Facebook Developers

AM I GETTING RIPPED OFF?


mmmm, probably as all these ad networks seem a waste of time whether a advertiser or publisher. Maybe it's time to look at other means of generating revenue from apps from outside sponsors who are looking for short to long term unlimited exposure and clicks. We have looked into this and it seems by far the best way to go and that way you have control!

Great post Dennis.  My favorite affiliate networks are neverblueads, maxbounty and marketleverage.  I've had a pretty good experience with them, although I'm not running the ads on my app.

George,

You make some great points.  The brutal reality is that if your goal is to maximize earnings, then CPA networks are the only way to go right now.  There is a lot of talk about how social advertising is about deeper engagement, truly leveraging your relationships, and somehow showing more targeted ads because of the amount of information we have on you.  While there are a few examples of such ads that do generate decent CTR, I do not know of ONE that has actually monetized well-- even at a quarter the performance of CPA offers.  Here's why:

Most of the traffic on social networks is female teenagers.  They don't have money, don't want to spend money, and aren't looking to get credit cards so that they can spend money.  So that eliminates all ecommerce ads for travel, real estate, gifts, clothing, and so forth.  Facebook apps weer born to be free, so an ad that urges you to give a gift is implying that the gift is free.  For Valentine's Day, I ran over a hundred million impressions to actual flower and gift sites.  We got maybe a few sales off over a few hundred thousand clicks.   Same results for Black Friday-- the big shopping day after Thanksgiving.  Tons of clicks, no sales.  Same for itunes--- great ads that promoted songs that users would like based on their profile information.  eCPM of less than a penny.  In other words, less than a penny per thousand impressions served.  If you are an affiliate marketer doing PPC, then you know that conversion rates should be in the 3 to 7 percent range for commerce offers.  I ran credit card and financial offers--- Chase Platinum card for students, payday loans, automobile insurance--- conversion rates were a fraction of what you'd expect from traffic from any other source.

I'm not saying that advertising on social networks is no good.  Rather, right now, there is a lot of hype from folks that want to sell the concept, as opposed to dealing with the harsh realities.  The reality being--- at the end of the line, someone has to pull out their wallet and pay for something.  You might buy impressions to to promote your dieting app-- that app then shows ads to users promoting diet tea, those users will tell their friends about the tea (and their progress in losing weight) to bring in more users, a certain proportion of these users will buy diet tea at a certain cost.  So whether you buy inventory directly to show diet tea ads-- or try to create an app that generates more traffic, and then show ads, the economics still rely upon a purchase at the end of the chain.  If no purchase-- it's a ponzi (pyramid) scheme. For example, many of the free gift apps generate tons of traffic, as teenagers enjoy giving each other virtual gifts-- which are basically just sending image files to each other. But will that person who gave a virtual flower actually shell out $50 to buy real flowers for their boyfriend?  It's akin to Youtube's problem-- lots of usage, but little money being made.  Usage and profits are not the same thing.

So if teenagers don't have credit and don't want to spend money, how does an advertiser get a return on his advertising budget?  There are 3 methods: get them to pay via their phone bill (premium text messaging services), get them to download software (for example, the webfetti toolbar), or get them to volunteer private information (their email, phone, address, and preferences).  We can go into the mechanics of how each of these three areas work (post a reply here, if you'd like me to break it down)-- but suffice it to say, the advertiser is looking to make back more than what he spent on ads.  For example, when users download the webfetti toolbar to "pimp their profile", that toolbar shows ads from the ask.com search engine.  We know that users perform X searches per month and keep the toolbar installed for Y months on average, and that each search generates Z cents in clicks from ads shown alongside search results.  The product of XYZ must exceed the average cost to acquire a new download.  So if the toolbar user is worth $4.50, then IAC (the makers of Webfetti) might pay out $3 to the ad network, upon which the publisher (you), might get $1.50.  If 1 out of every 15 clicks results in a toolbar install, then you are earning 10 cents a click.  And if the CTR on your ads is 0.7%, then your eCPM is 70 cents.

