Ask HN: I wrote this, what now?

13 points by nl 15 years ago

Over the past few months I've been experimenting with affiliate marketing on a blog I run. To my surprise it has been moderately successful.

Being more a software person than a writer, I wrote my own contextual ad delivery system, using affiliate ads for inventory.

Last week, Google rolled out a very similar feature for AdWords [1], claiming it "leads to more clicks, higher quality leads, and higher ROI for your search ads".

Encouraged by that I built a technology demonstrator for my ad delivery technology: http://demo.qontex.com/

Based on my experiments, the contextual technology works roughly as well as Google AdSense does (I do have an advantage in that I don't need to worry about click-fraud).

Here's a few samples, based on the top HN stories when I submitted this:

http://demo.qontex.com/?url=http://www.thestar.com/news/world/article/744199---israelification-high-security-little-bother

http://demo.qontex.com/?url=http://stacksmashing.net/2010/11/15/cracking-in-the-cloud-amazons-new-ec2-gpu-instances/

http://demo.qontex.com/?url=http://brokenbottleboy.tumblr.com/post/1583250140/facebook-messaging-scorned-by-nerds-built-for&categories=Books

I'd love some suggestions for what I should do with this? Should I try and licence the technology (to who?) or should I try marketing it to webmasters (and how would the finances work there? It seems odd to charge a website to run ads on it)? Or maybe I should be paying people to run my ads on their site?

[1] http://adwords.blogspot.com/2010/11/product-listing-ads-rolling-out-to-all.html

ig1 15 years ago

I had actually planned a startup building something similar (+ip geolocation to pick appropriate affiliate links), however I ended up moving onto another idea because I couldn't make the numbers work out.

To make the same amount of money as from CPC ads, you have to get a much higher click-through rate than CPC ads conventionally get. I wasn't convinced I could build contextual ad technology that would result in the higher CTR to make it worthwhile. Congrats if you've managing to pull that off, I'd be interested to hear what the equivalent CPC/CPM would be of your ads.

My own business plan was to target sites which can't use Google Adwords, primarily sites which require logins to access content (dating sites, small social networks) or do dynamic page generation.

Another good market might be blogs which don't monetize very well at the moment if your tech works well in that context. I believe jgc posted a while back saying despite his high traffic on his blog he couldn't see traditional adwords bringing in much.

My plan to monetize initially was to take a percentage cut, use your affiliate code 20% of the time and the websites affiliate code the other 80% of the time. That way you avoid having to handle the payments yourself but get income very early on.

  • nl 15 years ago

    For some niche markets affiliates work pretty well.

    I don't think I'm allowed to share Adsense numbers, but from the one (mildly) successful blog I make money on I get around 30 times as much money from affilates compared to Adsense.

    (and yes, it's becoming clear I'll need to do the geo-location based filtering of ads)

    • ig1 15 years ago

      Let me make some ballpark guesses here: Adsense for blogs normally comes to around $0.10 - $1.10 CPM, so I assume your making it the region of $3 - $30.

      If your affiliate ads are generating > $10 / CPM then you're home dry, your product will sell yourself. Less than $10 then you'll probably find it hard to compete against the major blog ad networks, however small-to-midtier blogs that use adsense may well be a good target market for you.

      • nl 15 years ago

        Your assumptions are valid.

notahacker 15 years ago

If you're marketing to webmasters, maybe a Skimlinks style model where you manage all the affiliate programs etc on behalf of the end user and pay them a cut of the revenues generated from their site?

  • nl 15 years ago

    I'd never seen Skimlinks. Have you used it before?

    Looking at it now that looks to be quite an interesting model.

    They claim "you can earn up to 110% of what you would have earned if you managed your affiliate links programme yourself." which sounds doubtful at first. Thinking about it though, most affiliate programs increase your commission as you sell more, so it's possible this could work out.

    • notahacker 15 years ago

      Haven't used it yet.

      I'm guessing that that economies of scale and quality control over their publishers they can potentially negotiate larger-than-normal commission arrangements with some providers, and they only need to take a small percentage to be profitable.

gregpilling 15 years ago

How about an openx style business model? or try to get some feedback from the affiliate marketers at WebmasterWorld.com and see what they think.

  • nl 15 years ago

    OpenX has an interesting model and it could work for me. I haven't been able to work out how much OpenX actually costs, though. http://www.openx.org/publisher/compare-ad-server-products says they charge "for overage impressions" (whatever that means) or "based on impression volumes".

    The obvious question is how much per impression because thats the difference between this being viable and not.