I created two FB mobile advertisements to direct traffic to the website, though the website is more eCommerce/service than any type of sign up. Budget $50 over 3 days reach was ~20,000+; the click through rate was .5% and .4% for the 2 ads; just under 100 clicks to the website with none resulting in conversion.
More disturbing was the fan page promotion through FB (paid "Likes" in my own words). $10 budget per day over 3 days; reach = 3,000+; total likes 34. What disturbed me though was when I would go to the profile page of the users who "liked" the fan page as a result of the promotion, many of the user profiles did not appear to be legit. Moreover, the majority of these users who liked the page had a single facebook post in their entire facebook timeline. As unlikely as it is that of ~30 paid likes nearly all were were inactive facebook users who were otherwise compelled to interact with my paid promotion, it is equally unlikely that facebook would be so brazen in committing fraud on advertisers by creating and managing fake accounts to click paid promotion/ads which could easily be proven. Nevertheless is begs the question what are these accounts (fake, bots, ect...) and who controls them and why?
It was posted on HN a while back. The gist is that there's a high chance that your ad was clicked by click farms, even though you bought it legitimately from facebook (in an effort to fool clickfarm detection, apparently).
What's even more interesting is that not only are you wasting money buying the ads, you also end up with less legitimate "likers" seeing your content because of the way facebook propagates new posts. So you're actually paying to get negative results.
If you pay for 500 likes, and 500 different accounts are created to like your page its obvious your page is gaming the system. Likewise, if you have a voting ring of 500, its easy to see what pages those 500 accounts are liking. However, if you have a voting ring that votes randomly while ensuring anyone that pays gets 500 of the votes then its much harder to find out who is gaming the system.
But you're right the strategy is not to raise suspicion is to click the targeted ads and like buttons among other random ads and likes.
If I was to learn that facebook was paying those click farms to do the clicking I wouldn't be much surprised.
Our example (country specific mobile app for doctors), spent 100 € on AdWords, end result was literally 0 app installs, 0 sign-ups, 0 everything. Medical keywords are expensive, no chance of sending them directly to the App Store/Play Store (that we saw at least), and no other useful targetting.
Here come Facebook mobile install ads. 40 € spent so far, 500+ app installs, 200+ sign-ups, great retention. We can roughly target medical professionals, take them directly to the app stores, and the clicks are cheap as hell.
I have no doubt that AdWords work much better in other cases, and that FB can be useless, but it's not black and white, you need to know which tool fits the purpose.
If this is actually true you should be doing affiliate installs which routinely pay out $1 with no (or high) caps. You could literally make tens of thousands of dollars daily.
For example, with less than 50 €, the ad has already been seen by around 15 % of the total audience. No matter how much we throw in, it's going to be going slower and slower from now on. But it is enough for us to get seeded among a few hundred medical professionals, and then rely on word of mouth to carry us forward.
Edit: It's also a free app, just to make that clear.
http://bigosaur.com/blog/28-facebook-ads-android-players
You need to use the targeting options, otherwise you're wasting money.
And as another handy tool, you can use the FB ads api without authentication to perform certain types of search: http://graph.facebook.com/search?type=adworkplace&q=hospital... and build your targeting criteria
Hope it helps :)
For workplace targetting, it looks like none of the hospitals in our current markets (Balkans) are in the system, so it's not as useful. We had a similar problem when trying to target people with "Medical School" as education and got something like 10 people total, seems education as an attribute is counted only for US-based users.
-Page owners have the ability to restrict access by country.
This means it's literally just a couple clicks, and you can completely disable the "click-farm" countries from seeing your page (+ advertisements).
http://sendgrowth.com/blog/simple-defense-facebook-click-far... (short post about it, shameless plug) http://i.imgur.com/snkv77Q.png (screenshot of facebook settings page)
Think of it like a computer program. If 99% of the program is right but one thing is broken, the entire thing won't work. Marketing is, in a lot of respects, the same way. You can be missing one single variable and your entire campaign falls apart.