There are but a few networks who understand the mechanics of these offers and have the relationships with these 3 types of advertisers to maximize eCPM's on social network traffic.  In the case of CPA Empire and Flingweb-- these are advertisers themselves that have decided to cut out the ad network and run directly with publishers.  That's not 100% true, for folks who will nitpick this statement-- since affiliate networks do share offers between one another to mix good and bad traffic and scratch each others' backs-- but for the most part, it's accurate. 

Zerostar07-- you can manage the advertising yourself-- you'll need an ad server and pick and choose offers that work.  I'd say that if you have at least 10 million impressions a day, then it would make financial sense for you to do that, since it's enough money to matter and you'd have enough volume to negotiate with the advertisers.  If you don't have the volume, then you're at the mercy of the aggregators that will take their cut.  At that point, it's up to you to determine who is best at monetizing your traffic versus what their cut is.  There are dozens of affiliate networks, but the majority of them are just resellers on top of other affiliate networks, who in turn have direct relationships with advertisers.

Someone asked if I started Flingweb-- it's not my company.  I am pointing out that you as a publisher should learn about how the money flows, so that you can maximize your hard-earned traffic.  I happen to believe that their payouts are higher than the other networks.  I'll tell you why and also give you the fine print.  They have the least overhead of any of the other major affiliate networks (of those that have their own offers).  There are a bunch of other small affiliate networks, but they are just reselling offers and taking a cut.  Flingweb is giving you mostly their offers, as they are the underlying advertiser.  There are only a couple folks larger than them in social network advertising.  They are by far the #1 advertiser on SocialMedia--- see for yourself who is running ads on Social Media pubs.  I've seen all the other ads that are running across Facebook apps, whether on SocialMedia or otherwise, and know that their offers outperform anyone else's. So a higher payout on a better offer is more publishers earnings.  If you are a growing ad network and want to become #1, then you will naturally be willing to take a lower margin to attract more pubs.  Of course, it can be said that they would increase their margins once they become the dominant player.  And that's why I say to always rotate your traffic-- with perhaps 80% going to your primary and 10% to each of 2 backups for insurance.

So why would you not want to run with a Flingweb, CPAEmpire, CXdigital, Azoogle, Neverblue, or another?  The offers are kind of spammy.  Toolbar downloads can be good or bad.  For example, the Webfetti toolbar is a legitimate product that is actually useful.  The Zango toolbar is spyware that is impossible for you to remove, except by formatting your drive.  Mobile offers are good and bad.  There are "crush" offers that trick you into thinking that one of your friends has a crush on you-- using messaging such as -- Inbox (5), You have (5) unread messages.  And there are not so bad (but not great) mobile offers for online dating and horoscopes.  Ringtones are another example--- they aren't really "free" (since when is anything actually free?) and you can get trapped into getting billed $20 a month if you don't read the fine print in 4 point font below the bottom of the page.

If you are interested in maximizing revenue and cast a jaded eye towards advertising general, then run with one of the networks mentioned here or elsewhere.  If you want to be squeaky clean, then completely avoid all ad networks.  That includes AdSense, where affiliates who are buying traffic from affiliate networks are also buying through Google.  Google monetization, since someone asked, is typically 50% poorer because they add another layer of middleman.  The advertiser brokers to the ad network, who then brokers to the affiliate, who then run through Google.  Then Google takes a 30% cut (+/- depending on what you negotiate).  And you get what's left.  The money has to pass through 5 hands-- if that's ever an example of the "leaky bucket" for you.

