Facebook Advertising for Lead Acquisition Case Study
My name is dk, CEO of Purpose Inc. John has very graciously allowed me to share my Facebook Advertising Presentation from Affiliate Summit with his readers.
When I speak, people repeatedly ask for actual case studies. The week before Affiliate Summit, in four days, I did the following test. I spent about 4 hours on it, and one of my programmers put in probably about the same amount of time making landing pages.
I first started 10 years ago in local internet marketing for my La Jolla Chiropractic Office. I got into Facebook advertising, after reading Shoemoney’s first write up on local facebook advertising, which is now part of the Shoemoney System
6X Brazilian Jiu Jitsu World Champion Robert Drysdale
I have been helping my friend Robert Drysdale with his new site where you can learn to do Jiu Jitsu online. Here is some info about Robert on my first slide.
Facebook Advertising Campaign Goals
I spoke in detail with Sean Rigo, Robert’s business partner on the site. We discussed expectations on the site, and agreed that if we could get quality people to visit the site, who were interested in learning jiu jitsu online, for $10 each we could probably make it work. The site is not live yet, so this is only a guess.
Whenever I do a campaign for any product, the first thing I do is put myself in the position of the user. Why would they buy the product? Who would buy the product? What do I know about them? The key question was…
PPC Advertising Landing Pages
Initially just put up landing pages, whose templates had worked before, but with new text and images.
Facebook Ad Demographics
The initial goal here was to find an image, title and ad text that would lead to at least some clicks as a starting point. We then tried many different ads with different demographics to see what we could discover. This would lead to the starting ad and demographics.
Then I came up with more ads with different demographics that we tested.
Click Through Rate
The CTR is the percentage of people who saw the ad, that clicked on it. We used that as a first indication of an ad that made people interested. The final ad that says De-Knot your stomach got almost no clicks, and Facebook stopped showing it. One of my staff thought the guy looked hot, so I imediately killed that ad!
The winner of the first round of testing was, “Afraid to Go to School” At $0.29 per click, it was in the ball park for traffic we might be able to work with. This had a grammar error in it, Bully’s instead of Bullys. It worked with the grammar error, so we left it.
Testing Color Variations
Now it was time to test for better landing pages. Without the site live, I preferred to get the traffic as inexpensively as possible, and verify the users were actually interested enough to at least give up some of their personal data. So we kept going with the testing.
We had made the following landing page when we first made the Bullys ad. No special thought had been used to make the page, but instead was made based on successful templates we had used in the past. The ads were written from the mindset of a kid who is afraid of going to school. When I was a kid, I went to Sun Valley Junior High in Los Angeles where there were gangs, drugs and bullies. This meant that I have firsthand knowledge of being a bit scared of going to school! 
We started with the ad that was already working and decided to test color changes. We could have tested any variable on the ad.
We used the same landing page for each of the following ads. We then changed blotches of color, and the color of the shirt of the bully. At this point we switched over to measuring the tests by the cost per lead generated.
The costs per lead generated varied from $3 to $9. We used tests of 50 clicks or more.
Cost Per Lead
It is possible the $3 and $4 ads really had the same result, and it was just random chance. Going onward using the green shirt $3 lead was a safe bet.
ADL – Ad, Demographic, Landing page
The Ad (including title, image, and ad copy), the demographics, and landing page together make up what I call the ADL (Ad, Demographic, Landing page). Remember you read A.D.L. here first on John Chow’s blog!
Now that we had a decent ADL, (Ad, Demographic and Landing page) we wanted to see if we could lower the cost per lead by improving on the landing page. At this point I turned the landing pages on as I sat down to watch Avatar. This means the ads were going to run at night, which often yields a different result than during the day.
Testing Landing Page Variations
We started with the landing page we had used in the initial test. We then did variations of that.
I gave this part to Mihai my Romanian programmer. Vampires often choose red as the first color to test : )= (that is a vampire smiling which you also saw first here on John Chow’s blog!)
Here Mihai made the title totally red. The orange submit button comes from Shoemoney’s research that orange submit buttons often work best.
You never know what works, so you try whatever sounds interesting.
The downside of course of having a non native English speaker choose the text of the ads can have results like the one below.
The winning landing page gave us leads at $3.20 per lead. This was more expensive than earlier ads, which I think was because these ads were run at night, and had a lower fill in rate.