Look at all of the variables in this campaign - title, image, targeting options, whether you do sidebar ads, newsfeed ads, or mobile newsfeed, and most importantly the product/service offered on the other side (not to mention the conversion rate of the specific landing pages). Apparently this campaign wasn't profitable, but I run a half dozen profitable campaigns on Facebook at any given time (most of them CPC), and I know people who spend $10,000/day on Facebook ads.
Facebook ads do work under the right circumstances. Concluding that they don't after one try is a little absurd.
"But it definitely doesn’t seem like you are getting what you pay for, and certainly the value, at least for my site is not there."
I've run a few campaigns and always managed to break even. I believe the key to success is very selective, and often slightly obscure, targeting.
The only thing you should consider to get to any conclusion is the overall CPA.
Facebook is a different beast from AdWords, you cannot simply target people searching for "Online budgeting tool". You have to find a good demographic and that requires lot's of testing. From that point of view $50 is actually meaningless.
Yeah ... he can say that Facebook is charging him for fake clicks ... that's an old news :)
I've spent mid six-figures on Facebook CPC ads over the last several years and can definitively say that they work very, very well - depending on your use case. Mine is not the OP's use case (though I've sold a metric a-ton of SaaS on FB).
I advise everyone here thinking about FB ads to do the following:
- If you try it, dedicate a serious amount of money. Nothing less than $500 will suffice as you need to get statistically significant data across all your targeting sets.
- Focus very narrowly on your target market. Trying women age 22-29? Do that in your metro area only. Keep your targeting sets small so you have fewer variables to contend with.
- Don't lose your nerve. If you give up too quickly you'll know nothing.
Finally, I do understand the OP's frustration with click numbers from FB vs. GA. Don't let it get you down, as this is common on every platform. Optimize for your actual logged data and you'll profit.
With every platform you're going to have some issues with click fraud. Ideally you figure out how to make your campaign work despite these problems. You focus on quality conversions that make a dent in your business. I agree that it sucks to have to pay for this bad traffic. If you spend enough or get an account rep, I'm sure they'd take a look at the report you put together. Its comprehensive enough that other sources I've worked with would probably credit you some traffic.
In the past, I've had some spend returned due to this issue from FB but its been quite awhile since i've even requested it.
For most Saas companies, retargeting and search will work best as they are solving a very specific problem/need. Its difficult to find the users with intent that drives them to convert to a paying customer.
You can target only desktop users, if you want to send people to a website
I don't know a single user who enjoy being sent to a mobile app instead of the website.
If being hated by your users is not enough of a deterrent, then factor in the cost of building said apps on mobile platforms and maintaining them through time and new smartphones versions while having a website versus simply having a website.
Unless you're a big corp with tons of money to throw out the windows, the wise path is a well thought and designed website.
I don't see why you would be hated by your users? The ad makes it super clear that it's an app, it takes you straight to the app store where you can read reviews / description / screenshots and then you just install it.
I think you are making the mistake of thinking that the average user is like you, most people will just want to get the app ASAP when they see an ad for something that seems useful; not having to navigate a website which doesn't have a mobile version (try budgetsimple.com in an android phone, it's impossible to use !)
Do people really install random apps because they saw a one-sentence advertisement on facebook? I really don't know what normal behavior is for the average facebook user.
FB do let you set a Desktop Only audience for ads. You need to use Power Editor (Google Chrome only) and select Desktop under Placements.
I'd like to see a re-run with Desktop targeting only.
Edit: https://developers.facebook.com/docs/reference/ads-api/targe...
I'd love to see one of these experiments that isn't trying to get website signups / sales which is something FB is notoriously bad for.
I spent a few years at an SPMD, and I would really advise against anyone who doesn't know exactly what they're doing to use the FB self-serve anyway...
I don't know what Facebook's long term business model is. IMO, this isn't it.
That's true. Since people spend a lot of time using Facebook (i.e. as a "third place" with no specific search intent), I would guess that brand advertising, brand-awareness marketing, and especially targeted ad remarketing would be their sweet spots.
This campaign failed after a few days and now it will go away. That's as it should be.
They're in a tough spot. But they should at least start to turn the ship in the right direction before their total ad business collapses as "ineffective".
While you can self-serve advertising, it is not necessarily a good idea, in the same way that representing yourself in court is not necessarily a good idea.
