Micro advertising

October 26, 2001 12:44 AM

Micro Advertising - a way to make niche internet publishing profitable?: "It's all about targeted micro buys of advertising space - the old principle of getting the right message in front of the right group of people. Buying 10,000 impressions of your message in front of a very targeted audience for a low fee. " Thomas has created a Yahoo group to hash out the issues further, if anyone's interested.
posted by owillis to general weblog-related at 12:44 AM (20 comments total)

Good, topical link, important for MetaFilter and its members and screaming for discussion. And, yet again, owillis's post is given the zero treatment. He put his money where his mouth was, was one of the first to buy a textad and reported to us on its effects. I don't understand.
Together with the demographics thread this is probably the most interesting theme for everyone worried about making weblogs pay for themselves. I wish I could contribute, Oliver, but this is way out of my league. But I'm rarin' to read what others think.
Niche publishing is going strong, btw, with Primedia axing Brill's Content and Inside.com but still raking it in with Monster Truck Advisor, Rambo Crochet News and all the other specialized rubbish they produce. So...?

posted by MiguelCardoso at 7:53 AM on October 26, 2001

It seems that I am now a Thread Killer. I've lost my game, perhaps I need to be sent down to the minors to get my mojo back.
posted by owillis at 8:30 AM on October 26, 2001

I told you, bud.
posted by Doug at 8:52 AM on October 26, 2001

Hey owillis and Doug, we already did that yesterday. This is an interesting article and deserves discussing!

"It's all about targeted micro buys of advertising space - the old principle of getting the right message in front of the right group of people. Buying 10,000 impressions of your message in front of a very targeted audience for a low fee. Micro since hobbyists can spend 10$ on getting their message out about their personal site and the small startup can spend 100$ getting the message out about their new product. It's all about having 50 small advertisers on your site instead of four large companies. No need to hire a salesforce or hire a company to represent your site - just advertise the possibility of buying adspace on your own site. Convert the readers of your site to advertisers on your site."

Ring a bell? Alright then!

posted by MiguelCardoso at 9:04 AM on October 26, 2001

Actually, this is kind of interesting. It seems like it could work, to an extent. I don't think the average weblogger would make enough to even almost cover rent, but bucks is bucks, right? So when are you going to start the "OliverClick Network?"
posted by Doug at 9:07 AM on October 26, 2001

I'm more of a logistics/idea guy, which is why I'm glad there's a few other people interested in this - who could actually make something happen. If we were even able to generate income so someone could buy their pooch some dog food, that'd be kinda cool :).

Here's a little "thought list" I posted to the discussion list last night:

Central Authority
I wondered if there could be some sort of non-profit (or
profit) "central authority" that would administer such a system -
possibly staffed with donations from the associated websites or a
percentage of the advertising income.

You could then have an association/panel deciding which sites
can/can't join "the network".

Tiered Payment
What if someone purchased an ad across the entire network of sites -
couldnt it be setup so each site accrues monies in their account
based on the percentage of the campaign they delivered?

So say Microsoft has a $1,000 campaign. Your site "Joe's Weblog"
shows 10% of the ads, then your account gets $100. At a predetermined
cash level the system automatically transferred your funds to a
Paypal account or some such thing.

Ad Choice
Perhaps each site could choose what sort of ad they're interested in
displaying - text/banner/rich media
posted by owillis at 9:16 AM on October 26, 2001

whoa, the guy from flazoom is doing it exactly like metafilter now, too.
posted by mathowie at 10:14 AM on October 26, 2001

If the problem with existing ad networks is that ads aren't targeted very well, I fail to see how creating yet another network is going to help. The lack of targeting is a direct result of the network concept.

Individual sites need their own ad systems. Fortunately, software for doing this is available off-the-shelf, and PayPal makes it easy to accept payment.
posted by kindall at 10:43 AM on October 26, 2001

I think both could make sense. The problem (in my eyes) is that there isn't an easy way to buy ads. You could create a targetting sub-feature with a higher cpm and still have it network based. I also think a network could help boost the member sites traffic...
posted by owillis at 11:00 AM on October 26, 2001

I gotta agree with kindall. I think perhaps the missing step is something that will help individual sites build their own ad systems. like a blogger for advertising...no particular programming skill needed.

- form/code to accept ads (including interface to PayPal, etc.?)
- interface to review (accept, refuse, edit?) ads
- code to print ads onto site

(note: this would either have to be stored in a central repository, ala blogger, or create code that could be managed by each site, ala php-nuke)

because I think the thing about this that is quasi-revolutionary* is the "converting readers to advertisers" part. I've got a few nifty new things in mind for my personal site, and when I get them done, I'm gonna buy a MeFi ad! why? (a) it's cheap & (b) I know my audience. (and (c) it would give me a little thrill to see my site there. actually, that's an important part of the equation, and not something to be taken lightly. one's desire for fame among one's peer group is a useful motivating factor.)

creating a central network, while possibly allowing for interesting cross-fertilization, would reduce the (b) & (c) factors, as well as probably increasing complexity for the people running the sites. (reducing complexity, it seems to me, would be the primary reason for a blogger-esque system.)

