A Complete Idiots Attempt at Media Buys

[MrSmith]

Hi,

Background

I am a complete Newbie but have read a few threads but I really only learn by doing it myself. The only threads I pay attention to are other journeys/case studies as they generally show where things go wrong rather than a tutorial on something that in theory should work. I prefer the real world examples.

So keeping in mind I have absolutely no experience I will make many mistakes and hopefully learn a few things. My budget isn't great but around $5-$10 per day but there is no use risking lots of money if I don't know what I am doing.

So here are a few things I have done :

1. Found a few offers worth promoting. I picked 2 offers to test. But have a few others as options if necessary.
2. Picked tracking software and set it up.
3. Created a few simple banners. I have created 3 main banners which vary in colour, images and wording. The variations on these add up to 11 banners in total. I am only testing 3 banners to start with which are vastly different and then will split test the variations of those banners.
4. Decided on a traffic source which is going to be direct media buys of ads on websites.
5. Purchased a banner ad space at the top of a forum cheaply.

Actions to still do:
1. More research into sites. I need to determine what to keep, what is important, etc.
2. Buy more ads to do more testing.


My approach
So while I am a total newbie at all of this I am very logical in my approach to the numbers. I love to delve into the numbers and find angles to use.

So here is the logical way I have approached this:

Offer 1 : Payout is $9.95 per sale (PPS).
It says it has a coversation rate of 9%. That means I should make a sale approx every 12 people. To test this properly I would need to send at least 3-4 times that amount of people to the offer which would be about 48 clicks so lets call it 50 clicks.

So I have found a site (lets call it Site 1) which says the ad space is expected to get 0.75% CTR. I'll assume that CTR is possible for a good IM marketer, which I am not, so lets assume 0.05% as worst case. At 0.05% I need roughly 100,000 impressions for my test.

The site gets approx 600 impressions a day so that is 166 days I would need to do a test at worst CTR conditions. The cost of the ad placement is $2.80 for 30 days. That means I would need to get 6 placements for the ad which is $16.80 in costs.

This seems low risk to me. I am risking $17 to gain some data and learn the ropes. I have also read spending approx 2-3 times the offer is a good test.

The Things I expect people to say:

Q: Why not do a RON to do the test?
A: I see lots of newbies do that and lose money fast.

Q: 166 days of testing is too long.
A: I agree so I will be doing this over a few sites but only have 1 site selected at the moment. Also 166 days is worst case and if I can achieve a higher CTR then 0.05% then it will be much shorter.

Q: You have no idea what your doing !!!!
A: Yep exactly right but I'll learn by doing.

Q: Your budget is too small to test.
A: Maybe but why risk big money when you don't know what your doing.

Stats so far
I purchased the site ad space with a few hours left to go yesterday so my initial stats only cover a few hours.

Impressions: 92
Clicks : 1
Conversions : 0

I welcome all and any comments both negative and positive.

 

[game333]

Keep hammer yourself onto a brick, you will learn more than those who seek for tutorial, at least I'm. Goodluck.

 

[AF-Roger]

Welcome to AffiliateFix. We look forward to reading more about your journey.

 

[azgold]

You've certainly put some thought into this! That's really good.

Nice to see you going in with such a positive learning attitude.

When you decide you're ready to venture a little farther afield and spread your wings and advertising dollars, check the site's stats before purchasing ad space (Alexa, SimilarWeb, PR, traffic, etc.). A really highly-targeted, high traffic site will definitely cost more but the potential ROI is also much higher. You don't have to jump right into the $1,000/month slot, work your way up as fast you're financially able to, assuming you're getting a good handle on how to do this.

Also, make sure that you give the site a good look-over, see what other kinds of ads they're running, demographics of the commenters (at least the type of people they seem) and the typical content topics. Then design your ad to fit in yet stand out. Confusing?

You don't want to try to sell car insurance to bicycle enthusiasts, or have a graphic game ad on a serious, professionally designed business site. Makes sense, right? Align yourself with the site's audience and go in with your best test ads.

