Let's Talk About Research (12)
Ok, let's talk about research. After reading the thread about the guy with wife's whos back went to shit and he was quite desperate, smax suggested that he would learn the target demographic inside out.
Personally, I'm failing here so I want to make this follow along a bit different. Instead of focusing on a campaign, let's focus on finding out different ways to drill into your demographics. I know this might sound somewhat giving out your competitive advantage but the way I see it, only people who are really competitors are the ones who put in the work and just by chance happens to pick the same niche you have - so it's actually quite unlikely but will still happen no matter if you share 
So without further ado, let's start researching.
-------
This is what I do which isn't working all to well so instead of copying, try to figure out why it's failing and post your insights
Niche: Dating christians
Go to POF AffSpy and get all ads containing "christian" (doing POF right now so...)
- problem 1: You can't know what other ads were tested and was it the ad/lp combo that didn't work or the ad.
- problem 2: When you "click" the ad, are you getting working LP or LP that's under a test
- problem 3: You don't know anything else about the profile than area (ad copies will tell you that), age and gender .. so you can't really know if the ad was targeted for fat chicks from new york and then you try to run it for all and get shit conversions.
- problem 4: In a nutshell, you don't know the full picture
Visit offer landers and try to figure out from their copy what converts
- problem 1: These are usually so general that no use, unless you consider "Find Christian Love" as a great headline.
Brainstorm a profile
- problem 1: The fact is that I'm not 43 year old single woman whos life is revolving around 2000 year old imaginary fairy tale (did I just say that out loud... sorry)
- problem 2: What is she looking for? New husband? Some good old christian spanking time? Honestly, you can make a guess but it's likely to cause your campaign to be everything for nobody.
- problem 3: 43 year old single woman from some small city in Texas is a bit different cougar than 43 year old single woman in California .. you can tap into 8 life-forces but I still feel that you are not pushing the right buttons.
Reading forums for their problems
- problem 1: You might get good stuff here for brainstorming but this type of research takes a lot of time.
- problem 2: If your niche is a bit tighter like let's say "bb" christian divorcee women, you are going to have a bad time finding those people from forums.
---
So, that's the way I do it. I want to do that awesome research with a lot of facts but where to get them and more importantly, is it worth the time? I think that's the question I would like to get answer from our professionals. Is there some general key factors you need to know or do you really need to go deep? And how do you justify the time spent on doing that?
This helps you target a much tighter demographic rather than " 43 yr old Christian Women in some small city in texas" so ultimately you get a much higher converting demo but at the same time, the tighter it gets, the harder it becomes to profit at a long term..
Generalising the demo problems get you much farther in the long run, here cones the idea...there MUST be a common mentality of "43 yr old women" and may be thats as you just said TOO LOUD 
- living in a 2000 yr old story !!( and thats true with atleast those 18-20 yr old girls on who i tried this, real time )
You can target the general demograpgic in ur own unique way, many might not work, but the one that does-GOLD FOUND...
Originally Posted by sean3
Generalising the demo problems get you much farther in the long run, here cones the idea...there MUST be a common mentality of "43 yr old women" and may be thats as you just said TOO LOUD
- living in a 2000 yr old story !!( and thats true with atleast those 18-20 yr old girls on who i tried this, real time )You can target the general demograpgic in ur own unique way, many might not work, but the one that does-GOLD FOUND...
There's always a theory what makes something work. Some people know the theory and win because of it, some people have experience & they don't know they know but intuitively make right decisions, and some people just lose because they have no theory or experience. Based on writings here, it looks like people are going for getting experience route much often than focusing on compiling a theory. To me the common way of getting experience is like learning to fly by looking how bird fly, strapping feather wings on my back and jumping from the cliff. Why people can fly is because few smart folks understood the theory of flight and physical laws that are in play. Turns out you don't need flapping wings with feathers even 100% of other flying animals have them...
Now this is not to say you don't need experience, it's more like saying that I believe successful affiliates have certain models in their heads (and really smart ones have them on paper *cough*checklists*cough*) and that's why their average winning campaign ratio is much higher than average aff's.
Originally Posted by qhead
I have experienced immense success in scare tactic, so I scare my visitors first with the LP and they tend to convert really well.. I have replicated this model in almost every niche you can imagine off !!
About the Science, i highly recommend you read " predictably irrational ", many such stuff like above are well illustrated which im sure will amaze you..
i dont think there is a spy tool can make u profit instantly.
what spy tool do is provide u some ideas,angels and demographic.
i think we dont need to find the exactly profitable ads, if affiliate marketing like that, every single member here is gonna be millionair.
so what i do is ignore the problem, and stop seeking the "copy & paste then profit" job, and begin find a profitable angle by myself.
and yes, i did found some easy money thru spy tools, but what make me the most profit is from my own idea.
let me said it like this: we are not expert for marketing research( do you?)
the best way to do marketing research: START SENDING TRAFFIC THEN ANALYZE BY YOURSELF
Originally Posted by sean3
Originally Posted by adgale
I know you can't reduce this to pure algorithm (or can you?) but that's not even the point of this thread
The point was to find out how other affiliates research their ideas.
--
Another research tactic is Quantcast and other website demographic profilers but I find these tools pretty inaccurate in so many ways. I think the best way to use them is to discover other similar sites that you might want to investigate for either media buy or PPV.
The best research you could do is to sign up for the websites, and start sleeping with older Christian women. This will gain you more insight than any other tools out there. The issue is when you start getting into niches like BBW adult hookups.
On a more serious note, I think you're overanalyzing it.