Ad Title Testing
Now we had a decent ADL, (Ad, Demographic and Landing page). We then tested new ad headlines to see what changes this would yield. There is no correct order to test in. You can go on and on like this, forever, or start completely over with a new concept. Remember this was all based on the thought that people learn to fight to prevent getting their butts kicked. We could have just as easily started with doing Jiu Jitsu to get stronger, improve balance, enhance self defense, get ready to get into the UFC, or a million other ideas. Another important thing we can test is what my co-speaker Mark Colacioppo CEO of Globalizer said, that you can test running the ads just during certain times of the day, and see when they perform the best.
We then tested 15 different titles on the ads, while keeping the rest of the ADL (Ad, Demographic and Landing page the same.) I honestly came up with these different titles in about 3 minutes, asking my staff who happened to be standing around me for any ideas they had.
The Winning Combination
The winner was, much to my surprise:
We now had an ADL, (Ad, Demographic, and Landing page) that got us leads at $1.43 per lead. So, no more fear for me! I had a success to present!
When the site goes live, we will be able to test these leads, and see if this is profitable. Whether it is or not, we will be back testing, testing and more testing to find the lowest possible cost per quality lead, and ultimately the greatest return on investment.
If you have facebook questions, just send an e-mail to me, dk or hit me up on twitter.
Those of us who get to read John’s blog are very lucky. If you read between the lines, you will see a story of a guy who comes from very humble beginnings, who worked very, very hard, studied hard, and pulled himself up to the top. John’s life is a fantastic and inspirational story that never says, why me?, but instead just pushes forward hard to success. It is a story that cuts across generations, cultures and languages. It is a story that is still being written!
John, thank you very much for letting me post this on your blog. I am very honored.
Much Love,
dk
Tweet This Post!- Posted in Make Money Online, The Net
- 91 comments what's your take?
Related posts from the blog
- Affiliate Summit West 2010 – Killer Facebook Advertising Tactics
- Top 50K Blogs Made $500 Million in Ad Revenue
- Would You Like To Be My Friend?
- Digg In Acquisition Talks
- How To Make Money with Facebook Ads
- Google Posts Record Fourth Quarter Profits
- Limited Time Offer: FBAdsGuide Facebook Coaching Forums
- Facebook Ads Launching Tonight
Trackbacks
- Where do you want to go today? » Facebook Presentation now on JohnChow.com - January 29, 2010 at 4:09 am
- uberVU - social comments - January 29, 2010 at 6:09 am
- Facebook Advertising Heats Up on The Open Source Marketer - February 3, 2010 at 12:53 pm
- Facebook Advertising Heats Up - Posted In Blog | I-Checks - Ready Mix - February 3, 2010 at 5:19 pm
- Facebook Ads For Lead Acquisition « DrivenTide - February 9, 2010 at 6:02 am
Loading ...
i believe facebook is the most effective way to adsvertise our products
you know that millions people update their FB account everyday more than one hour
WOW best post on john chow so far and so much testing.
Also how long did this whole process take?
Thanks Dana,
I spent about 4 hours of my time on it, and it took my programmer about 4 hours to make the landing pages, or maybe less. It is more about coming back and checking the numbers often, than in spending lots of hours on it.
Last comment was from me, dk. Dana had been using my computer and I was still logged in.
Thanks for the comprehensive FB advertising post.
Well, its nice and thanks for sharing and millions of ads in facebook is always rotated. How about click and sale ratio.. I mean an example or proof..
Fique, the site is not live yet, so we don’t have sales numbers. Usually with something like this we will need to get the site live, and find out if these users will actually buy or not. This was lead acquisition pure and simple.
DK, are you guys going to try call tracking, as well? I’m doing something cool for my BJJ instructor. Would be happy to share what I am doing with you. Let me know bro!
Cool Eric,
This campaign was for their online training, but for their studio, as well as the other stuff I do, I would love to hear about it. Feel free to post what you are talking about in a comment here, to benefit everyone!
Hope I will see some results from the live site. Wondering, how you do the test without live site?. I’m just want to know more about it so that why I ask hope its ok with you
In this case, the target was to get a lead which consisted of a name and an e-mail address filled out. We then dumped duplicate e-mails, and ones that were obviously made up.
So its can actually work effectively on list building? If so, that the power of list building on facebook..Hmmm
Me thinks you are on to something here. I have a big project I am working on that does just as you said above.
And here was me thinking you just spent time lying underneath bunnies…
Excellent post and a technique that you can use on any type of cost per click traffic where you have control over the environment.
I like the isolation of the variables, the use of colours, the asking questions, the variety in the demographics.
