So you were thinking of using adsense to generate income? Think again | Hacker News Reader

So you were thinking of using adsense to generate income? Think again

jacquesmattheij.com

150 pointsjacquesm85 comments

Google has historically not been very open about this, but they're very gradually becoming more transparent on ad pricing: their cut is 32% for ads on the content network, so the advertiser is getting charged $1.47 for every $1 you see in your stats.

See: http://adsense.blogspot.com/2010/05/adsense-revenue-share.ht...

Their customer service is abyssmal. This is intentional. Everyone doing business with Google needs to go into that relationship with eyes wide open in that respect.

reocities.com probably got banned because -- and this is a pretty important concept -- there is no incentive for their Indian quality rater to think "Hmm, although in five seconds that sure sounds like it is a typosquatting spam site which is infringing on the Geocities trademark and built on industrial strength copyright violation, I'm going to give them the benefit of the doubt and investigate this thoroughly."

Typically, AdSense pays much, much more poorly than individually negotiated advertising buys, which is why most major sites use it only for placing remnant inventory. (It isn't called "webmaster welfare" for nothing.) You also get the best results of it if your page is essentially an extension of someone's Google search, which is why eHow gets $2.50 CPMs from just one of the thirteen ads on their bingo-related pages: their traffic is overwhelmingly (60%) direct from Google for queries with demonstrable intent in commercial areas. (This is better than quadruple the performance of most sites I advertise on. sigh I'm giving aid and comfort to the enemy.) Demand Media gets the psychology of advertising. Other people wanting to do it should study and study well.

"Typically, AdSense pays much, much more poorly than individually negotiated advertising buys"

What you are saying is true in some but not all cases... It's a bit more complicated than that and depends on how targeted your app/site is, as well as how much traffic you are getting.

Small startups with a lot of possible advertising topics can't solicit every possible niche. You would have to have a sales person selling bingo ads one minute, then the other niche, then the other niche etc. and then manage all those relationships.

What Adsense does is manage all that for you. No salesperson investment, no ad software investment, no prospecting etc. Also, Adsense does some other things like customizes the ad to the user's behavior (across sites), not just the page content.

Maybe what you are saying makes sense in the case of Demand Media because they have a lot of sales people.

But it's not practical for a small startup if they cover more than a few topics. They will end up with less-targeted run of site ads with lower CPM and more ad management headaches.

If you're a startup with lots of possible advertising topics, and monetizing with advertising, you're probably a UGC/social media site. Advertisers will see you as not much more than a glorified general discussion forum, and pay accordingly.

Forum traffic, unless it's very targeted to a specific niche, does NOT convert well. You won't get a good CPM no matter what. The exception to this rule is for social networks, who know enough about their users to allow demographic targeting. Even then, the CPM isn't great(I pay less than 20 cents CPM on Facebook for example). Social networks make up for this on volume, because social apps tend to generate a lot of pageviews.

That said, you will still do better with ad sales people even if you're selling general, untargeted inventory, especially if you're selling to brand advertisers.

Every couple weeks I get pestered by ad sales people from some second-tier social network who are willing to do almost anything to get an IO and commitment for a large number of impressions. Sometimes this works, even for low-quality traffic. It's possible to sell almost any kind of traffic as long as the numbers are right, and the benefit of having ad sales people, at least good ones, is that they can sell large blocks of inventory to big advertisers at once and hopefully maintain a relationship if the traffic backs out.

As for Demand Media, you're wrong about how banner buys work. If I go to cracked.com, which has lots of diverse content(and low-quality traffic), I don't see ads that are targeted to the categories or specific niches. I see banner ads from big CPG advertisers doing branding sitewide.

"Forum traffic, unless it's very targeted to a specific niche, does NOT convert well. You won't get a good CPM no matter what."

No, I'm talking about informational and news/blog sites that have hundreds or thousands of subtopics. They can get excellent CPMs with Adsense.

"That said, you will still do better with ad sales people even if you're selling general, untargeted inventory, especially if you're selling to brand advertisers."

No that's not true, unless you have a big enough company to support a lot of salespeople and really high untargeted traffic -- which is not most startups. (That's why I mentioned in my original reply that it partly depends on your traffic.)