Facebook cares much less about fraud than Google does, because FB has been under much less external pressure from shareholders to do it. That is not to say that various Google properties do not have fraud issues still. This is reflected in the price differentials.
After all, entire IPOs built around Adsense fraud occurred in the mid-2000s. There have been countless small businesses built around link fraud. It is quite likely that some of the major media names built on social traffic are also based in part upon defrauding social advertisers, because as of yet, few have cared about it, and many investors will just reward companies based on trivially faked traffic metrics.
But guess what? Circulation fraud is a problem that has been with us for over a century in media. Some combination of the price system, auditing, direct response ad testing, corporate incompetence, the good ol' boy network, and other methods have kept it from making advertising either totally useless or totally risk free.
Despite this, here are some issues that could help you advertise better in the future:
1. This is not a good ad. The copy is bad. The illustration is bad. The call to action is unwieldy. The logo placement is haphazard. The headline is Wrong. The human figure is in the wrong position. The button placement is haphazard. You would be better off plagiarizing ads from Mint and swapping out the logos and colors. If you want to keep the lady accountant mascot, put her to the left of whatever copy you want the visitor to read, and make her look at it.
2. The demographics you selected might as well have been at random. Market research is not throwing a dart at the entire planet and targeting whatever the dart landed on.
3. FB != Adwords in the same way that a newspaper != the yellow pages != a niche interest magazine != radio != flyers != e-mail spam != direct mail != a catalog and so on and so on and so on.
4. Your budget is so small that it barely qualifies as a test campaign. You ran a test campaign and discovered a hazard to avoid. That is the point of the early tests. If you run out of budget before you can discover a profitable marketing strategy, your tests will uncover that you are out of business.
In this case, you are dazzling yourself with your measurements because it is easier for you to do so than it is to think at a higher level about your objectives and the methods that you want to use to achieve them given your resources. You could call this Silicon Valley Degenerative Metrics Dementia. Sadly, there is no known cure for SVDMD.
I could personally care less if Facebook goes out of business, but as long as real people with wallets continue to use it, it will have some utility to advertisers, so long as they put forth at least some good faith effort to control their bot/fraud/misclick problems.
Considering some of the things that I have seen with Facebook, I am not confident that they really care, because many investors will reward them when they count bot users (or human users living in third world conditions) as if they were humans with first world bankrolls. There is no comparable Matt Cutts figure for Facebook. I think the real money on the platform, like was the case with Google for a long time, is on the criminal side.
Hopefully some short sellers are paying attention to these stories, because terror is the only thing that will induce Facebook to stop its absurd gyrations on the product side and actually police their platform. Short sellers can orchestrate a PR campaign and either pressure Facebook to start caring or can just make a lot of money by torpedoing the firm through aggressively publicizing its failures.
All that being said, I hope that this is helpful to you, and I am glad that more businesses are learning that online advertising is difficult, complex, and risky (like advertising everywhere and always in all mediums over all time periods using all sorts of technologies).
The fact that fraud exists is huge, and that it runs so unmitigated on Facebook should be subject to class-action lawsuits. That you should work to avoid click-farms, to avoid Facebook from profiting from click-farms, is unacceptable.
FB is in a perverse incentive position to not fix something that makes profit for them, much like any other running scam. And the size of this beast is unknown, because its not recognized.
I'm not the OP, but WRT point number one, what in your mind would qualify as better copy, and better a better call to action, and logo?
It's just in a haphazard location. Humans read left to right and are attracted to faces. The logo is off there in the lower left corner. The lady is to the right of the copy. If this ad were in Hebrew, that would be fine, because that language reads right to left. English goes left to right.
'Better' copy is what tests better with the market. I do not know what is better. I have inklings on what direction to move in to improve the campaign.
I would test the current image copy vs. 'Join the 100,000+ people like you who have saved money with BudgetSimple.'
Body copy test: You can become financially secure using our free budget planning service.
Button test: 'I want my free budget' or 'Teach me to save money' -- 'learn more' is too generic.
Like most ads for tech of most kinds this is a little too heavy on the features and not enough on the 'what's in it for me?' No one cares about 'getting a better understanding of their finances' except accountants. No one gets excited about budgets except for politicians eager to spend money that belongs to other people.