*why only quasi? too many parallels with classified ads.
posted by epersonae at 11:09 AM on October 26, 2001

you know, kindall is right.

I've been chatting with people all day yesterday and today about this.

A central ad service is exactly what is not needed.

why? here's why:

- I think one of the only reasons why the ads worked here was because this community is fairly well-defined. You know if you plucked the average metafilter user from behind their computer, they probably surf the web a lot, like gadgets, and are interested in hearing about new sites they haven't seen before and new, geeky products. There are also obvious issues of trust. I've keep this site up for over two years, and now when I ask for money for promoting a site, people know who they are dealing with and what they are going to get out of it.

With a central service, you lose that definition. Each participating site has to somehow broadcast an identity of their own. There's no trust inherent in the participating sites, when you are an advertiser looking to place an ad.

- the ads work here (at least for me personally) because they are coming from the community itself. I love seeing new blogs or personal projects through an ad to the community.

A central service loses that. Anyone can place ads and they probably end up not talking to the community being advertised to.

- Standards go out the window when you have to spread things out among a network of sites. I have denied an ad in the past week. It was to a spam-looking site selling a get rich quick book. If I would have accepted it, people would start disliking the textads or ignoring them, since their experience with a spammy popup inducing site would sour their taste for clicking on those.

A central service would have to have some extremely powerful tools for letting participating webmasters block certain ads or approve them all.

- what is really needed is an easy DIY package in php or asp that someone can download and create their own, site-specific service. Everything I'm seeing in the past few days seems to sound like banner ads, circa 1996. We're just going down the same roads and making the same mistakes if we do a central, non-specific or targetted network of ads. Don't confuse the need for programming and software with a need for a new central authority to muck up anything good coming out of self-serve ads.

When you use a central service, as a webmaster, you get a little bit of conveinence. Someone else does the dirty work and gets a cut. But, you lose things like trust (who is this network run by, who are the participating sites and why should I trust them?) and specificity (if I want to put an ad on MetaFilter, would the users of FilePile also count? What if I just elect to broadcast my message across all blogs, since I have the money for it?). I think those two things are the only reason why they work here, and if you threw them out the window, you'd have all the effectiveness of a new doubleclick.
posted by mathowie at 11:16 AM on October 26, 2001

Matt, epersonae, Jerry:

You guys are making a lot of sense. I was part of a few ad networks in the past, including one with some proportionality built in - and they did go to poop or close to it eventually.

I must exorcise the demon of Kevin O'Connor from my head.

"Out charlatan, out!"
posted by owillis at 11:56 AM on October 26, 2001

Check out the talentless hack who just jumped on the bandwagon.
posted by owillis at 1:35 PM on October 26, 2001

OK, maybe I'm just not looking in the right places: where is the off-the-shelf software that I can use to put textads on my site?
posted by bjennings at 2:08 PM on October 26, 2001

WebAdverts is a good package.
posted by owillis at 2:24 PM on October 26, 2001

I think one of the only reasons why the ads worked here was because this community is fairly well-defined.

Don't forget the artificial demand created by Matt's long-standing policy of "no self-linking." So all the creative types working in the field on interesting projects are dying to put it before their peers (and what better place to do so than their beloved MeFi?), but cannot without breaching etiquette and policy. All of a sudden--pow! They CAN put proudly display babies...for a small fee. Brilliant! But the concept probably won't translate too well to other sites.

...oops... :::looks at all the people looking at him strangely::: I quoted Matt, didn't I...doh! Now I've stepped in it!
posted by rushmc at 5:33 PM on October 26, 2001

If it helps, the textads are great. i hate advertisements but have read the MeFi textads. They are brilliant. unobtrusive, small yet uniform. maybe this is not what is being discussed. but matt did a good move with the textads.
posted by clavdivs at 8:02 AM on October 27, 2001

I agree that a central server or network is the last thing that this initiative needs.

It would be useful for there to be a centalised list stating who offered such advertising opportunities, giving a profile and outlining what type of audience they get.

My initial reaction to this is that it is a very good idea, especially for metafilter. I would suggest however that to get this to really rock websites need to start profiling their users more clearly. It will work as a 'gimmick' for a while, but at some point you really need to start creating the data that is required to demonstrate that your users are the 'highly targeted audience' that is being touted.

Also, it may be worth pointing out that from a marketing perspective there is a direct contradiction between the concept of a highly targeted audience and cheap advertising, generally speaking.




posted by RobertLoch at 5:30 PM on October 28, 2001

I just started reading it, but this book seems to have relevance to the whole micro-market meme.
posted by owillis at 9:56 PM on October 28, 2001

I think there is plenty of good reason to have a central point from which to buy textads for all sorts of sites. The key is in filtering. Already pyRads has some filtering built-in: you will be able to select which of the sites across the network carries your ad. Google allows you to specify which kinds of words your advert will be targeted to. I think the key is to build some software which will target the ads to the subject matter and generate the price accordingly. I'd love to be able to advertise across pyRads, MeFi, Google etc etc all in one go - and I'd be happy to pay extra for it. If, that is, I had anything to advertise!
posted by skylar at 2:06 AM on November 16, 2001