"you're wrong about how banner buys work"

No I'm not wrong. First, I'm not interested in developing an app to sell low CPM banners. Low CPM means you are untargeted--I'm targeted. Second, the vast majorty of people on this site and doing startup do not have boatloads (100s of thousands of pageviews per day) to justify the type of ad buying your are recommending. They don't have the time nor overhead capacity to hire a staff of salespeople. Nor, do they need the headache of managing them.

They would be better of with Adsense.

"As for Demand Media, you're wrong about how banner buys work."

Actually Demand Media does both -- they are using Google Adsense, its lower on the page often integrated into the articles.

> there is no incentive for their Indian quality rater

I don't get what being Indian has to do with this, could you explain?

I doubt that actions of quality raters of non-Indian nationalities would differ if they were paid the same wages.

I've said it before and I'll say it again - Adsense is remnant advertising, it's what you do to fill up the crap space that you can't sell yourself. In every other medium, network advertising is considered an option of last resort.

In the traditional media, advertising sales is a crucial function. A publisher lives or dies on the competence of their ad sales department and their ability to convince advertisers that space in their publication is worth more than just eyeballs.

It's obvious to all of us that a click isn't just a click. There are quality clicks and bad clicks. There is clear value in advertising even with no direct response. It's up to you as a publisher to convince advertisers that your readers are richer or more spendy or more enthusiastic or more easily persuaded than Joe Random.

Print publishers laugh at us for our naïveté. As online publishers, we just don't have the cultural sense that our job is to match up advertisers with people who want to be advertised to. That's the real value-add of ad-supported media, the content itself is just a hook to attract and retain the right demographic.

Ad sales is an active pursuit. If you're waiting for people to find you, you're doing it wrong. If you're expecting advertisers to know what space to buy, you're doing it wrong. If you're not thinking in terms of campaigns and strategy, you're doing it wrong. If you're not pursuing ongoing relationships with agencies, you're doing it wrong.

If you were an advertiser, would you buy advertising from you?

> Ad sales is an active pursuit. If you're waiting for people to find you, you're doing it wrong. If you're expecting advertisers to know what space to buy, you're doing it wrong. If you're not thinking in terms of campaigns and strategy, you're doing it wrong. If you're not pursuing ongoing relationships with agencies, you're doing it wrong.

Ok, sounds fair. How do you go about doing that? I have a site, http://langpop.com that could use some better advertising than AdSense, but I don't have a very good idea about how to track someone down that would pay to advertise on it.

I have some other sites where there's a more direct connection that I could make between who might want to advertise, and thus chase those deals a little bit more easily, but with that, and others that I've done, selling ads just isn't in my repertoir of skills. You've convinced me that either it should be or I should find someone who is good at it... any ideas how? AdSense is very easy to get started with.

> How do you go about doing that?

Check out this video by Gary Vaynerchuck, covered on HN here:

http://news.ycombinator.com/item?id=1467017

It's... impressive.

Go find companies that sell programming related products like IDEs and libraries, go to their corporate site, find the person most like a marketing representative, and pitch your ad space to them.

Hrm. Are there any steps in between that and AdSense? I'm not sure it's a terribly efficient solution:

* I'm not a great sales guy, and the opportunity costs (consulting money) are not insignificant.

* I don't have a good idea of how much to sell space for.

* Other people have to have a better handle on this stuff than I do. If they can make me more money than I would alone, why not split it? Or at the very least, make me more money than AdSense and not make me waste my time trying to track down companies ("Hi, uh, IBM, want to buy some ads?")

The first and last step is to start thinking about the 'true' purpose of your website - as a lead generation tool, a system that converts interest into sales prospects. You're the very top of the sales funnel.

The essential purpose of langpop.com is to establish which languages are most popular. Implicitly, most people coming to your site are trying to work out which language is 'best'. The obvious strategy is to work your website towards answering that question and in the process gaining more information about your visitors and their purchasing intent.

As it stands, the only thing we know about anyone on your site is that they have some vague interest in programming languages. The more concrete our picture of your visitors, the more precisely we can target advertising, the better it will convert and so the more you can charge for it. The Adsense algorithm tries to do this implicitly, but it does it relatively poorly.

