My First Try At PPC to CPA Affiliate Marketing
Introduction
I have recently been trying to figure out PPC to CPA affiliate marketing. My first attempt was through an offer on NeverBlue Ads promoting a $250 grocery coupon. This CPA offer required a zip code submission to be counted as a conversion. The offer had a low $1.20 CPA, which means for every sign-up (acquisition) I would make $1.20.
This was the first offer I had with NeverBlue, and PPC -> CPA marketing in general. This means I had no history with Neverblue and they would be less willing to accommodate any special requests I had. Also, since this was my first time trying this I did not want to take a huge risk with my money. After reading countless forum posts I realized that you have to take a risk to make a profit. That being said, choosing a low paying CPA deal often means that the CPC’s will also be proportionately lower, reducing my risk.
Some Quick Math
With a CPA of $1.20 I did not have much room squeeze out a profit. Assuming I want an ROI (return on investment) of 100% and my conversion rate (clicks compared to people who actually signed up) was 15%, I would need my CPC’s to be $0.09. This is based on the formula: (CPA/ROI)*CR=CPC. This formula says my CPC should be equal to my cost per acquisition ($1.20) divided by how much I want to make (100% profit) multiplied by my conversion rate. I had asked my account manager to raise the CPA amount, which he did to $1.35. You can use the same formula to calculate what my target CPC’s were at this rate.
Creating the Landing Page
Many affiliate marketers will tell you that creating the landing page usually takes the most time with regards to setting up a new CPA offer. I also found this to be the case. Since I am not a designer in any way I had a difficult time coming up with ways to make creatives for my landing page. So I did what any beginner would do and I put together pieces of other people’s landing pages along with some customization to create mine. I went through many iterations before finalizing the design, but here it is:
The main thing to notice is the zip submit area. Originally, that was not there and was actually occupied by a “continue to the next page” button. The actual zip submit form is located on the advertiser’s landing page, which lowers the CR (conversion rate) because it requires users to click an extra page before signing up. I was not happy with this so I iframed the zip submit area of the advertiser’s page onto my page. Can you tell? 
This increased my CR. I’m not going to go much deeper into creating landing pages, but feel free to ask any questions and I will answer them.
Setting Up the Adwords Campaign
I will give a quick overview on how I was taught to setup Adwords campaigns. Basically, every keyword should be in its own ad group in the singular and plural form along with exact and broad match types. With this method you should have 4 ad groups per term. This is to control quality score and to better analyze at a more granular level. Each campaign should only have one root word. My original root word was “coupon” so I created a campaign called coupon. I used both the singular and plural form of the word and also had the broad and exact match types. Here’s what it looks like: Coupon, Coupons, [Coupon], [Coupons]. I did this with every variation of coupon-based keywords I had.
The Data
One thing that I was told very early is that you have to run a campaign for a little while before you start to see CPC’s lower automatically. After about 3-4 weeks, here is what my campaign was shaping up like:
This data is for the life of the campaign. In the beginning, my stats were poor and I was losing money. Towards the end I was able to start making money but at that time it seemed almost impossible to actually turn a profit. At that point I decided to cut my losses and stop the campaign. Based on this data it is obvious that for every conversion I had, I lost $0.57. This multiplied by 223 conversions means I lost $127.11. I do not think this is very bad for a first time and I had gained a lot of experience.
Conclusion
Although I had lost money I had learned many valuable pieces of information. I learned how to optimize an Adwords campaign, how to extract keywords from broad match type, how to lower CPC’s through optimizing my campaign and much more. You can read forums and blog posts all day but unless you get your hands dirty, you’ll never know a thing. As always, feel free to email me about any questions you have about my experience.
Related posts:
Twitter Me!
Do you know if its legal (permission from ad networks like Neverblue) to iframe the zip submit/email submit?
good question. that was actually something I had asked them, and they said it was perfectly fine. I would definitely ask your CPA network before implementing something like that though.
Google won’t take kindly to your iframe.
that’s what I had thought at first. But I believe google only has a problem when you do a frame redirect (100% iframe).
i only iframed a small part of a website onto mine, so that was ok. my quality score didnt drop after implementing this, so I do believe it is ok by google.
thanks for the comment.