I know this has been a long post.  If you have any questions, please post them here and I'll keep answering them.  As for why I'm doing this?  I don't run the Social Media ad network any more.  I've walked away from the industry because of the very things I've explained above.  Facebook contacted me several times about the nature of the ads that we ran-- it doesn't feel good at the end of the day to participate in this kind of advertising.  It's like making sausage-- great to eat, but don't ever watch how it's actually made.  This is my gift to all publishers who really wanted to get answers to the monetization question, but where I was not allowed to provide an honest answer until now. 

You can see in my signature line that I'm an analyst at BlitzLocal.com, which is online marketing for local service businesses--- doctors, dentists, therapists, and other folks who sell a similar service, differing only by geography.  Knowledge of running an ad network here does apply, strangely enough, since you're still calculating ROI on advertising dollars.  But instead of promoting webfetti toolbars and ringtones, we're helping folks who have a cavity find the local dentists.  Completely different product/service, but underlying traffic management being similar.  So I'll happily cheer publishers from the sidelines-- and if some of you make money because of my shared experience, I would definitely like to know.

Last edited by dennisyu (2008-06-15 22:27:06)

andy wrote:

I think Facebook really ought to consider adding a revenue sharing option for developers who choose to disable the ads on their apps.

Or start their own ad network based off the same technology that runs theirs.  Not sure why they haven't done it.. this seems like an easy way to boost their revenue.  The system is already in place.

Dennis, sorry if I you mentioned it but I did not pay attention... but what kind of products does Flingweb advertiser for? Is it the same spammy products you mentioned (i.e. ringtones, freebies, toolbars, etc)?

Great to read your posts BTW. Keep up the good work.

Hi Jasek, and others,

I'm Adam Foroughi from Flingweb and work with a partner of mine to head up our network operations. As way of background, we've been one of the largest advertisers on social networks the past couple of years and have finally launched our own network as of 3 weeks ago! I know Dennis from his past businesses and he is mentioning us because he knows how good we are as an advertiser and it sounds like he's interested in pointing developers directly to the advertisers to cut out as many middle men as possible. We buy lots of our advertising direct, but also do lots through networks and our hope with our own network is to get to you developers directly.

We offer a new, proprietary serving technology that we'll be optimizing over the next several weeks, weekly payment, and 80% of our inventory is filled with our own in house offers. We have a variety of products but have strict policies against any advertiser running any offer creatives that are deemed to be deceptive in nature. We are already live on a few applications and are doing quite well already and the numbers will only get better as we optimize our systems and our creative.

Please email me at adam.foroughi@flingweb.com to get a test live.

-Adam

Very interesting post Dennis. I would like to shed some light as the CEO of AdParlor.com.

I agree with pretty much everything that was mentioned earlier. Yes, if is hard to get clicks on ads for external sites such as credit cards and travel, and Yes, better a CTR is achieved by using 'spammy' type advertisements. In fact, there is always an active discussion around here on what types of ads we should sign on to our network.

Thus far, we have been quite successful at delivering high payouts for our publishers under the following two internal guidelines:

1) ALL ad sales are purchased from us directly by the advertiser, cutting out the middlemen.

2) A MAJORITY of our ads are for other facebook applications (or in the case of myspace, other myspace applications). This has actually turned out to work quite well. It is a lot more compelling for a facebook user to click on an ad to install another facebook application than to click on an ad that takes them to an external site. Additionally, we have all advertisers and publishers tag their respective ads and applications and we try to serve up an ad for a related application. For example, someone taking the 'See how Kissable I am Quiz' would get an ad for the 'Kiss Me' application.

I hope this sheds a bit of light on the considerations we, as ad network administrators, think about all the time.