For example, the "about the languages" section could be expanded massively, with a (keyword-dense) review of each language and (affiliate) links to books, tutorial material, IDEs, code review tools and so on. Adsense links on these pages should see better bids and better clickthrough, but that's entirely secondary.

Spot on. For an example of a site that takes broad interest in tech and tries its best to funnel it, look at http://builtwith.com , particularly the "build" tab.

They provide a list of affilaite linkbait for each technology you might be interested in. It's not that bad.

The difference between selling your own ads and using a network is the middleman. Google takes a cut, so you potentially end up with less.

If the paid time that would be lost while selling ad space outweighs the cut Google takes, it might not make sense.

I suppose there's also the possibility of hiring a firm that deals with that kind of thing, but I don't know anything about that. It's probably not cost-efficient if your startup isn't big enough.

I was sort of curious about that middle ground. LangPop is not 'my startup', it's a side project type of thing that I'd like to make some extra cash from.

The middle ground is something like iSocket or Federated Media.

We wrote a an entire 7-part series on how to sell ads online. The main points were

   1. Have an outstanding product (i.e. audience) to sell.
   2. Hire a dedicated sales person.
   3. Provide a self service ad management system.
   4. Show detailed traffic statistics.
   5. Sell exclusivity.
   6. Let the market determine the price via an auction.
   7. Simplify the advertising process.

Give it a read - http://www.trafficspaces.com/blog/2009/12/16/how-to-sell-ads...

"That's the real value-add of ad-supported media, the content itself is just a hook to attract and retain the right demographic."

I'd like to point out that this must also be true of the content you yourself are currently reading. HN is a hook to attract and retain potential YC candidates.

the problem with adsense is 3 fold:

a) ad blindness...people have learned to ignore those text links just like they've ignored the banner ads. The only way to actually profit off adsense is to design your entire site around the ad.

b) adblock - more and more people are blocking ads, so you get less and less revenue

c) ad efficiency - more and more advertisers are starting to recognize that the quality of traffic on the display network are crap so they pay MUCH less for it. The CPC on the keyword tool may be $2 per click...but in reality you'll only get 15-20 cents for adsense

So once you pair all of that up...you realize that it takes 30,000 visitors to make $50 a month on adsense.

And yes at scale you can still make a buck...but why would you bother if there are many better options out there?

Sell a single ad space for $100/mo and you double your profit. Sell 3 ad spaces, and you increase it 6 fold.

Or forget ads altogether and go premium. If you get conversions down to 1:200 after you optimize the sales page/price...that 30,000 people turns into 150 sales. So even if you go with a cheap $5/mo subscription option, that turns into $750/mo...recurring. Or if you decide to charge $19.99, you end up with $3K.

The only reason to use adsense is to have something generating revenue until you decide to get serious with your site.

> The only reason to use adsense is to have something generating revenue until you decide to get serious with your site.

I see styleguidance.com still has google ads ;)

and auto-conversion to affiliate links, and some ad sales.

but yeah I could do more with it

AdSense was designed to help Google offload their abundance of keyword inventory. They have millions of people bidding on keywords. Web pages have keywords. So AdSense is a great tool to help them sell what they already have.

Adsense wasn't designed with increasing publisher revenue in mind. Google doesn't operate it as a primary business and the revenue ig make pales in comparison to search advertising. That's why they don't act like any sane business when it comes to customer service. Ads might stop working. They certainly won't give you answers as to why. You may make a lot of money or only a little this month. On the whole, they don't care.

Google has actually changed the name to "Google Display" and a lot of the ads that I see are retargeted. I don't know if this is actually increasing publisher revenues, but I would love to hear first hand about that from the HN community.

By and large, AdSense will work well for you if a high percentage (>60%) of the traffic to your site is based on search queries. It works best as an extension of the search advertising process. If you have a community or user generated content, then you'll need to find another solution.