We normally don’t pay for Iframe traffic. almost every network will tell you up front what are accepted forms of traffic. It is appreciated when asked up front if you can experiment with other unique forms of traffic so the network can clear it with their client. Surprises in this biz don’t always go over well…
Well it looks like to me if you dropped your max bid per click to $0.10 or lower you would be generating a profit….
right, but then i wouldnt have gotten any clicks;)
the lowest i could get my keywords was above what it took to make a profit, hence my stopping the campaign.
thanks for the comment.
Hi Matthew
That was a great post, thanks for sharing all your data, its great for an analytics junkie like me.
Although your campaign wasn’t profitable overall I would deem this as a success.
I downloaded your stats into excell and took a look at the data.
A lot of your adGroups are profitable so there really is no reason to turn off the entire campaign.
You could just disable all adGroups with conversion cost of $1.35 or higher.
Or how about split testing the landing page? Even some basic A/B testing on different colours, try a few different headlines, add a security seal or some quotes from magazines.
Did you make changes to the adGroup after you put them up? You could try different ad title and headlines.
I think its more than possible for you to atleast double the conversion rate on this landing page.
What quality score did you get? I doubt you managed to get a great quality score so you could try improve that and get your CPC down.
These are just a few thoughts of the top of my head, If I was in your position right now I would put a small battle plan together for this campaign.
Disable/pause all adGroups that aren’t atleast breaking even, try some of the suggestions above and push down your CPC and up your conversion rate, once you are confident about the campaign bring back the paused adGroups.
-Matt
matt,
“A lot of your adGroups are profitable so there really is no reason to turn off the entire campaign.
You could just disable all adGroups with conversion cost of $1.35 or higher.”
Hmm…to be honest…i didnt think of that:(
Great idea….i guess its too late now that I gave all my data away.
maybe i will try turning this campaign back on…but its hard to start again after pausing for so long.
one last thing…
I do not think enough of the campaigns were profitable enough to justify turning this back on. also, the profit margin was so thin that i dont know if it was worth my time.
that being said, since you have my data…if you decide to make a run at this campaign…please keep me/this blog updated:)
thanks!
Hi Matthew,
Congrats on your test! Since you gained valuable data, this was certainly not a “loser” – you effectively paid about $127 for some conversion data. This data can ONLY be had by “buying” it the way you did. No matter what a networks says a conversion stat is (10%, 20%, etc.) you always have to test for yourself.
That being said… your decision to go with a lower paying offer because there is “less risk” is a logical one, but I believe is also incorrect. It’s a bit counter intuitive at first, but higher paying offers will give you a LOT more wiggle room once you get a single conversion. I use the lottery as an example a lot of the time… people think it has lower risk because it only costs say $1.00 but actually the risk is tremendously high because the probability of winning is so low. There is so much more to risk than merely the financial amount at stake.
Something that I think very few beginners realize is that the first thing you have to do is find out – QUICKLY – how your offer will convert. To do this, I always overbid initially… say .30 CPC. I will do this until I feel I have enough data to go on (cunber of clicks, hours, days… whatever feels right to me at the time) and then begin to lower my bids. This is a strategy I learned from some VERY big PPC -> CPA earners, and it works very very well. But of course it requires an up-front willingness to spend some money. Like I said initially, you have to “buy” the data, and then act on it.
One other thing I think you may want to consider is your target audience. In my experience, this kind of thing works best when you are chasing a very “hungry” market… people who are rabidly and passionately searching for something. I don’t know that grocery coupons really fall into that category… more market research could yield you a much higher conversion rate, especially with a higher paying offer.
I hope some of this helps… everything I know about CPA traffic brokering, I learned from one specific course, plus my own experience and that of others in a private forum. If you’re serious about this business, let me know and I’ll point you to it.
Take care and good luck!
Jonathan van Clute
–
http://www.LPGen.com
http://www.ToolTrainer.com
Hey Jonathan,
Great comment, seems like you know what you’re talking about. What course are you speaking of when you say you learned everything you know from?
“your decision to go with a lower paying offer because there is “less risk” is a logical one, but I believe is also incorrect. It’s a bit counter intuitive at first, but higher paying offers will give you a LOT more wiggle room once you get a single conversion.”
Good point, but it seems the lower the CPA offer, the lower the CPC’s I will have to pay thus lowering my total cost. But you’re right, I started my second campaign ($26 CPA) and its doing much better already and I have MUCH more room to breath.
Thanks for the comment, what are the links you dropped by the way? Looks interesting
Hi again Matt… btw your comments don’t appear to render properly in IE. FireFox is fine, but IE7 on Vista shows only the first comment from each person, and not any subsequent replies… at least for me.