Thanks,
Hussein Fazal
www.adparlor.com

Nathaniel,

You are spot on about redirection, where there are multiple middlemen taking a cut.  Running ads for other Facebook apps does add another step to the monetization process, since that advertiser has to make money somewhere down the line, by in turn running an ad.  Let's look at the economics of this:

ARE ADS FOR FACEBOOK APPS JUST A PYRAMID SCHEME?
If you're buying clicks for 10 cents each, but they are worth only 8 cents on average (based on the lifetime value of all ads a user clicks on), then you're upside down.  If, however, you buy at 8 cents a click and that user is worth 10 cents, on average, then you're making money.  So the short answer is "not necessarily".  However the majority of folks who buy app installs are bleeding money, may or may not realize it, or might not even care.  Some companies do things for reasons other than maximizing profit-- they have a budget to spend (use it or lose it) or just want to grow their brand, metrics be damned.  The key to making it work is ....

THE VIRAL FACTOR
You've heard the story in many incarnations-- you install an app and then tell 10 friends about it, who then install and who each then tell ten of their friends.  Pretty soon, you have 3 million installs. And the app has hit the "tipping point", where it takes a life of its own, as install activities are posted to the friend feeds and drive yet more traffic.  All you need to do, goes the common thinking, is to make the viral factor work in your favor-- keep the viral factor ahead of the decay factor, so that more water gets poured into the bucket than leaks out.

THE UNSPOKEN LIE
So all you need to get your app to take off is to initially buy enough installs to "jump start" your user base to "critical mass".  The the app has the boost necessary to become viral.  Completely false.  If your app is viral, then it will take off on its own--- if not, then no amount of advertising can stave off decline.  I have seen many ad campaigns and have yet to see this proven false.  Your viral factor will significantly be above your attrition rate or significantly below. To be even is like flipping a coin and having it land on edge, instead of heads or tails.  Net net: If your app sucks, no amount of advertising will help it.  If it's great, it will take off naturally.

IT'S LATE  IN THE GAME
Remember back in June when you could build a simple app and grow 100k installs a day?  Facebook now has "app fatigue"-- just look at how many apps are on my profile page-- and they also ban forced installs.  So apps now have to have significant entertainment value and complexity of interaction.  Pong won't meet the bar now-- it's got to be more like GTA4.  Same is true of advertising-- static, run-of-network ads blasted to everyone worked at the F8 launch.  Remember all the webfetti ads?  Now that this has burnt out, the ads have to take advantage of things like dynamic insertion, where you can insert name and profile picture directly into the ad.  The bar for advertising is growing higher (not by that much, but it's going up), so advertisers and ad networks have to target better and take advantage of what they know about the user through the FQL calls.

AD PRICES ARE NOT LIKE OIL PRICES
There is still a vast amount of ad inventory, so we'll continue to see plenty of cheap traffic.  Just look at how MySpace is still unable to get high CPM's for their traffic and look at how much unsold inventory there is on Facebook.  But I do see ad prices creeping up as big brands (who often have large, unaccountable budgets) and affiliate networks come in and gobble up the higher quality US/UK/AU/CAN traffic.

CONCLUSION
So how does that bode for appvertisers (folks who buy ads to promote their apps)?  You have increasing costs per click, but are partly offset by monetizing better.  Yet, the net result is that the math rarely works out for folks who buy their traffic, instead of trying to grow it virally.

heres my biggest concern, when one was at social media/appsholics, there was nary a whimper of compliaint, now one is at another company, all of a sudden, another option seems compelling. theres a conflict of interest issue here, and you cannot treat all devs as so effing naive,

Yeldarb wrote:

You can more than make up your investment of the $500 on ads with the extra users you have and it gives you a much bigger head start on other people looking to emulate your good idea.

but the initial spend required for an app to jumpstart -if it's truly viral- is negligible. I 'm more with dennis on this one, applications that are truly viral can grow on their own even with $10 in advertising. the problem here is that facebook is a saturated viral platform. Try launching an app on bebo to see how fast it grows. it's also other factors, like the time of launch that matter [ like hatchlings, that took off as an Easter app ]

Best post ever.  Well done.

Someone sticky this.

Oldie but goodie thread, worth while bumping. One of the first I really remember reading when I first came here.

CB
AVPI

Last edited by Lethos (2008-12-22 10:23:07)