While adsense was my main route of raman profit, I'm considering to converting to something like fiverr. Creating domains and websites every 2 weeks for a niche is fun/mind numbing/building {research|programming} skills, but if I can just score two gigs a day on fiverr I can typically match/overcome my adsense on 2 sites. Without the hassle of updating, maintaining, and worrying about my {linode|sites|keywords}. There's also the fact that domains go stale after a niche dies off, so, domain renewals etc.

Now if I have the opportunity to snag a domain about the bp oil spill before it gets in mainstream media and snag in $30+ a day... adsense might be my piece of cake again.

You can't completely knock adsense... its basically nonsense advertising. You never contact google, they never contact you (unless you violate the 1 page, 10 sentence ToS). My girlfriend and my uncle both use it and had a harder time setting up a facebook account.

daz.com looks like a fing spam site to me; I've got Markov Chains that make sites that look more honest than that. If you want to make a living making sites like that you've got to spam the fk out of them so you get a lot of money fast, and be careful to make up phony identities so you can keep getting new adsense accounts and not get your new domains burned.

Once you've got a good content gen system, the really interesting question becomes: "how do I create protective coloration so that I can pass site reviews?"

daz.com obviously answers that one poorly

> daz.com looks like a fing spam site to me; I've got Markov Chains that make sites that look more honest than that.

I'm sure the 60,000+ users that put it together over the last couple of years will be appreciative of that comment.

> If you want to make a living making sites like that you've got to spam the fk out of them so you get a lot of money fast, and be careful to make up phony identities so you can keep getting new adsense accounts and not get your new domains burned.

That was not the plan.

> Once you've got a good content gen system, the really interesting question becomes: "how do I create protective coloration so that I can pass site reviews?" > daz.com obviously answers that one poorly

Well, for some reason it passed those 'site reviews' for the first couple of years with flying colours, and then overnight the ads stopped working and there is no way to get an answer out of google why that happened, given a multi-year business relationship that wasn't too much to ask for.

It's their right to disable those ads based on whatever the terms-of-service are, but the least they could do is:

   (1) tell me about it

   (2) do that without being asked

   (3) point out which part of their terms-of-service was violated

Rumour has it that the site was blocked because it contains the word 'hard-core', which refers to a music genre, not to porn, but we've not been able to get that substantiated either.

Well, when I look at the site I see a random jumble of one-liners, non-sequitors and meaningless links. If you got 60,000 people to make that site, you really wasted their time (and yours.)

Just take a look at a page like:

http://daz.com/artists/Glenn%20Miller.html

I mean, the top of that page repeats "trombone 0000 - 0000 delete" hundreds of times... That page just screams "this is trash"

there's no human voice obvious there, no attempt to contextualize or organized anything. I could scrape that stuff off some other sites and make something better looking in a week. Then I'd drop a few million spam links on it, expect to collect adsense for 2 months, then close the bank account.

> I mean, the top of that page repeats "trombone 0000 - 0000 delete" hundreds of times...

It does that 6 times, once for each member of the band. The '0000 - 0000' texts are placeholders for unknown dates, the delete links allow you to remove that entry in case it is mistaken.

All the edits are then piped in to a holding tank where they're verified before being made part of the page. This results in very little attempts to spam the site, on the 'no broken windows' theory, nobody ever made it through.

> there's no human voice obvious there, no attempt to contextualize or organized anything. I could scrape that stuff off some other sites and make something better looking in a week.

I believe you. But it has to start somewhere. Maybe you'd be scraping this one...

> Then I'd drop a few million spam links on it, expect to collect adsense for 2 months, then close the bank account.

I see.

I'm beginning to understand why adsense is the way it is, thank you for enlightening me.

As trollish as the guy you were responding to is, his criticism about the site's design doesn't strike me as totally inaccurate. I would encourage you to tweak it to, as he said, make the 'human voice' more obvious.

That's tricky because the whole idea behind daz is to register the 'graph', so the majority of the data consists of artists, bands, song and album titles, dates and connections between those.

It's like a giant network of musical collaboration, hard to give that a 'voice'. It's major use is to find music by bands that are related through one or more members.

The first things I noticed on daz.com were the words, "share your music". That might have rung the alarm bells somewhere. Just saying.

This article is a misguided gross oversimplificaton of the issue.