But, to your response… =)
The course that taught myself and a number of other HIGHLY successful traffic brokers, is know as the PayPerClick Formula. My affiliate link is http://www.lpgen.com/ppcf.html and the direct link is http://www.payperclickformula.com. Gauher Chaudhry is the real deal, but he’s going to be pulling PPCF from the market within about a week, in order to prep version 2.0 this fall… existing owners will upgrade for free, so if you seriously want to get into this business, I can’t recommend this course & the private forum more highly.
As for lower CPCs on lower paying offers, again it’s a bit counter-intuitive… CPC is largely about competition in the market you’re chasing, and there are many low paying but highly competitive markets. It depends more on the type of offer – zip submit, email, short form, etc. For example, dating is a VERY competitive market, but you may only get $2.50 or $3.00 for a dating offer. Can be tough to convert profitably if your CPC is .50 or .75 like it can be on some keywords. I’ve also seen very high paying offers with very little competition where you can get cheap traffic… I’m running some now. But they also may convert less often… so you might have to blow through $15 or $20 in traffic to get one conversion on a $30 payout offer. Just takes the nerves to be OK with that.
The links in my sig are my sites… LPGen is a tool I developed for myself and the private forum I mentioned earlier (PPCF) but it was so well received that I turned it into a “real” product. It helps create properly optimized landing pages for PPC, and has numerous other uses as well.
ToolTrainer is, well, me!
I spend so much time in forums and email helping people out, that I am building a whole brand around it. I made the videos for Affiliate Radar that are on their home page, did the original training videos for SpeedPPC, and have made numerous videos on various tools that are available at http://www.ToolTrainer.com for free. I have a gift for explaining the highly technical in a way that people seem to understand, so I’m just pursuing that. Of course I’m also a full time traffic broker running CPA offers, as well as more general affiliate marketing.
Oh yeah and I’m one of the contestants on the upcoming reality show Top Affiliate Challenge, at http://www.TopAffiliateChallenge.com so you’ll be able to watch me and others make money live starting in about a week! =)
Take care,
Jonathan
–
http://www.LPGen.com
http://www.ToolTrainer.com
You did not include the Ads.
Hey Geiger,
Yea, decided not to include the ads because they were not difficult to make. If this is a common request, I will include the ad copy next time I report on my current campaign. You can always contact me directly and I will send them to you.
Matt
note – i am no longer seeing any comments at all in IE 6.0. i’m using Firefox which is mostly working but the name of the posters + the URL has some spacing issues for me.
Anyway, my question relates to how did you get tracking done on a per keyword basis. is there some sort of dynamic keyword insertion thing you did so the keywords were trackable at Neverblue? Hopefully my question makes sense….
yea, i dont know why that is happening. i have to try to fix it but i dont have IE on my mac.
i tracked keywords in adwords but used statcounter.com to track incoming traffic in real time. i had all keywords in exact match so i saw how they all performed individually.
Greg – just thought I’d share something here, hope you don’t mind…
You can pass the keyword that triggered your ad by simply using Google’s own DKI (Dynamic Keyword Insertion) token. For example, your link might look something like this…
http://www.neverblue.com/affiliate.php?subid={keyword}
The important part is to put {keyword} in the subid that goes to your CPA network. Then you’ll get the actual keyword that converted in your subid reports. This is the core principal that pretty much all tracking softwares use to keyword track, stuff like http://www.tracking202.com, http://www.affiliateradar.com, etc.
Hope that helps!
Jonathan
–
http://www.LPGen.com
http://www.ToolTrainer.com
I have a question regarding this: http://www.neverblue.com/affiliate.php?subid={keyword} post by Jonathan, (Hi Jonathan)if you use that KDI aren’t you giving away your converting keywords to the CPA company and therefore giving away your good keywords?
Tito
High-risk means high return. Low risk means low return. But life is full of uncertainties. Even a high-risk investment could turn unprofitable.
Jonathan – thanks for that post! i knew how to do the DKI for the ad itself… I didn’t realize you could do it for the URL though i had assumed that’s how people were tracking on the keyword level. thanks for making it clear to me.
ALL-
I turned off threaded comments, so now it works on IE. I will try to fix threaded comments this week.
Matt,
Is there any place that you can recommend a lesson on what html I need to create the iFrame? I am not code savy but, I am looking at getting started.
Suggestions?