I guess taking it one stage further you could try to identify the top states/towns/schools with the highest rates of bullying and target those to make it even more pinpoint.
I’d love to know where I could get one of those radio-active t-shirts, that would definitely stop you being bullied just being that.
Other titles you could use :
Stop being a wimp.
Hey pussy, tired of being pushed around?
Looking forward to reading the rest of the comments on here, I am sure there will be a ton.
Heading off to retweet this post now.
LOL, Yeh, I am sure the facebook guys would love the Hey….tired of being pushed around! It is true the more scandalous the title, the more clicks it gets, but at the same time it may offend some of the users, and could actually get your account banned.
It is cool how many variables there are. The combinations that you can make between all the variables is astronomical.
Great analysis DK. It shows how we can’t just throw an ad up and expect the same results. Please don’t kick my ass.
O.K., but you have to be good
DK… Thank you for this post… It couldn’t have come at a better time for us. I’ll be referring back to it all day tomorrow as I set up our Facebook campaign for our new product…
The best thing I like about your post is.. Our product is free and the example you did was on a free product… Outstanding… Thank you again…
Brian, if you want me to take a peek at it, fire it over to me dk (at) purposeinc (dot) com
Great analysis John! In my view, Facebook is become the most effective way to promote our new product.
great sharing… keep it up…
hope that i can make money out of facebook..huhu
Very informative post dk! Thank you so much but I am confused, or maybe a little out of touch with our youth but where would a 18 y/o kid and younger get 79 bucks a month to learn jiu jitsu online.
I would think targeting an older crowd with something like “Defend Your Wife” because i would think they actually have purchasing power.
Or am I just way out of left field on this one?
No Vini, I think you are possibly very right. At dinner after the presentation Steve Poland said the exact same thing to me. I used to spend plenty of my parents money, and there are plenty of kids in the world who DO have that kind of money.
The real test of course is to run this ad when the site is live, and see how it converts to cash, and or do an e-mail campaign to the leads.
Would “Defend Your Wife” work? Sounds like an idea worth testing! I am never arrogant enough to think I know for sure about any add before I run it. Testing gives the answer.
I always tell people who doubt the purchasing power of teenagers to check out x-box live. Millions of kids with full access to their parents cards – a lot of which have no problem cheating, getting banned and then creating a new account several times a month.
Yeh Paul, Games played by Teenagers Online and Off are of course a powerful example of teen money spending. I never had trouble finding money as a kid when I really wanted something.
I forgot to say thankyou for the case study. It’s already highlighted to me where I’ve gone wrong in the past. Too lazy, not trying enough variation in my split tests. Seems the key is to really pump out as many variations as you can.
You are totally right. The crazy thing about it, unlike any other industry in the world, is that testing a new ADL is virtually free except for a few bucks for the clicks, and the time making the landing pages. When I did direct mail in the past, it was almost impossible to do a test for less than $500.
Wow adding of changing color can make that much of an impact? Time for me to go back to the drawing board. My Facebook ad have 39k impression 36 clicks LOL. If i was paying per CPM i would be getting a deal LOL.
I predict you will be shocked how much of a difference you will be able to make.
Great sharing! It is surely help me to know what possibilities in facebook advertising.
Honestly this was a crazy post to me. Ok I have done some facebook advertising but never so extensive as this one. The time and effort for putting this all together seems like a lot if you are a newbie.
Very good information though really opens my eyes on how much work you need to put in to finding the lowest cost per lead. Seems pricey for someone like me that does not want to invest that much money into from the beginning. Screwing up in the early stage can be crucial.
Brian, a real secret is to find a group that you can advertise to with inexpensive clicks. Searching for these groups is free, and facebook’s interface allows you to look around all day and gather information at no cost. Then you can run tests literally for a $1 or $2. As far as making landing pages, you can get an old copy of photoshop for making images inexpensively, or some other free image software. For making web pages, some type of notepad is free on all computers, and learning html is free with all the info you need online, so making landing pages is free as well. Just keep a tight eye on the daily budgets, and turn off tests the second you prove to yourself they are not making money.
Wow now that’s genius. I think I learned a lot from that. Never knew that FB advertising could be this successful. Unbelievable.
Thank you, I am blushing from the compliment
Well you’ve done something that most people fail at. And it was always thought that it wasn’t possible to make money from CPM and social networking sites. So you’ve just re-opened a market to me by showing me that it can be done. One of the most inspiring and concise posts on John Chow. Cheers!