I agree that you shouldn't only rely on adsense the rest of your life. But if you are starting with either solid CPM targeted traffic or a lot of traffic or both, then it's a good way to get begin booking some revenue.

Why? If you are a developer you probably don't want to spend your initial days soliciting people for advertising instead of developing your product.

Plus, Adsense can more effectively target customers than the few companies you solicit with your meager resources.

All I know is this: I'm making $50 / day from adsense. That pays my rent in San Francisco.

I'm currently in the process of acquiring direct advertiers, but I'd have to disagree with the author about using adsense as direct income.

I'd like to know more about what other ad networks were appraised, and why they were no good. I'm going to have to rely on ad networks for my own web app, at least for a while, and it would be pretty helpful if there was was decent customer service on the inventory side.

I've looked at: adbrite, adroll, tribalfusion and a bunch of others after asking HN if I should go and make an inventory of advertising networks and how well they perform relative to each other.

None of them comes close to adsense in terms of performance (and adsense absolutely sucks, if you can get $1.00 ECPM out of adsense for a large volume website then you're very very lucky), so even if they would handle customer service better than google (most of them don't) they would still be pretty bad income wise.

You should try to find vertical ad networks, these are the ones getting the higher ECPMs ($5-$20). I run a couple programmer focused networks, but there are networks out there for just about anything. Here are some off the top of my head:

http://www.sportgenic.com/ - sports focused

http://www.glammedia.com - Gender based

http://www.halogennetwork.com - Influentials

http://www.longboardmedia.com/ - Consumers/Shopping

Kudos for not pumping your own stuff, but...what are these programmer focused networks?

The Lounge (.NET developers) - http://theloungenet.com

Ruby Row (Ruby developers) - http://rubyrow.net

Thank you, that's quite helpful!

Is that $1 ECPM for the whole site, or just one ad spot? I mean, for the whole page, which might have up to five google ad spots, or just one spot, or channel as google calls them.

ECPM per channel on large volume typically is 16cts.

So if you have multiple ad spots on a page you might increase that a bit but it's not like two adspots work out to 32 cents, the amount per spot goes down as the number of spots goes up.

I tend to do better than that.

I have some sites that do around 25 cents eCPM, most of the time I wouldn't bother with a eCPM that low... $1.50 or $2.50 is more like what I want.

I've got a project in the pipe that I'm projecting about $2000/month in ad revenue... That's based on an extrapolation from a series of smaller projects covering similar domains. Both my black hat SEO friends, legit "Web 2.0" people and people who are experts in this subject domain would all think I'm crazy, but that's exactly why I can pull this off... Nobody thinks this will work so ranking for this is like stealing candy from a baby.

Unfortunately, the average webmaster is drawn to low-eCPM topics like a moth to a flame... That's their loss.

Again are you speaking of an eCPM of 1.50 per channel or the entire page?

My page eCPM is well above $1, even £1 and per channel it is well above 16 cent and even 16 pence. So I guess it depends on the site.

I gave an average over a much larger number of sites though, some do better, some do worse, it seems to average out to about $0.16 ECPM.

If you don't mind graphical ads instead of text, you may want to look into Advertising.com (AOL). Good experience here (surprisingly high eCPM at times--many times higher than AdSense). Then again, our site performs better than most as far as click-thru rate is concerned, so YMMV.

Thanks, looks like text ads are also available: http://www.advertising.com/publisher/ad-formats

The author's point about Google's support is the key one. Even if your site is generating Google thousands of dollars, it's impossible to get a human if something goes wrong. As an AdSense publisher (and purchaser) I'd gladly pay some monthly premium to be able to contact Google support.

I used to get a £0.20 ($0.31 US) eCPM for 0.33% CTR (http://news.ycombinator.com/item?id=1136710), however, after getting a call from a Google account manager who had some extra inventory to shift for June, we've since been seeing (£0.39) $0.61 eCPM.

Tip: if you have some UK or German traffic, consider Adjug (http://www.adjug.com/) you can set a CPM floor (we set £0.40) and your own backfill. We are setup like this:

Premium Ads (via our agent) > £1 guaranteed backfill > £0.40 backfill (Adjug) > Adsense & House Ads

The only problem is slight delay in tags loading is one of the servers is slow.