@Not John Chow:
The code is as follows:
iframe src=”http://www.iframed-page” scrolling=”no”
I removed <> from the outside of it, so be sure to put that back in.
reminds me of what I was doing in the beginning of this month. I was able to pull $700 gross in only 2 email submits offer! in one week. In the beginning, it was doing well 10-$20 profit. After the week ended I was getting a very low conversion rate. Nothing changed but maybe i was getting scrubbed. After a lost I just cut it. However im still looking for other ways to figure this out
@Entrips at least it was a learning experience..that’s what i took away from this campaign.
Entrips –
Did you thin out your losing keywords? If you weren’t tracking at the keyword level, that is absolutely critical… you may have found that simply by not spending money on the keywords that didn’t convert, you’d have had a VERY profitable campaign. Might be worth revisiting the campaign!
Typically if I can hit even close to break even at first, I get VERY excited because it means I have a sure winner on my hands… once I thin out the poor performing keywords, and spend more on the winners… it’s an instant goldmine.
Jonathan
–
http://www.LPGen.com
http://www.ToolTrainer.com
Jonathon,
You reference Gauher Chaudry’s PPCF as the program you learned from. Gauher says he does not track at the keyword level due to using such a large volume of keywords. Have you found a way to effectively track large volumes, or do you just operate with fewer keywords?
@Jonathan,
Thanks for helpin out my readers:) Good suggestions. Do you need to drop your links in every comment though:)?
LOL
Habit… I always include a sig. I’ll refrain henceforth…
Jonathan
P.S. happy to help!
Jonathan, well i know what you are saying and you want the converting keywords. It was only about 3-4 keywords that was getting impressions + clicks. When I saw I started to lose money, I lowered bids from .19 – .10. At that time I was in mystery. I was getting clicks like crazy but no conversions. Maybe I should do another $20 test.
Jonathan-
no worries, i’m liking your email auto-responder too. i’ve gotten like 20+ emails from you so far:)
random question: i can’t seem to get brian’s threaded comments plugin working in IE (works perfectly in FF)…any experience with it? I dont know what to do since it works flawlessly in FF. i turned it off for now, as you can tell.
Were you using the search network, content network, or both? You can get cheaper ads on the content network, but I hear they tend to convert a little less well.
I am going to be trying something like this in the near future as well, except I want to use AdCenter and YPN as well. we should compare notes when I done, and see how it goes.
@Glenn,
I was only using the search network. i didnt try content. on my current campaign, i am using search only, but plan to expand to content very soon. i bought some YPN today, but tracking was not setup properly so i paused it.
I absolutely wanna hear how your campaign goes, and sure we can compare notes. email me with updates or post to my blog, either way works for me. thanks for stopping by.
matt
Very interesting things to note here. Have you ever tried this with any other search engine (not adwords)?
i tried it briefly on YSM, but i didnt have much luck.
Looking at post 22 and 23, @ Matt and Glenn,
Great post and discussion! I’m unclear on the difference between the search network and the content network. You aren’t talking about organic searches vs. adwords, are you?
Any clarification would be appreciated.
Hey Bob,
The search network means Google’s search results, like when you go to Google.com and type in a search query.
The content network is all the websites that run Google Adsense ads, including my blog. These ads are much less targeted but have more volume and are generally cheaper.
I am writing up a blog post right now discussing my second attempt at affiliate marketing, so check back soon!
Nice info there. I face a lot of trouble because CPC is so competitive these days, that often I really fall in a dilemna whether to risk the sum for a little margin of profit.
thanks!
yea i’ve had some difficulty getting into affiliate mareketing, but i think there’s plenty of room for competition still.
I just posted my second attempt at PPC to CPA marketing, check it out on the homepage.
Good luck…I’m learning affiliate marketing now myself and it’s a tough subject to learn. I’m doing an e-mail only offer right now and I figured it’d be easy to turn conversions, but no luck so far.
thanks. I’ve never tried email marketing, does it work well?
This is great. I just signed up with NeverBlue myself, and have been researching ways to promote a few offers via PPC. .09 CPC is brutal. I’m not sure how this could ever be profitable.. and how NeverBlue recommends these types of offers? Thoughts on that?
Did you try higher payout offers so you could raise your CPC?
Thanks!
hanji
Yea, neverblue seems to love that offer. I couldn’t turn a profit, but apparently a lot of people do.
Also, yea i’ve been trying a much higher payout. In the high 20′s. It has been much easier to turn a profit. Its really all about the niche though, with enough research i believe any niche can be profitable.
