But I can't forgive the hotdog/hamburger/caviar analogy in this full length blog post. It does not convey his message well and really does the article a disservice. This writing should have been more thought…

What's wrong with the metaphor? How would you have described the problem differently? I thought it was a really good description of the situation Facebook is in. People have speculated for years about all of the different ways they could produce insane revenue/profits (the hot dog -> caviar machine), but their current revenue (the hot dogs) is very lackluster.

I think the issue is that metaphors work best when they are about real situations. The idea of a metaphor (usually) is to help a person understand one situation by mapping their knowledge of another, more familiar, situation to the first. No one is familiar with the problem of turning hot dogs into caviar, because no one is trying to do it. If anything, the actual problem is more familiar to those on HN than the metaphor.

Re: Hot Dogs & Caviar

#47
post #26

A new “social ad unit” will be created at Facebook or Twitter in the next 12 months that will manage to be far more profitable than current ad units, not piss off users, and immediately be embraced by advertisers. I call them "tweets". Seriously, most tweets are ads, so just charge people to send them. If you have over ~1,000 followers then you are a de facto marketer and you could pay per tweet per active follower.

Won't that just incentive quantity of tweets rather than quality of tweets? Where is the value in that service?

If you're paying to send tweets, I'd think you'd want to maximize their quality.
Post reply on HN