Ultimately you should be looking at selling your own adverts anyway (or at least using a genuine advertising agency who can pull in decent targetted adverts), then you won't have problems with Analytics, and you'll also make a hell of a lot more money.

I'm looking at selling my own adverts and I have advertisers wanting to give me their money and copy.

Does anyone know of a system to automate that stuff?

What I'm looking for needs to have:

* Allow advertisers to upload copy and request an advert slot.

* Allow an advert slot to be time-based regardless of number of impressions (1 month in a slot that rotates 10 adverts).

* Send me an email when a slot has been booked so that I review the advert against the end user agreement (that it's not abusive, totally inappropriate).

* That I can approve an advert and then it sends a message to the advertiser with an invoice and requesting payment.

* That the advertiser then pays via Google Checkout, PayPal, etc.

* That after all of that the advert goes live for a scheduled period of time.

* That the advertiser can see impressions, stats and reports letting them know the success of their campaign.

* It should look professional and ideally be themeable/skinnable or self-hosted so that I can make it seamless (the advertisers want to deal with me... I want to reduce my work).

Basically... aside from me approving the content I want a system that hand-holds the advertiser through the whole process and shoves the advert on the site and just gives me the money at the end.

I have 4.5 million page views per month, on a very strong community website.

What I've seen out there that does part of the above doesn't do the whole story... I'm quite willing to pay to get the above.

I'm probably just not googling for the right terms, what should I be looking for?

At the moment I'm leaving money on the table because my time and sanity is more important to me than the money. I realise that this isn't the best business sense and so either I find some tool to help me, or find the time to build one myself.

The self serve ad market has been grossly ignored. A good startup opportunity.

http://buysellads.com does some of it, but is very limited in the options (30 day buy only).

https://www.isocket.com/ promises to do more, and seems like a very promising option. it's in private beta now. So, I haven't tried it.

One thing that is very important to understand when selling ads on your own is how those ads are actually sold. Depending on your site, the niche, and how much time and effort you put into the ad sales process, results can vary greatly. At the end of the day, the only thing that REALLY matters is if ads are sold or not sold. Software alone, in my opinion, won't get the job done. With roughly 2 and a half years of data from running BSA, less than 20% of ad sales come from a publishers website directly - sure, some sites are higher and some are lower, but 20% is the average (and we're in a tech savvy space).

One of Arn's sites, for example, came into BSA with a handful of ad sales. And, once we worked with him and his team on pricing and marketed the site to our existing advertiser base in the iphone niche, his sales soared. I don't know the exact numbers, and I doubt Arn would appreciate me sharing these here - but it's safe to say that his ad sales at least tripled once he joined BSA. In this case, it wasn't the software, it was our ability to sell ads for his site.

The point I'm trying to make: there are a billion ad networks out there. Do some research on them before using them, try some out, and find the ones that sell for YOUR site the best. And, if you can sell ads on your own, do it - hard code them into the site, have people pay you via PayPal, and use a network to supplement those sales.

Full disclosure: I run BuySellAds.com

Would be interested in exploring opportunities to work with you guys on ipodtouchfans.com (site in same space as arn's). My email is username at gmail.

sent you an email a couple days ago, maybe I got the address wrong? :)

if you didn't get anything, can you ping me at todd@buysellads.com so that I make sure I have the right email address?

we would kick ass for ipodtouchfans.com, very nice fit with our existing inventory and advertiser base.

Actually that is a fair point. AppShopper's ads have sold much better through BuySellAds than us doing it direct.

isocket looks to be exactly what I need. I've just signed up to the private beta and if accepted will feed back my experience.

Thanks for pointing it out (I'd never have found it).

I work at isocket - ping me at my HN username at gmail and I'll bump you to the front of the queue.

Thanks alabut, another isocket HN member ( jasonshen ) has already bumped me.

But it's appreciated.

Sweet, glad you're in. And definitely let either of us know if have any feedback. We've even been running occasional batches of usability tests, if you're open to going through the onboarding process with one of us.

