CPV流量 個案研究

I actually forgot to post this up for you guys, it's meant to be a one week case study on CPV traffic for an offer however this quick writeup and stats on the first day test should give you enough insight to understand more about successful CPV campaigns.
( G2 ]7 J- u6 I- B0 y) M5 Q' H
The offer we're talking about is https://www.buyregeneratemd.com/index2.html This is one of our internal offers and the CPA we pay on it is $40

9 |* ^0 x2 w5 gFirst we need to touch on a couple key points I keep mentioning for anyone new here or those who may have missed other posts I've made in the forums:

3 i) p$ E' {) i, [! r# e0 V   1.

      CPV traffic IS definitely harder to convert for anything that requires a credit card, but certainly not as hard as some people like to claim. Interruption marketing is fantastic for targeting but is really most effective in lead gen and the less fields needed the better. With that said, I still have multi thousand dollar days each week with risk free trial products since the cost of entry is so low, but email submit, short form, etc basically quadruple the conversion rates. I've also found that cost per sale can work with CPV but you have to be really smart about the offer. I will be showing you two offers that require an actual purchase which have done extremely well for me but I've also tested almost 300 other cost per sale offers which tanked, again because people aren't quick to pop a credit card out with this type of offer serving.
   2.

      You have to track and most people continue not to with CPV. Much as I hate to place a pixel for a company I'm positive steals campaigns I've got two choices, make a lot of money which may not last long, or make no money. I'd rather make a lot of money which slowly starts to drop off. If you aren't placing the tracking pixel from your CPV network then you don't know which targets are generating the conversion. People tend to treat CPV as a second class citizen because of it's low cost. You don't want to fall into the trap of the 「If I spend $100 I'm happy as long as I'm making $110」 That's loser thinking and with five minutes of extra work you could have made that 10% profit 40% the next day by eliminating the losing target but now you have no choice but to keep spending on all those targets and hoping for the best.
' z7 g3 c% I9 O, w9 \; Q) X8 z
. d0 g0 l# V( s% v  H3 |; }3 \9 ~3 CFirst I'm going to get into the stats, and then the breakdown:
8 W9 H% `2 ^; R% E: W8 J( q/ |  G; `
4 b/ m: {4 ?9 m) UHere's the TrafficVance campaign, 27 targets were used in the test, we were served about 4700 impressions at a cost of $46.94.
4 [2 I7 E4 X; U2 f
Here's the Elite ROI stats for the day, 3 conversions, the CPA is actually $40 not $38, our campaign manager made a mistake when putting it up so we actually made $120 for the day.

% o+ I$ j3 d+ g& f, e- XIf you do the math, the profit there was $73.06 which is pretty good. Keep in mind I'm not showing you this campaign because it was a success, it just worked out that way because of how I set it up, had it failed I would have been able to show some other efforts in optimization which I'll now have to do on a later campaign demo.

. H. J2 M/ X  D- {, RMore importantly, those 3 conversions only came from 2 targets which of course I'd have never known if I wasn't tracking. While most sloppy marketers would be thrilled to pull a $75 profit for spending $45 a pro is more concerned with slicing off any wasted money.
. N8 o# ~" t' X: h+ Y
25 out of the 27 targets generated nothing for me however they all received significant traffic which I'll get to in a moment and that's why I really had no reason to run those same targets the next day. If a target gets a couple hundred views with no conversion we typically pause it in favor of new targets and will revisit it later.

. E; ?1 l0 u( \& a. @+ {& |! WThere's really no 「formula」 only personal interpretations. I've heard many of the guys who call themselves 「experts」 in CPV say that you should run a target until it at least gets to half the CPA or more. Personally I think this is a huge waste of money, to spend $20 or more on one target for this offer is retarded, why the fuck would you do that with thousands of other potential targets to test?
) c( ~1 I4 b. {  Y3 J% \
, C# j% O% n& H# ^I do a lot of what I call 「revisiting」 I'll pause a target after a couple hundred views in favor of new tests and when I'm satisfied with completing those I often come back and run another few hundred to the list just for the hell of it, even still, I never get close to 「half the CPA」 that's like the losers who keep chasing the break even point in blackjack when they're down $1000.

2 K6 q5 B" ^% Q1 m- Y4 m# Q0 ZCPV traffic isn't about cost, it's about conversion like anything else, if 1000 people who should be highly targeted saw your offer and weren't interested then what does it matter if it only cost you $10 so far? Spending another $10 or $20 just because one conversion will break you even or even put a few bucks in your pocket has zero business sense.
  Z. d0 j6 R! o4 u3 K. P* v, p
Anyway, let's talk about the targets used.