Individuals want more money. They want safety. They want financial security. They want control over their own life.
If the OP wants to run a female-targed campaign, it'd be a great opportunity to pose as the authority and quote a magazine article about the problems that some women have with credit card debt or general financial planning, and reference that in the ad. This just comes to mind because a friend of mine wrote a feature for Cosmo on this topic a couple years ago. Ads that do not come off as pushy and instead present themselves as trustworthy, useful information can win over the trust of the target audience better.
If it is really just Oz, then Oz up the ad. Australians would care more about joining 20,000 other Australians just like them than 100,000 Yanks, for example.
There is a dude who runs a budgeting business off of Something Awful banner ads that I remember from a few years back that uses goon jargon to establish rapport with the target audience. He succeeded by matching the tone and content of his message to the readers in the right context.
If FB's traffic is almost or even largely from mobile devices, paying to show ads for a non-mobile site to that traffic seems just silly. The site is downright hostile to mobile users; the text loads last, it starts with a video and a worthless image, and the actual text ping-pongs across the page to accommodate the clip art and screenshots.
Given this exact same data, the OP could spend a week making at least his landing page mobile, run another FB ad, and make a blog post about A/B testing your landing page for mobile users. But no, it's all Facebook's fault, because bashing Facebook will always, 100% get you upvotes on this site...
I find it hard to believe that Facebook is not capable of separating mobile (smartphones, wearables and tablets) and desktop traffic in a reliable fashion. I don't know if that's intentional, or why it takes so long to address this problem. But it's hard to deny that right now the only beneficiary of this foggy situation is Facebook.
Two comments:
1. This media is sold in an auction. If the quality of the traffic vs what you pay for it is bad value then the bids are set too high. If I pay over the odds for something on ebay it isn't just ebay that is at fault.
2. Doing online advertising well is harder than Facebook and Google are incentivised to make clear. In some cases this stuff is very hard which is why there are people whose full time job it is to get it right.
As someone with some expertise in biddable media reading posts like this must be like a coder reading about how a programming language is flawed because the Todo app scaffolding doesn't quite do what the author expects.
As the adverts are sold in an auction, shouldn't the cost per click reflect the proportion of fraud?
E.g. if value of true click is $1 and 50% of clicks are fraudulent then CPC should surely approach $0.50?
(Sorry, this is a little off topic.)
Can the OP or someone else fill me in on how he was able to target people who like other pages (that he doesn't own)? Is it through lookalike audience or is there a more direct way to do it? I've been trying to do the same (target similar pages) but I'm clueless as to how to do it.
http://www.quicksprout.com/2014/02/19/how-to-steal-your-comp...
And try different ad text. Acknowledge that this is a different platform than search and you need to advertise differently. Don't be so quick to dismiss it.
Edit: and I was comparing apples to oranges anyway. If I use your Google Analytics data for both measures, we get a range of 0.39%-7.7%. This upper bound actually exceeds your Google CPC result. You don't have enough data.
There are unlimited examples of failed advertising campaigns on every single medium where failure can be seen measured. Most campaigns fail. They are a cost of doing business. Generalizing based on those would be very mistaken. Facebook is a new but giant ad program. The tools are still rough and "best practices" are even rougher. The consultants...
That doesn't mean that good campaigns can't be run on facebook. Facebook allows campaigns to be run that would be impossible to run anywhere else. In some cases the ROI is ridiculous. In others it's one of few things that works.
The number one reason for all these Facebook sux rants seems to be "it's not adwords." People want their adwords campaign to work on Facebook. If Coca Cola wanted to tell you that they're "the real thing" on adwords, it would be an uphill battle. A budget app on Facebook might be hard going on fb. Maybe not impossible, but it's a squeeze.
If you want to advertise a local children's art exhibition taking place this weekend, Facebook ads will work like magic. 'Friends of friends of the gallery who live close by and have kids.' There is no other platform that gives you anywhere near the reach, relevance and context that FB gives you for a campaign like that. I would expect the "ROI" to be under a dollar per physical ass-through-door.
I don't think the article in anyways suggest that they have a perfect advertising campaign.