Yeah, isocket can do pretty much all of that (disclaimer: I work there), including that it looks professional (disclaimer: I'm a designer).

Techcrunch was our first customer, check out an example of how you can request to buy a leaderboard:

https://www.isocket.com/socket/ZOT76H6IF9

I said "request to buy" instead of just "buy" because being able to approve an ad before it runs is an important part of direct sales (#3 in your wishlist of features). Building in those kinds of checks while improving the overall process has been my main usability challenge in my 6 months or so here.

OIO-Publisher is available as a WordPress plugin or stand-alone and is quite powerful. It has all of the features you've requested except stats for the advertiser.

http://www.dotsauce.com/2008/04/20/oiopublisher-coupon-code/

Check out Open X: http://www.openx.org/

I haven't checked it in detail against your requirements, but at the very least if it doesn't do what you want, it'll give you the terminology to search for things you can spend more money on that do what you want...

Unfortunately the downloadable product seems to be close to dead. There have been serious security problems that have not been addressed for many months. Take a look at the forum to see what I mean. Their server side product must be their focus.

That's a shame. I'd heard good things about it in the past. I believe Google are investing in their admanager product with the doubleclick acquisition, but I haven't heard anything about it recently: https://www.google.com/intl/en_US/dfp/info/welcome.html

Do you find that self service ads pay significantly more than remnant advertising/ad networks? I know that a negotiated direct sale will pay several multiples of what AdSense will, but do the self service ads come close to that?

With adsense I was making a very poor return as click-through rates were quite low (it's a discussion forum).

However I've got some advertisers really pushing to get in front of the audience for page impressions and exposure rather than click-thru.

Basically the market I have is niche, and represents the largest site on a given subculture (fixed gear cycling). This subculture happens to be very cool at the moment so even beyond those wishing to advertise bike products to the community I've been approached by EMI Music, Adidas, Puma, etc all to put adverts on the site.

I was going to just operate an intra-community rate to help support the cottage industries (sole traders who make custom courier bags for example) at a low rate. But now I've got very large brands wanting to put adverts on the site it's beyond the amount I can control, yet they are perfectly happy paying full market rate to get the brand exposure into this hard to reach niche crowd.

So basically... for me, for my site, self-serving is incredibly lucrative.

It's just that I have a full-time job, am doing an MSc in the evening, run several web-sites and am trying to keep a social life intact and a partner happy. So whilst self-serving is lucrative my trade-off is that time for my is very expensive. So I need as near to full automation as possible to take all this lovely money off the table and to put it into the community coffers.

I'm estimating it's worth at a low-end GBP 500 per month, and at an upper end GBP 4k+ per month if all of the slots are purchased by large brands. So yes, it's way in excess of what AdSense click-throughs will generate, and all because those who want to advertise on my site value exposure and eyeballs more than click-throughs... we're just hard to reach.

why not hire a salesperson and pay via commission?

Because I don't need a salesperson, I need an administrator. And what needs administrating is something that can be done easily by some software... so why demean someone by making them do a highly repetitive and dull task?

I am working on a SaaS ad server, it has much of what you want but doesn't have full self-checkout and self-service just yet. It does reduce the rest of the work though so if you are interested drop me a line (in my profile).

There is a time cost for this though, unless you reach a certain size its not really worth the time organising it when you can just slap adsense or the like on there.

Couple of things I've noticed about adsense - have no idea if they apply here, but maybe it hasn't been considered.

If the colour of the links and whatnot are two similar to what you are using on the site (particularly if they are positioned to look much like links) the ads can fail to appear. This has happened to me a few times and I've changed the colour to have them up and working again within a few minutes.

Another thing - I've also had pages fail to display adsense where the content is exactly the same as another page serving adsense (botched url structure in my case).

I assume you've tested these things - but figured it may be worth mentioning anyway. You never know.

"And today, it happened again. This time on reocities.com, my rescue attempt for geocities where it helped to pay for the hosting. After serving up the ads for a while suddenly I'm shown a nice blank rectangle where the ads were before. No explanation"

Uh... maybe the fact that it's a bunch of stolen content?

