Concept for flat-fee CPC folksonomy directory

[ggibson400]

Hello all,

I would appreciate your comments/feedback on a new concept I'm working on. It's a web directory where everybody pays the same amount per click (say ten cents), relevency is driven by peer review, and it uses a "folksonomy" to organize information.

Allow me to explain:

Flat CPC: Advertisers don't bid for top placement, instead everybody pays the same per click.

Relevancy (how the top results are ranked): My company, Helium Exchange, has developed a peer review technology that allows competitors to rate each other while mitigating the tendancy to "trash" your competition. To get listed at the top, you have to be rated highly by your peers. To get rated, you have to do ratings - and the more ratings you do, the more ratings you get. The system uses side-by-side comparisons rather than scalar (e.g. one to five stars) ratings to make this work. You can see an example at www.squik.com

Folksonomy: The site will use tagging to organize the listings (see del.icio.us or squik.com for examples). You can choose any tags you like for your listing.

Example:

Bob's Boots decides to advertise on the site. Bob submits bobsboots.com to the directory and chooses the tags 'shoes boots "work boots" "cowboy boots"'. Bob then rates other sites that have chosen those same tags. He does as many ratings as he can because he knows that the more he rates, the more his site will be rated by his peers. Bob pays ten cents per click no matter where he ends up being ranked on the lists. Since his peers know that he runs a good company, he gets a top ranking and cheap traffic.

Notes:

People can't easily rate themselves. They can try to create a duplicate account and rate themselves, but they can't choose which sites to rate, so it will take them a while before their site comes up.

The concept relies on people to drive relevancy, rather than algorithms. Also, it leverages the self-interest of the advertisers to drive the relevancy/rating engine (because the more they rate, the more they get rated). The hope is that by harnessing the self-interest of peers, the end result will be a directory that delivers more relevancy than machine-organized results.

Thoughts?

[Marketing Guy]

I like the principle, and although I'm slightly cautious that it could be abused, it looks like you're certainly knowledgable enough to take a lot of those factors into account.

One potential flaw is that click fraud to burn out budgets could still take place (thereby removing a high "value" competitor). Perhaps consider a set fee advertising cost structure with potentially the option to review regularly to relfect traffic levels?

MG

[ggibson400]

Originally Posted by Marketing Guy

One potential flaw is that click fraud to burn out budgets could still take place (thereby removing a high "value" competitor). Perhaps consider a set fee advertising cost structure with potentially the option to review regularly to relfect traffic levels?

MG

Click fraud is definitely an issue. A set fee might work. PPC is nice, though, because you pay for performance.

[graeme]

Hi,

I like the concept , it's fun !


Graeme

[kidmercury]

there's gotta be a simpler way of explaining the concept, else folks like myself will be left scratching their heads.

[seobook]

the problem with flat cost per click is that leads have different values in different industries.

this leads to two fundamental problems
1.) expensive categories will get heavy coverage and inexpensive categories will get minimal coverage
2.) the inefficiency of universal pricing means that you will make lower profits and have less money to spend on infistructure and marketing

also the best sites may not be likely to want to pay at all (what do you do with sites like w3c) and people will nead heavy incentive to rate enough of the web to beat out modern search algorithms

[ggibson400]

Originally Posted by seobook

the problem with flat cost per click is that leads have different values in different industries.

this leads to two fundamental problems
1.) expensive categories will get heavy coverage and inexpensive categories will get minimal coverage
2.) the inefficiency of universal pricing means that you will make lower profits and have less money to spend on infistructure and marketing

also the best sites may not be likely to want to pay at all (what do you do with sites like w3c) and people will nead heavy incentive to rate enough of the web to beat out modern search algorithms

That's exactly the idea: the more valuable (expensive) categories will get more ratings done because there's more of an incentive for the advertiser to try to make it to the top of the list. Efficiency still happens - but rather than earning more revenue from more valuable categories, those categories get more ratings done (and therefore better relevancy).

[I gotta get better at explaining this stuff...]

[ggibson400]

Originally Posted by kidmercury

there's gotta be a simpler way of explaining the concept, else folks like myself will be left scratching their heads.

I completely agree. Check out ...//...squik.com which, in a sense is the first version of this concept that I launched a few weeks ago. I think it does an okay job explaining the idea of community ranked content. But I still need help. Suggestions?

[kidmercury]

Originally Posted by ggibson400

I completely agree. Check out ....//....squik.com which, in a sense is the first version of this concept that I launched a few weeks ago. I think it does an okay job explaining the idea of community ranked content. But I still need help. Suggestions?

in my opinion, complex/innovative/disruptive business solutions are best presented online using multimedia technology.
google's adwords demonstration, for instance, is a great example of what i mean. it's more engaging, which people will be more receptive to; most people dont have time or discipline or desire to learn new things, so making the explanatory process more sensory can increase hte likelihood that they'll pay attention to what you're talking about. even if it's just voiceover and a few slideshows with a bunch of clicking from the end user, that can often do the trick, in my opinion.

[kidmercury]

Originally Posted by ggibson400

That's exactly the idea: the more valuable (expensive) categories will get more ratings done because there's more of an incentive for the advertiser to try to make it to the top of the list. Efficiency still happens - but rather than earning more revenue from more valuable categories, those categories get more ratings done (and therefore better relevancy).

i think i get it now though. i like the idea. makes a lot of sense, in my opinion.