This is a great landing page. How do you iframe just a zip submit form like that? It’s not the advertiser’s landing page, is it? When i do the iframe, i’ts only show the full page of advertiser’s landing page.. Please help, thanks
That’s a nice landing page. I wonder how do you get those list of keywords? And how do you do the iframe just for zip submit area? I only know how to iframe the whole page from offer’s landing page… Can you share with us?
Thanks,
Martin
hey martin, you have to use css to hide the rest of the webpage. wrap the iframe with a div and set the width and height. it is the advertiser’s landing page that is iframed.
I still don’t get it. Can you show me some quick example? I’d really appreciate it.
Also I just subscribe to your post. Looking for more showcases…Keep up good works.
hey martin, thanks for subscribing.
wrap your iframe with a div tag, i can’t seem to comment with code so i can’t paste what the div tag looks like.
then, the CSS looks like this:
#container{
width:425px;
height:115px;
border:none;
overflow:hidden;
margin:auto;
}
#container iframe {
width:500px;
height:500px;
margin-left:-5px;
margin-top:-325px;
border:0 solid;
}
you have to play around with the width and height till you get it to look how you want. let me know if this works.
It works well, I just need to play around with it a bit. I’ll update the comments when success. Thanks, I’m thinking about running blog too now. Cool..
nice. glad it works.
Hi,
This article is great, especially for people like me starting with this business, I really appreciate this kind of precise information, like the affiliate program, the landing page…
thanks!
The iframe trick is nice & neat; however, once the ZIP is entered, how can the user continue filling the merchant’s form in the tiny iframe window, especially if scrolling is set to “no”?
hey greg, that was actually one of the problems i encountered. but i ran it by the aff. network and they were ok with it, go figure.
What was your QS like and the min. bids on average?
hey greg, my QS was usually great and my min cpc’s were around .04
Excellent article and very helpful. So, going back to the iframe size question from before…did you just leave the page this way while it was live? In other words, do you feel it’s ok to just leave the iframe window that small size even though the user won’t really be able to continue filling out the rest of the form? Just wondering if there is some other way to do this.
hey james, there’s not really another way to do the iframe. you just have to ask your affiliate manager if this method is legit with your network. thanks for the question.
You can do a frameset instead of an iFrame, and it will fill 100% of the screen and you won’t have to worry about the iFrame size.
replace all ( and ) with the appropriate HTML brackets.
[code]
(frameset rows="100%,*")
(frame src="http://your-affiliate-link")
(/frameset)
(noframes)
(body)
(p)
Browsers with noframes support will only see this.
(/p)
(/body)
(/noframes)
[/code]
Should do the trick…
thanks for that tip.
Matt – Great Post ,, Thanks for the information. I was wondering whats going to happen when i go live with my first campaign.
Jonathan — Great comments. I will be checking out the tools and affiliate link you threw up.
BTW Matt can i shamelessly plug my website here as well ????? Just kidding
tony, thanks. let me know how your first campaign goes.
Tony Gatto/Matthew Berman can’t blame the guy for getting backlinks. Its always been my opinion if someone contributes something they should be able to get respect at least in return.
Good write up i was on the fence as far as email/zip submits go.
And that tares it.
thanks, glad i could help.
Matthew,
You posted this last June. How is your CPA stuff going now? Looking at starting some campaigns and want to find someone having some success.
Thanks,
Peter
hey peter CPA is going ok right now. i did really well last month and have just been testing since. what are you trying?
Hey Matthew,
Thank you so much for your input with this article. Not many people are nice enough to share their experiences.
I am not a very good web developer however have really good graphics experience.
The iframing or frameset for only the zipcode area is really messing with my head. I read all the comments above but still cant get it to work.
Is there anyway you can send me an example landing template you have so I can study it or teach me? Thanks.
I like how clean and simple your landing page looks and I am a little bit confused and disheartened at the lack of response for a consumer discount offer in this tough times we’re in.
Hi there,
Thanks for the srticle it is great. I am interested in CPA marketing, but dont know how to start. Can you plz advise. Secondly I dont have an income, but I have a great network of contacts. Is it possible to start by just sending mails to folks and getting them to take the neccessary action? Or do I have to have a landing page?
Would appreciate your tips on this.
Thanks,
Sheen
hey sheen, sorry for the late response…you do need to have some sort of capital to start affiliate marketing. you may way to use a credit card for the first few weeks till you build capital…but be careful that you don’t spend too much that you can’t pay back.