Biography: The Rise of Ngo: Chapter 5
I’m going to analyze the campaign from my newbie perspective. My methods have obviously evolved since then, and some of these techniques aren’t best practices for 2013. I’m just trying to give you insight into how I approached the campaign as a newbie.
Reading Nick’s post, I came to a few conclusions about the campaign:
- I wanted to direct-link this offer.
- I needed to brainstorm images. I used in-game screenshots to begin with.
- I pretty much used the headline examples Nickycakes gave plus came up with my own stuff: Game of the Year 2008, Better Than Half-Life 2, Free FPS Game, Play Games In Your Browser, etc.
Remember that this was my very first time advertising on Facebook; all my previous campaigns were with Google Adwords.
In order to optimize, I wanted to keep things as simple as possible. I thought of it as a science experiment, which has controlled variables, independent variables, and dependent variables.
Test #1: What are the best age groups?
- Controlled variable = elements that you keep the same. The offer, images, and ad text would be kept identical. I just picked an image that did well, used “Game of the Year 2008” as the headline, and featured one offer.
- Independent variable = what am I changing or testing? I’m testing the age groups.
- Dependent variable = the new value that changes. Basically I’m looking at the click-through rates, CPCs, and the return on investment.
First thing I did was split-test the age groups.
I set the initial bids high. The results were that the 13-18, 19-21, & 22-25 demographics did the best, and the 19-21 group was the most profitable. Unfortunately, I lost money and was at -100% ROI. No big deal.
Test #2: What’s the best headline?
I used the 19-21 age group since it did the best, and I just used an in-game screenshot. Remember, I’m only testing one thing at a time. For the headline, I decided to test “Game of the Year 2008” vs. “Better Than Half-Life 2” vs. “You Suck at This Game.”
This was a big a-ha moment for me. “Better Than Half-Life 2” doubled the CTR of the other two. I was comparing this game to a more popular one, and it worked. It’s like saying your new band is better than the Beatles—it’s going to get people curious.
I did another headline test where I used the top 5 most popular first-person shooting games at the time. There was one game that absolutely demolished the rest of them. I’m not going to mention which game that was since I don’t want to get into trouble.
Test #3: What’s the best image?
While going through test #2, I realized it was important to find the best “angle”—the best way to approach the campaign on a creative level. Now that I knew the best angle/headline, I needed to find the best image to go along with it. I used Google image search and pulled in-game screenshots from that ultra popular FPS game I found in the headline test.
I couldn’t believe my eyes.
The images gave me an absurd CTR. I can’t remember what the CTR was, but I remember that my clicks were around $.02-03 each, but my earnings per click were around $.30 EPC.
The first day the campaign popped, I spent maybe $350 and generated $1,000 in profit. It was insane. I would repeatedly refresh my stats, and the numbers kept going higher and higher.
An easier way to think of split-testing is to imagine you’re having a tournament. You upload 10 ads at the same time and let them battle. Whichever has the highest click-through rate is your champion. Your goal now is to find other warriors that can be champions. Sometimes, your champion gets tired and worn out (banner blindness), and you need to find a new one.
Pro tip: Just because an ad has the highest CTR doesn’t mean it’s the best. You need to also check the conversion rate of the ad through a tracker. I’ve had some ads that got extremely high click-through rates losing money because they didn’t convert well. I didn’t have a tracker back then, so I just went with CTR to measure the effectiveness of the ads.
Expanding the campaign
Now that I found the best angle and best images, it was time to scale. Originally, I tested only one demographic, so now I took the best headline/image combination and expanded the campaign to the rest of the age brackets.
I also expanded the campaign to other countries such as Canada and the United Kingdom. Expanding to those countries was easy profit because these countries weren’t as competitive.
Cash flow issues
A big problem was coming up…I was running out of money. I was on monthly checks from Neverblue and requested to be upgraded to weekly wire payments. The issue was the first wire would come in about two weeks, and I could only last another two days at the rate I was going.