[seobook]

Originally Posted by ggibson400

That's exactly the idea: the more valuable (expensive) categories will get more ratings done because there's more of an incentive for the advertiser to try to make it to the top of the list. Efficiency still happens - but rather than earning more revenue from more valuable categories, those categories get more ratings done (and therefore better relevancy).

here are a few catches

general users will not be interested in using it if only highly commercial categories have relevancy, so you would need to bolt a regular directory or search system onto the undercovered areas. (like how uncoverthenet uses the yahoo api)

as a category becomes more valuable the users time is worth more money.

it is hard to bribe relevancy. if people are just rating themselves to the top it will not be relevant. and if users vote for something that you wont list at the top (because it doesnt want to pay) then you alienate your userbase. if you will list it without payment then there is also no reason for anyone to pay. paying would be like doing yahoo paid inclusion without the rapid refresh, etc.

the solution is, how can you make a system better than delicious or furl, then build a huge userbase, and then later extract profits from it.

many web businesses fail because they fail to make something useful that solves a problem. many others fail because they try to extract profits before they create value.

[ggibson400]

Originally Posted by seobook

many web businesses fail because they fail to make something useful that solves a problem. many others fail because they try to extract profits before they create value.

I couldn't agree more.

The problem I'm trying to solve is relevancy (i.e., search results that best meet my needs)

Right now when we search for something online we get results ranked two ways:

1. By how much an advertiser is willing to spend; and
2. By how some algorithm decides it should be ranked (link popularity, etc.)

In an ideal world the results would be ranked by how well a web site will meet my current needs.

My approach to solving this problem is to use human beings themselves to help decide which sites best meet which needs. The problem here is that it's time consuming for people to do this. They need a pretty strong incentive to sit down and match needs with sites.

Enter peer review. If the very individuals whose incentive it is to present their sites to prospective customers were asked to do all of this matching work in exchange for more customers, then they get rewarded for doing the work. That's my theory.

I've been testing this peer-ranking theory on two sites of mine: www.heliumfeedback.com (where screenwriters peer rate each other) and www.heliumknowledge.com (where answer givers rate each other). The experiments have so far been successful: a number of screenwriters have been discovered and signed movie deals from making it to the top of the list, and the better answers make it to the top on Helium Knowledge.

I'm now trying to apply this concept to search. Will advertisers rate each other in order to get rated themselves?

[seobook]

>Enter peer review. If the very individuals whose incentive it is to present their sites to prospective customers were asked to do all of this matching work in exchange for more customers, then they get rewarded for doing the work. That's my theory.

if your system is easier to spam then it will be hard to get a large courpus of users.

if your system is harder to spam then marketers and business owners would likely rather just spam other easier to spam systems.

the key though is getting demand to use it and some sort of viral marketing angle. and the profit later. IMHO

[glebarr]

I predict great success! Keep up the good work....

[rcjordan]

>success

It depends on the marketing experience and brain-power of your niche. If you're in a modestly sophisticated marketplace then I think there's a chance. However, if you're in a generally luddite-filled category (independent bed & breakfast inns come to mind) then my own ad/sponsorship sales indicate that you're likely in for a serious disappointment.

[ggibson400]

Originally Posted by rcjordan

>success

It depends on the marketing experience and brain-power of your niche. If you're in a modestly sophisticated marketplace then I think there's a chance. However, if you're in a generally luddite-filled category (independent bed & breakfast inns come to mind) then my own ad/sponsorship sales indicate that you're likely in for a serious disappointment.

That's a very good point. Thanks.

[seobook]

Originally Posted by rcjordan

marketing experience and brain-power of your niche

perhaps I was being a twit

I thought it was going to be a general non niche type thing.

[rcjordan]

>twit

No, your posts were dead-on target, Aaron. Very perceptive. Tempered by experience, as I recall.

My own experience confirms all of the negatives you've outlined. In a nutshell, 'properly' structured directories are often too granular for their own good when it comes to driving revenue. As you slice and dice the categories, sub-cats, and sub-sub-cats you're doing the same to the traffic.

[mrdch]

I am not sure, but there seems to be a fundamental taxonomy issue with your suggestion.

You talk about a 'directory', but you never refer to CATEGORIES. Instead, you are mentioning KEYWORDS, which are SELF CHOSEN.

That implies, I understand, that people will not be able to BROWSE the directory - they will only be able to search on it using keywords.

If so - then I can fully understand why you are concerned with relevancy. After all, a site owner may choose ANY number of keywords for his/her site, isn't it? As this a CPC model, a site owner may choose HUNDREDS of keywords! Mind you - they may all be relevant (in his mind), but it will take a hellish amount of effort for the 'peer review' to go through all of them and rank/rate them. I am not quite sure why would anyone make that investment.

To get listed at the top, you have to be rated highly by your peers. To get rated, you have to do ratings - and the more ratings you do, the more ratings you get. ... Since his peers know that he runs a good company, he gets a top ranking and cheap traffic.

The scenario seems quite far-fetched. It is very doubtful that the rater knows anything about you in the first place, and if he does - he may not have a good opinion about you anyway! It is naive, at best, to assume that the more ratings YOU will be doing - the higher YOUR site will rank.

That's my 2p anyway