He's just talking about the discrepancy between what FB reports and what he sees. I don't see how any of what you wrote actually relates or is pertinent to the article.
It looks like you just took offense, along with a lot of other commenters, to criticizing Facebook.
Here is a good presentation from the quantcast guys about the "natural born clicker" problem. The people clicking on your display ad are probably anything but actual potential customers.
Clicks is just an easy holdover metric from the paid search side of digital advertising. It doesn't make sense in the context of early funnel ads. You need to measure the effect your display ads are having on your purchasing endpoints. Which is what the whole cross channel attribution industry is about.
Its quite possible your are getting good value from facebook ads, you've just inadvertently focused in on the worst subpopulation, the clickers.
[1]http://www.slideshare.net/hardnoyz/display-ad-clickers-are-n...
Simply exclude the countries that are known to be click farms from seeing your page at all.
On your page settings, you'll see a "Country Restrictions" section. http://i.imgur.com/snkv77Q.png
When your page is not visible to a certain area, Facebook will not serve ads to people in that country.
Bam?
I say this only because I had a habit of flagging ads with the reason of 'other' and entering 'wrong language' when FB showed me ads in the wrong language.
FB ignore browser language for ads, preferring to use IP address. Duh.
It might just be that x% of people flagged a number of your ads for similarly odd reasons and you got banned automatically.
Also I'm testing a theory that FB will at some point just tag me as an 'ad hater', because any ads that show in the timeline I mark as 'don't show me again', reason 'not interested', purely because I don't like the feature.
I'm not going to say that copy was not good or that the number Facebook tracks are correct. I find the copy of the ad used pretty good overall. However I've some consideration about it:
- I totally agree that Facebook must improve its tracking and must do more to prevent clicks fraud ... a problem which is still very relevant
- Lot's of Facebook Ads traffic comes from mobile nowadays. This can be good or bad. If you're promoting a website and aiming at conversions on a non mobile-friendly website you MUST disable mobile targeting.
- Overall $50 budget is not enough to get to any relevant conclusion.
- On a product like this (budgeting, finance, etc.) it's critical to find a very good audience to target. I'd suggest using a lot custom audiences.
- Facebook Ads bounce rate & overall quality is very often lower than Google, Yahoo & Bing, this is implicit in the nature of the platform. On Google you're getting traffic from people who are actively searching for a keyword strictly related to your product. On Facebook you're targeting people based on demographic profile and a vague interest. However Facebook is very often much cheaper than Google.
- CPC & CTR are meaningless metrics. You should always have conversion tracking and measure the overall CPA to acquire a customer. Click frauds, wrong reportings etc. ... they exists. You cannot do anything about it. You should not give a crap about it. Just check your Cost to acquire a customer and see if it makes sense.
- Sometime for some markets Facebook Ads for direct conversions simply don't work. Create valuable content like eBooks, webinars etc. to get cheaper leads and then close the sales funnel with targeted emails.
My 2 cents, hope it's useful for someone :)
I did really well running dating ads in every English speaking market, and a lot of Spanish speaking markets as well.
FB ads were the second step in my post-college process of bootstrapping myself as a viable economic entity amidst the fallout and financial devastation of the sub-prime mortgage crisis.
So thank you Mark Zuckerberg, if it wasn't for your creation I might have had to get a real job.
Note a photo of a smiling female is the best creative for CTR, and narrowing the demographic in her use case was simple: females within 15 miles of her office, aged 25-45, in certain income range. We think the average lead who actually called probably saw her ad 6 or 7 times before clicking.
OP may have valid criticisms of FB ads, but in our case it was a massive success. Spending a couple bucks to acquire a client w a LTV over $1000 is a no-brainer.
Again, YMMV but if you use it right FB can be a fantastic tool.
Compare the author's: "Easy to use, free online budget" to "Scared of being in debt? Get your FREE budget report instantly. Click here to request info."
I'm not saying that's the ideal copy, but you have to get people's interest and explain more. Make it specific to a location like "Virginia" or "Sydney" or "Melbourne" or "Kentucky" and target those specific places you'll get a higher CTR and conversion. The mobile vs. desktop part is a whole other discussion.