I needed a few thousand dollars, and I needed it fast. I didn’t even consider asking my parents because I wanted them to think of me as a man. If I achieved success, it would be 100% on my own.
The only person I knew who could let me borrow that kind of money was my roommate at the time (I was living in his house). I had never borrowed money in my life and didn’t want to do it now. Instead, I was going to make him a business proposition.
I told him I have a business that’s doing 500% ROI. I pulled up Facebook and Neverblue and showed him the numbers. My proposition was if he loaned me $5,000, I would give him $7,500 within a month. That’s a 50% ROI for him. He gave me the cash with no questions asked and said if I didn’t pay him back, he would kill me in my sleep :-).
Analyzing the campaign from my 2013 perspective
There were a few things that made this campaign successful for me. The biggest one was that I simplified the optimization process and made a very clear plan. This was the beginning of developing my “system” of optimizing campaigns.
My Google campaigns were way too confusing. I had to optimize the landing page, the offer, the keywords, the bids, the quality score, the ads, etc. Which do I test first? Do I test them all at the same time or one by one? I didn’t have the experience yet to handle this many variables.
With the Instant Action campaign, many variables were eliminated.
- Offer – I didn’t have to split-test offers or offer pages; the offer was instant action, and there was only one page.
- Landing page – I direct-linked, so I didn’t need to have a landing page.
- No quality score to deal with.
- In Adwords, you have to group your keywords in a certain way, and you don’t know the best way to do it because it’s subjective. It was so much easier to understand when I grouped my campaigns by age brackets.
I realized that you can’t master every traffic source. Some people just understand the optimization process easier for one place over another. One of my friends is amazing with media buys, but he can’t grasp mobile. I wasn’t really good at Adwords despite my experience, but Facebook came as naturally to me as Koreans and Starcraft.
The biggest difference maker in this campaign was the fact that Nicky said this campaign works. I had affiliate managers telling me that this campaign was hot, that campaign was hot, etc. Having someone like Nicky saying the campaign works gave me the confidence to continue working at the campaign even if it was losing money at first.
Gaming on Facebook still makes money. If you are a newbie, I highly encourage you to get in on it.
Here is what this campaign taught me to avoid:
a) don’t advertise to such a young demographic. My ads were directed toward people under 21. People under 21 don’t have credit cards. The leads won’t pan out, and you’ll get kicked off the offer.
b) don’t use screenshots from other games. Even though the CTRs are nice, it lowers the quality of the leads, and you risk getting sued.
Spending my first check
When I got to work at the end of my first month, I saw a direct deposit pay stub on my desk: I was paid $2,500 for my first month of working there.
I had made a $2,500 profit the previous two days from affiliate marketing. Two weeks later, the first Neverblue wire came in for $7,000.
How’d I spend my first wire? You have to thank the people who got you to where you are. I wrote a check to my parents to help them out with the bills.
In the last chapter, I mentioned that I needed to be frugal in order to cut costs to invest in my business. I didn’t mind it at all because I knew where I was heading. I was quite fortunate to have a girlfriend at the time who never complained about how cheap I was. I drove her to Phipps Plaza and told her to pick out any handbag she wanted (oh, man, thank you for picking a Coach bag instead of that Chanel).
A Samurai needs the sharpest sword, and an Internet marketer needs the fastest technology. I upgraded my $10-a-month shared hosting from HostGator to a dedicated server from LiquidWeb (I’m still using them today). I was using a really slow, 5-year-old computer at the time and upgraded to a new Mac.
I wasn’t thinking I “made” it yet. Anyone can have a lucky campaign, but I knew I needed to test whether I could be profitable in other traffic sources and verticals to see if I had the skill.
Now that I had a taste of success, I smelled blood and was hungry for more.
I knew that my life would never be the same again.
Next time on The Rise of Ngo: running more successful Facebook campaigns, heading to Affiliate Summit East, and quitting my job.