I'm not going to give you the two targets that generated the conversions because they are very profitable for me on multiple offers but I will give you some of the other targets and the thinking behind it all so that you would actually add those profitable targets to your list today if you followed along.
! }/ }/ W3 J) x  b" a0 E- D
3 w! N: o2 n( b0 L; i, t% aIt of course, as always goes back to my favorite word, demographic. You need two things in CPV, demographic and targets with decent traffic.
3 m  h! D, a. F2 m( m. G; O
4 ?3 N. e7 [* FWe know our main demographic for Regen is Caucasian females 38-57, image conscious, typically professional with an income of over 45k per year. We know this either because we are experienced or because we asked our AM or merchant directly for this information.
( r0 C: I5 j- A! B0 n
4 z: E' o5 R1 A% h; L# b) ]That was easy, so now on to step 2, finding that demographic on sites with decent traffic.
' Z% f' }3 `. d  k- n5 U6 [
Is there an easy way to do it? Ehh, I think it's pretty easy but I prefer to do it the old fashioned way which is the same way I've been doing it for three years of buying CPV, the alexa list.

% n$ f9 D, ~1 `I'm not saying you shouldn't use scrapers and all that other shit if you want, but my brain does a way better job than any keyword search and believe me, yours does to. The good news is, you can really compile your main lists once in spreadsheets and not have to do it again.
+ o( j2 Z4 s# G9 ?6 v
Up until early last year I was paying thousands of dollars for this list twice a year like a lot of other media buyers, I really don't think the gravity of the fact that it's free now has been appreciated enough.

While alexa's numbers are pretty bullshit in a lot of ways, they are still good enough to make it one of the best sourcing tools for media buying and traffic I've ever come across and a lot of guys agree with me.
' Q& k) H& b: B
It's now updated daily and you can grab it from this link http://s3.amazonaws.com/alexa-static/top-1m.csv.zip

I suggest using it the old fashioned way for a lot of reasons, the main one being that you will find thousands of sites you never knew existed that have huge traffic and are perfect for so many offers.

9 R! T1 S+ h; G" y" zIf you haven't caught on, I want you to start at number 1 and go to at least 50,000. This might take you a week or two, maybe three, but I promise you if you haven't done this before you will tell me it was one of the best things you've done in perhaps your whole time as a marketer for the ideas and energy you will walk away with just in the first few days alone.

Here's some of the test targets for the Regen campaign:
1 ]8 a0 E" p7 U/ R! k, Z
. M2 e! J' z0 J( k/ R7 _http://www.usmagazine.com
2 C& m5 C/ i" L6 e2 O$ I+ j$ U0 ^http://www.peoplemagazine.com
http://www.righthealth.com
% \% K" G. s! h# ]: E& zhttp://www.cooks.com
http://www.cvs.com
http://www.bedbathandbeyond.com
http://www.qvc.com
http://www.foodnetwork.com
http://www.barnesandnoble.com
. G# P2 w) C2 Y4 D- G# [) o; P
1 }$ ]) h4 z& x4 z2 j, G9 [As you can see, what they all have in common is that they all serve the target demographic for the campaign and all have high traffic.

. `6 j7 C3 B) G! v5 w' t# ?$ [This is what heavily increases your odds of success on every test. Most people just use a URL harvester and drop in piles and piles of targets hoping for the best and of course they don't even bother tracking.
' ~& W# t& L$ ^% j* A" I; V! U& {" b
) c5 b' ?7 r. D( P6 A/ J+ c" |" aBy being intelligent in my target choices (you know, doing some actual work) I see more winning campaigns than losing campaigns in CPV compared to most, it's always the people that want something for nothing, the whole copy paste click that always lose their money and give up.

What else is important to mention is that those targets I just gave you weren't the ones that were profitable for Regen but all of them are profitable for me for multiple other campaigns.

5 l) s, O8 H/ z' P( u1 x: vI have a belief that I'll share with you, and that is a target like a barnesandnoble.com or a foodnetwork.com just can't be a dog, it can't. There is SOMETHING that is going to do well to that traffic, it's just too good. What separates the winners from the losers is the people who work to find it.

$ y1 m" L. O0 B9 A/ F+ YMaybe the offer is totally different than what you'd expect to do well like the snuggie for dogs, maybe you need to make your own email submit, maybe you need to offer a free ebook on recipes to build an email list off that target and monetize it by marketing to it or renting ad space for it. There is however, always something that will work.

One thing I've mentioned before is that I typically rotate dozens of offers to the same targets. I didn't do that this time because I wanted to show you a case study similar to what you guys starting out would be successful with and with a smaller cost but if you have the cashflow there is no reason not to test multiple offers to top targets.

半夜竟然沙發，哈。好好拜讀下先……

看的懂洋文的真好啊！

慢慢看吧！！！

慢慢看　　感謝幾何了！

收下，慢慢翻譯

感謝幾何老大，慢慢翻譯看

感謝幾何老大，慢慢翻譯看

真高端！！！

完蛋~~  純英文的  找GG去了