Lets say I'm doing this campaign 100% blind, like not spying on anyone. First I would direct link. It's a dating website, I wouldn't bother with wasting time on a landing until I know it's a winner.
Pick a simple demo, women 35-40 on POF. We're going to test one variable at a time, the most important is offer. It doesn't matter how brilliant your creatives or angle is if the offers don't convert. Split test like 5 of them, from DIFFERENT NETWORKS.
Ok now you got your winner. Next we're going to test your creatives. I usually focus more on Angles at the beginning. Once a good angle is found, then start testing the creatives. Get that CTR as high as humanly possible. Brute force it, test out 30+ creatives.
Hopefully by now you've found a winning combination of offer + angle + creatives. Whatever was the winning creative, I analyze WHY people liked it. Cute guy? A guy with a kid in the image? The guy is showing muscles? Basically if you have an awesome creative, you'll burn it out so you need to make more similar to it.
Next you can expand the age demos, 18-21 girls, 22-25, 40-50, blah blah.
Start testing direct linking vs. different landing pages
Get higher pay bumps
Test out other countries
blah blah.
Basically I don't research much in the beginning. Why? Because you start ASSUMING things without data.
It sounds like you're scared to lose money and you're trying to minimize it as much as possible. Most campaigns start off unprofitable, and it's through rigorous testing that it becomes a money maker.
It looks like this
Day 1: -100% ROI. Test Offers
Day 2: -50% ROI. Test Angles
Day 3: -25%. Test Creatives
Day 4: Break Even. Lower Bids
Day 5: +25% ROI.
Blah blah I'm simplifying it. But yea quit analyzing so much and just...launch, test, launch, test, launch
You only analyze after you have tests and data.
Good points. I come from analytics so I tend to overanalyze stuff but your point about assuming was good reminder.
Actually you answered one of my questions probably without even realizing it Now I have more or less tried to test angles, creatives and offers in day 1.
Well, I think I took your bruteforce advice a bit too literally (here's my black hat background) because the first thing I did with POF was to code an automation tool that uploads creatives, keeps 5 alive and shuts them down if my rules about CTR are not filled. Makes everything much easier because less manual hassle with the creatives but the problem is that I tend to do fuckton of different creatives and then it takes forever and a lot of cash to find profitable ones.
I think I'll start doing it by hand carefully trying to find out what works and why instead just throwing shit in the wall.
By the way, how do you guys keep your pictures organized? I have a lot of photos I have gathered from various places but organizing them is a bit problematic, for example how do you remember which creative you have tested and which converted great to who?
That was Boss advice dr_ngo. Simple but Boss.
Originally Posted by dr_ngo
So if you first test offers, how do you come up with creatives? You just write few out of your ass or get "inspired" by some spytool?What I'm getting at is that test offers first, fine, sounds logical BUT am I wrong if I assume that your headline and adcopy is going to affect on which offers wins? I mean if we stay with christian dating and I whip up few basic headlines to test "Meet Christian Men" and "Date Kind Christian Men". Meet headline SHOULD convert better with the offer that promotes meeting on their lander and date headline should convert better with the offer that promotes dating. Just your basic framing.
Or is the ad really irrelevant and in general no matter what the ad says (date, meet, msg, etc. nothing radical), the better offer is going to reveal itself?
I might be making this too complicated but I've just read too many copywriting/psychology books that highlight how humans are predictable that there must be an overall pattern which leads to average conversions. Not sometimes but all the times like valid theories do. And also partly I'm not looking for anybody to tell me exactly what to do, I'm more interested understanding the inner workings of everything
I rarely use spy tools for inspiration, more for finding new niches / offers I haven't come across. What happens is someone does Angle A, and everyone copies Angle A. So you go to the spy board and see everyone's using Angle A and you go with it because everyone else does. Instead if you went with a blank slate, you could've come up with much, much better angles. Seriously, most affiliates are lazy / stupid.
Yes there are certain patterns. I could enter a new niche and re-use headline "templates" I know work from older niches I've already done.
For example that (1) weird trick to ... was originally a diet headline, but it's been used on basically every niche. 1 weird trick to a bigger penis, 1 weird trick to lower insurance, etc.
I'll throw out another random CTR trick. You can target more niche than other people. Random example. One of my first campaigns 5 years ago was eharmony. Most guys just do the generic headlines, Want to find a girlfriend? ZZZZZZZ
Instead I did adgroups on the person's age. Ex I had an adgroup for 25 year olds. I made my ads TARGETED towards that person. Ex. "Looking for 25 yr old guys." I keyworded military guys "We want Military guys". Age is just one variable. But think about what targeting options you can have, and if you can call the ad out to them. (btw the age thing isn't allowed on facebook anymore)
How do I know what works? I've been doing this for 5 years. It's just like basketball, only way to get better at shooting the ball is to shoot the ball. I've also read almost every advertising / copywriting book out there: Caples, Cashvertising, ogilvy. Also it helps to have an outside perspectives. I always split test other people's ads against my own. Also I have a professional copy writer I hire sometimes to help me write ads. I don't know what it's like to want to lose wrinkles and be a woman, but maybe someone else does.
The ad, landing page, and offer all work in synergy. Some ads will make the offer convert higher. But you also have to balance it with a high enough CTR. Only way to know is have everything tracked. I keep it simple. Even split test on ads. Then you look at how much the ads spent vs. each other, and then look at how much $$$ was attributed to each ad.
Maybe it's better to put the books down and just launch campaigns. After you launch a few thousand, you'll learn more about human psychology than what any books can tell you.
Originally Posted by dr_ngo
Home > Affiliate Marketing Forum >