Great post DK. I had the pleasure of seeing the presentation live. I found it interesting that you targeted people who are afraid to get beat up rather than people who want to train in jiu jitsu or MMA for the sport. I wouldn’t have considered that angle. Obviously, it worked.
We hear time and time again… test, test, test but we don’t necessarily do it. This was the kick in the rear I need it.
I’ll be calling you soon about a campaign.
You know I got your back Lyndon. I don’t know why people would not test. I bet people interested in training in jiu jitsu would work, but would not have been as funny of a presentation. Because of the controversy surrounding the facebook session at Affiliate Summit, I wanted to deliver something funny and entertaining as well as educational.
Interesting, I was wondering why that approach was taken. A number of years back I join a muay thai club for a few months and it was intimidating registering and getting started, I also spent a lot of time on Sherdog and it seemed like a lot of guys trained at home for a number of reasons. I think there may be a big market from the guys who won’t join a gym because they can’t afford it, are not near one, or want to learn the basics before they walk into a club. If you run those campaigns I’d love to hear how they go.
I think the smaller (cash flow) and lazier people get discouraged at a buck a click with no conversions, and shut down their campaign right away assuming it does not work.
I think I speak for us all when I say we would love to see a follow up post talking about your conversion ratio once the system goes live and we would love to see what other campaigns you run in the future for this same offer (if you are willing).
Thanks for the free education
Great analysis DK. This was lead acquisition pure and simple. Facebook is become the most effective way to promote our new product.
I’ve never taken a look at Facebook ads. First of all, how much money did you spend on this testing. Secondly, where did you host the landing pages?
Thanks
A quick loose count looking at the facebook account I used added up to $752. $390 were at the end, confirming my final add to make sure the results were correct. I did not keep a close eye on it, because I was more concerned with getting my results done in time for my presentation, than I was concerned about making a profit. Often at the beginning of a campaign, especially if I own the product and plan on following through no matter what, I will focus on finding the best add, without too much concern about the cost of finding it. Then when I find a good add with a good ROI, then I can run it for a while while making profit off of it for a while. Obviously if you screw this up too much you will not be able to make up the loss.
Answer to your second question, I have a dedicated server with secondary accounts on them that I can allow different programmers access to. In this case Mihai
= put the pages up there one at a time as we created them.
Thanks for the info.
One other question. Do you assign a unique domain for your landing page? Like as in go out and purchase a domain specifically for the decided upon landing page.
After trying all kinds of things we ended up doing the following. I will grab a domain name that I don’t care too much about, that I can use for many things. One of mine is a name that is not, but something like, offerofthecentury.com so it makes sense with any industry and any landing page. I have quite a few of these mediocre domain names that make sense. If you are going to be doing a lot for a certain industry, then get an industry specific domain name for those pages.
Then each landing page we make is in a seperate folder with a number on it. Examples would be dealofthecentury,com/10003/index.html and the next one will be dealofthecentury.com/1004/index.html and so on.
I actually feel very low tech doing this, but it has worked out very inexpensive, and incredibly fast to make. Each add goes to a unique landing page.
Alex at Facebook suggested a more sophisticated way to do this, and another friend who is a huge facebook user confirmed that they did it exactly the same way that we do. The way we do it is a technology that anyone can learn easily.
My buddy Eric just came out with a wordpress plugin, literally today . I haven’t played with it yet, so don’t know if it would work or not, but kind of an interesting coincidence. If someone could go check it out and let me know if it would work for this, I’d love to hear. I know it is for adwords for sure.
Great! Thanks for the info!
So you guys are targeting one particular segment … interesting.
That’s great info right there! Great post!
Thanks dk!
dk,
This post illustrates the difference between the bottom-feeding affiliates looking for a guru to “tell them how to make money”, and the guys who actually go out and do something and make money.
There is absolutely no possible way that you could have come up with what ended up being the winning combination, without testing tens or hundreds of combinations. Someone could easily have tried some of the combinations that you tried, not made any money, and ended up saying “facebook ads sucks”, or “there’s no way this campaign will work”.
Kudos to you for doing a GREAT job sharing. I hope you and your client have great success with this campaign.
Yeh Greg, I strongly believe in quality products. Matt Cutts of Google was one one of my earliest influences online, and he helped me a ton, and as a result I have made a lot of money off of doing well in google. I am a pretty good guy, and also as thanks to Google I really try to get quality stuff on the net and not just crap. But believe me I am not perfect or even close.