To the OP, I realise you're trying to do a noble thing, but that content is not public domain. Trying to preserve it is cool. Trying to profit off it is not so cool. Bit of a grey area and it's easy to just turn you off than get into some heated, obviously biased discussion with the site owner.

> Trying to profit off it is not so cool.

I don't think that 'profit' means the same to you as it does to me. At adsense income levels reocities will never be profitable, and I have a couple of thousand thank you notes collected by now that prove that on the whole people are pretty happy that I did what I did.

People that request removal of their pages have it backed up now (I do that for whoever asks to create a zip file of all their content), and those that don't want their old stuff online any more because it is either inaccurate or because they can't update it (reocities is 'read only').

In some cases they want it removed because they put private info online back when everybody did that or because they are now professional designers and don't want their geocities stuff to haunt them.

And, of course, whoever asks has their content removed, once every month or so I filter my inbox on 'reocities' and 'geocities' to process the removal requests.

It's still a manual process, that's how little it gets requested, but I probably should automate it somehow.

If in the long run reocities contains no pages and all the stuff has been moved to the proper owners again then I'll be more than happy, that would be the best outcome.

Having some adsense ads in there was a way to offset some (but not all by a long shot) of the hosting costs.

"The data on this site was bootstrapped using the freedb.org data set (yes, our database will remain GPL'd, no strings attached)'

What version of the GPL are you using, where is the (database)download link and a copy of the GPL? It is an oversimplification to say no strings attached.

Unless the content is all already GPLed, I think there is a problem with you trying to assign a license to work you didn't create. You certainly can't say all rights reserved (restrict rights) if you're dealing with GPLed work of others. (what's free must remain free) I see some band photos. I seriously doubt the the rights holders released them under the GPL.

The banner "share your music" and taking MP3 uploads seems like a serious issue too. As does distributing 30 second clips. iTunes and music stores that do that have agreements allowing it. Do you?

And posting peoples geocities sites without their permission? It isn't your content. A notice saying you'll taking it down doesn't make it okay. Just because they haven't discovered your site and had a note survive that mail you check once a month... Right holders should not have to comb the net looking for abuses. Even the many that may like the idea may not know to update their content. (I'd expect you'd also have trouble authenticating them)

You really need to contact every former page owner and have their advance permission (and give them editing rights along with some sort of terms of service)

A bunch of thanks yous are great, but you'd get that for MP3s too. It doesn't mean there's not a problem.

Good luck sorting it all out in a proper way.

Why don't you spend an hour or two and make it read/write ?

Because I'd be flooded with spam.

That's not necessarily true. I agree with the other guy that you are really just scraping content and trying to profit off it. If people really wanted to save their content, they could have done that. Most of what you have is probably abandoned.

> I agree with the other guy that you are really just scraping content and trying to profit off it.

You can agree with him as much as you want, it just isn't true.

> If people really wanted to save their content, they could have done that.

Part of the problem here was that yahoo did a pretty bad job at communicating their intention to the people that had their sites there. Essentially Yahoo tried to convert the 'holdouts' from the free portion of the site to paid hosting accounts 'or else'. When they pulled the plug it left 10's of millions of accounts up in the air.

There were several attempts at rescuing as much of geocities as could be done, the fact that you don't know any of this isn't really your fault, after all your account is all of 40 minutes old, but when it happened it was fairly widely publicised both here on HN and in the regular media.

> Most of what you have is probably abandoned.

You'd be surprised.

Adsense was amazing for the first year, but after it became mainstream, the rates fell drastically.

If your website has a strong brand and design (i.e. isn't a blog) then there's lots of alternatives that pay better.

> there's lots of alternatives that pay better.

Such as ?

It depends on the topic that your website specialises in. Determine that, then search for ad networks and rep firms that represent those areas. If you find the right service and you are well know enough within the niche, they will personally pitch your website to their clients.

Good catch, thank you! (fixed)

Why put a date on the copyright at all?

A few minutes of Googling:

http://forums.digitalpoint.com/showthread.php?t=48127#post53...

My father makes about $400.00 a month from adsense. He's been working with it for at least a year, and he'll probably be making five as much as that in another year.

Reply on news.ycombinator.com