FB Ads is a very stubborn creature. There's a lot to learn in order to make it work, their editorial team is trigger happy with account bans... but the volume is massive and the targeting options are amazing.
Running a 60$ is nothing on FB, you need to run volume and optimize.
I"m doing a lot of mobile right now and you can go anywhere from .10 to .50 per install and basically scale to infinity if you like.
First facebook ads would drive online ads pricing towards the bottom, then it would make obvious something almost all of us know: online advertisement is mostly an overpriced scam that doesn't work and most netizens despise.
Then the usual business model to support costs for running a website would crumble and disappear.
Sadly I can't find this article now (thanks to google tweaking its search engine, it's now hardly possible to find an old results or anythine relevant past the first half of the first results page), but I remember it pointed out that facebook users are much less receptive to ads than google search users. People using a search engine are actively looking for something and ads can be actually be useful to them, but for people looking for social interactions with people they know ads are quite useless and an annoyance.
Right now facebook lack of transparency and accuracy in their ad business means more profit and less trouble for them while hiding the elephant in the room, so don't expect the situation to change soon unless they're given incentive to do so.
...and therefore every single derived stat is completely nonsense. A percentage you say, on a sample size < 100?
Whats your confidence level on that?
(I also think that Facebook ads are a waste, and the conclusion is plausible; but the stats in the post are meaningless and probably deceptive)
Simply run your campaign for $X and measure your resulting sales, $Y, and now you know if you are wasting money or not.
If Facebook 'fix the problem' then the CPC rate will simply increase.
The post's conclusion is that there's a strong indication of Facebook charging for mis-clicks and double charging for non-unique clicks.
Google in the same timeframe has had a measurable ROI and is converting at ~10% for us; even mobile clicks.
This is just data, it's worth experimenting for yourself but I definitely feel that something sketch is going on. Make sure to use utm_ codes and something like MixPanel so you can track the originating source for your paying customers.
If you stick with those, you're pretty much guaranteed to be targeting real people, and not bots or fraudsters.
Also, learn to use the Facebook Power Editor, as you get a lot more control over how your posts appear, how your ads work, etc.
I have dev'd for an advertising company, have worked with several campaigning networks like HasOffers, and have found similar results. This is more than common.
Still very interesting that the bulk seems to be Android (mobile traffic). A must know if you are not targeting mobile..
For this type of website, he should be bidding desktop - will pay maybe 40% more per click, but much better site engagement.
FB Ads still have a long way before the tools are as robust as Adwords, but learn the platform and run more tests before you trash it. Unless you're going for something ultra-targeted it's rare to nail a CPC platform on the first go.
Call me a hater, I am one, and completely revel in the privilege :)
However, I have had little luck with adsense for the same company. Honestly I think picking an ad network for your market is a much bigger decision than "tuning" a network you are set on using!
It is a bit more complex but you will get more possibilities with this editor: https://www.facebook.com/ads/manage/powereditor/
Even if you optimize for mobile web, I'm sure they won't experience the true power and wow-ness of your app unless they visit it thru a desktop.
This sounds like a bug (or a feature, i.e. fraud).
What is Facebook CPC??
How can Facebook fix this? They need to work like Google.
- Why fix something that's not broken? It's working exactly as Facebook intended it.
Don't do business with scumbags.
Facebook is professional scumbaggery.
Just a taste
https://en.wikipedia.org/wiki/Criticism_of_Facebook#Privacy_...
https://en.wikipedia.org/wiki/Criticism_of_Facebook#Data_min...
https://en.wikipedia.org/wiki/Criticism_of_Facebook#Inabilit...
http://www.dailyfinance.com/2010/06/03/facebook-ceo-mark-zuc...
http://www.socialmedianews.com.au/zuckerberg-in-trouble-over...
http://www.techrepublic.com/blog/it-security/why-you-should-...
There are so many ways this post is wrong. First, a .4% CTR for a newsfeed ad sucks. That means either your demo targeting sucks, or your ad sucks, or both. Second, if android visits don't convert, change targeting to desktop visitors only. Third, traffic sources behave differently. You can't jump to the conclusion that they're scamming you just because one traffic source worked and another one didn't. Another possibility is that you haven't tried hard enough.