Ideas For Startups: Advertising Ideas | Hacker News Reader

Ideas For Startups: Advertising Ideas

blog.awesomezombie.com

14 pointsig111 comments

Truly awesome ideas; ad feedback is something that makes a lot of sense. Context based ads based on intelligence embedded on ads is also a great idea. This would change advertising from static to a dynamic in nature. Giving advertisers the control is something that would create a lot of value.

A third-party network that sites like these could easily integrate into would mean more money for the sites, more relevant ads for the end users and an alternative to Facebook Ads

I couldn't agree more. Trafficspaces already provides it. See screenshots - http://www.trafficspaces.com/tour/

You can integrate demographic info from your database to show targeted ads just like in Facebook Ads.

Any plans to offer a service to let people buy ads across sites that run your product ?

(or alternatively an index listing your users so people can go to them directly to buy ads)

I'm doing a targeted demographic buys at the moment and I'm struggling to find places to advertise. Facebook is giving me solid results (good CTR, cheap CPC) but not getting enough volume so I'm looking for other sources. MySpace's demographic targeting is awful, PlentyOfFish has very low CTR, poor targeting and the interface is buggy.

We get a lot of enquiries about that so yes, it's definitely going to be available shortly.

We'll probably start off with an index because that's easier to do considering that each social network/dating site/news portal integrates their database in different ways.

Advertising idea: Can you build a website that aggregates both display and text ads. Basically when searching for X all I would only see ads for X, not content pages.

Some day back, I wanted to combine this (searching in ads and define your own selection algorithm) with the "real-time adsense" idea.

Then people could build the ad widgets, themself. Possibly in a plugable way, with a store/subscripton for them.

For example, a youtube ad widget that gets the relevancy from the transcribed text.

I actually pitched that exact idea at Launch48 earlier this year (and registered the domain adynam.com for it), but ended up building something else instead.

The main problem is getting the initial ad inventory, I thought about using affiliate links to start with, but my back of the napkin calculations showed that they were unlikely to bring in enough money to be competitive.

Although last week someone posted on HN saying that they'd built an adsense competitor using affiliate links and they were getting a decent CPM out of it, so maybe that is a viable route.

Yeah affiliate links seemed the right way, but I was also thinking about crawling the ads first, as it looks like some interesting data to have.

Of course money making would be a bit harder, but getting traction would be easier I think.

Actually, it should be validated and that seems to be the hardest point for me.

A Real Demographics Ad Network

quantcast? And since its inference, not regdata based, we can offer significantly more reach than eg blue kai or excelate

NB: for jackasses -- all opinions are my own. My opinions are not endorsed or authorized by my employers: past, present, or future.

edit: also, small site owners should get themselves into a realtime exchange so their inventory is available for purchase, assuming that nothing else monetizes better

I didn't know Quantcast operated an ad network. Interesting. In quantitative terms, how accurate is your inference based data?

On a different note, I have mixed feelings about exchanges cus I feel they overly commoditize inventory. Whilst that may be okay for large sites, it's disastrous for small sites. IMHO, the better strategy is to either join a vertical ad network or sell directly to niche advertisers.

Hey DJN (Niyi?) -- if you're reading this, please email me, and consider putting your email in your profile.

Reply on news.ycombinator.com