Matt Cutts is one of my biggest influences when it comes to SERPs. Great blog and super useful info.
woow… that’s very great. i readed every one, and thanks for your post.
Sharp Analysis… I get it now Thanks
This is the best analysis I’ve ever read relative to online advertising. Now I have a better idea why all of my efforts, especially in the travel affiliate space, have been terrific flops.
Thanks to John Chow’s advice, I enrolled in the ShoeMoney System and look forward to learning how to successfully run an ad campaign, just as you have.
Thank you for the nice compliment. There is no magic bullet to this stuff. It takes time, commitment, and trying lots of things. I am glad you got into the Shoemoney System before it sold out. Jeremy has a ton of good info to share.
nice post i always like case studies facebook advertising is one of the methods i have done very little with. this post definitely is giving me some ideas
Awesome presentation.
So it means we need to test test test and test some more.
So Test Test Test is answer. Thanks for the awesome info DK! I wish I still had a fb ads account I would def try out a new campaign…oh well was fun while it lasted
Come on..how hard is it to get another e-mail address, and a new credit card, and set up a new account. Just don’t do anything scammy this time, be good to the users, and you will keep your account! LOL
LoL I have a love hate relationship with fb. I never understood why they approve or disapprove things especially when you believe your following the guidelines to a “T”
I’ll give it another shot just because of this post. I’ll stay away from the dating ads and cruise under the fb radar.
Wouldn’t the new redit card need to be in a different name though?
At this particular moment they are approving pretty quickly. They are still learning and subject to change at any moment. Don’t be afraid of the dating offers, just don’t do anything dirty or scandalous. What’s the worst they can do? Disapprove an add?
It is surely help me to know what possibilities in facebook advertising.
Wow Nice post! I haven’t tested that much with facebook ads. I tried when facebook ad plataform was starting and it was more less…
I will test it and follow some of your advice.
I wrote on my blog a little review of this blog post
http://emailbuilding.com/blog
Talk to you soon.
Thanks for the blog and link love Jorge, it all ads up and flows.
@JohnChow, I am really impressed with your readers. For having 50k to 100k people read each post, there seems to be few or no really negative ones. Maybe you just delete those comments? Either way, awesome!
most people comment on blogs to gain traffic to their own sites/blogs…so whenever they comment its not always real…:)
btw–I don’t think he deletes -ve comments unlike other big shots like Mike Filsaime.
John has done good in his terms…
That’s amazing. I am actually stunned by how positive all the comments are here.
if you don’t write positive comments then you will have less chance of getting clicks to your site by other users….and like I said most the comments are just there for the sake of it. I am not sure they really mean what they write.
This is the reality…..
BTW–congrats for making the list of top commentators
Best post ever !
Many many thanks, very interesting
Awesome post!
But how do you only spend 4 hours working on this project. Did you use some automation tool to upload all the ads and ad changes?
Also do you have some automated way to collect all the data and results? Thanks
How many different ads did I run in this? Like 50 maybe? 80 tops? It takes about 2 minutes to enter an add into facebook, especially when you are copying one. I spent more time finding the pics, and thinking of clever titles.
I used to use the fb ads manager to help me bang out a campaign and ad variations. I dont know if that’s frowned upon but it helped out a lot.
Awesome real world example / case study. Thanks for sharing!
Great post, it shows the kind of testing that sometimes needs to be done! Thank you
Thanks a lot for this post. It’s a really great example of the process of tweaking an Ad for maximum conversion and I highlighted/linked to this post in my latest blog post.
you rock andrew! Will go over and see it!
Thanks for the share.. This is a great case study
Very interesting. Good information from lots and lots of testing.
This is the most amazing case study I have ever seen. I can now see why you are so popular- this is the perfect example of how great split-testing works.
I’m going to bookmark this and come back to it once my own company starts putting facebook ads out- thanks so much John!
I love this site! Thank you for this case study! It will be very helpful!
Thank you for the nice article.
this was extremely informative!!
Great Post DK!
I learned a lot! I think John Chow should have you as a Guest blogger for often!
Fantastic post DK thanks John. John you should have people like DK post such great case studies more often
I could do one if u like 
=
Great insight.
I like that you targeted people with the bullying technique.
Also, does Drysdale still think Forrest can submit Silva after he KTFO him? HaHa
Thank you.
I will tell you after watching them train. Forest looks unstoppable.
Let’s just hope we get to see Robert in that UFC ring one day.