It isn't hard, honest. Just takes some practice and I have a feeling you'll do just fine.

Not every campaign is going to be profitable but I think you know that.

Like @AF-Roger said, we look forward to reading more about your journey.

 

[MrSmith]

You've certainly put some thought into this! That's really good.

Nice to see you going in with such a positive learning attitude.

When you decide you're ready to venture a little farther afield and spread your wings and advertising dollars, check the site's stats before purchasing ad space (Alexa, SimilarWeb, PR, traffic, etc.). A really highly-targeted, high traffic site will definitely cost more but the potential ROI is also much higher. You don't have to jump right into the $1,000/month slot, work your way up as fast you're financially able to, assuming you're getting a good handle on how to do this.

Also, make sure that you give the site a good look-over, see what other kinds of ads they're running, demographics of the commenters (at least the type of people they seem) and the typical content topics. Then design your ad to fit in yet stand out. Confusing?

You don't want to try to sell car insurance to bicycle enthusiasts, or have a graphic game ad on a serious, professionally designed business site. Makes sense, right? Align yourself with the site's audience and go in with your best test ads.

It isn't hard, honest. Just takes some practice and I have a feeling you'll do just fine.

Not every campaign is going to be profitable but I think you know that.

Like @AF-Roger said, we look forward to reading more about your journey.

Thanks to both game333 and AF-Roger for the welcome/good luck responses.

And a big thanks to azgold for the detailed response. I am only starting small as you know but I have spread to a few sites now (total of 3 sites). Two were based on thorough research and the other was just to much of a bargain to pass up even if its mostly or all tier 3 traffic. My PPS allows for any country traffic, as they only care about the sale.

I have actually done my research along the lines you have said. I got the sites statistics and cross referenced them against alexa, and MOZ tools. I also checked into their keywords from there and checked that out on google keyword tool and have recorded all this data for each site. I expect after I have done 100+ sites this data may become useful. I have tried to find sites which are specific to the niche but also ventured a little off track with the keyword research to look for alternative sites worth pursuing.

Your comment on designing banners to fit in yet stand out is the exact issue I am playing with at the moment. I have a couple of banners designed to blend into the site and others which will stand out to split test. At the moment I have gone with those which stand out just on the hunch they may work better but its really me guessing till I have tested it. I suspect it depends on the niche and the visitors overall.

I am happy to spend $5 each day spreading out to new sites and trying new ideas and new offers and just seeing what happens at the end of a few months. I'm also thinking about what will work well in the new year such as fitness, health, money related niche, etc as people take action in those first few weeks of January on their new years resolutions. So I may target around those starting now to get the ads in place with the knowledge they may not really pay off till next month.

I'll update my stats later but I am up to 3 sites now. One of the sites allowed me to place add space on their site for 1c for 30 days. Yes 1c. All of their traffic was from brazil though so I may need to target with some spanish? ads. But for 1c for 30 days I just had to buy it.

I'll be back in a few hours with the daily stats up, with the good news that I have been getting some clicks but at this moment no conversions yet.

 

[newlxx]

Good luck and Subscribed.

 

[azgold]

Thanks to both game333 and AF-Roger for the welcome/good luck responses.

And a big thanks to azgold for the detailed response. I am only starting small as you know but I have spread to a few sites now (total of 3 sites). Two were based on thorough research and the other was just to much of a bargain to pass up even if its mostly or all tier 3 traffic. My PPS allows for any country traffic, as they only care about the sale.

I have actually done my research along the lines you have said. I got the sites statistics and cross referenced them against alexa, and MOZ tools. I also checked into their keywords from there and checked that out on google keyword tool and have recorded all this data for each site. I expect after I have done 100+ sites this data may become useful. I have tried to find sites which are specific to the niche but also ventured a little off track with the keyword research to look for alternative sites worth pursuing.

Your comment on designing banners to fit in yet stand out is the exact issue I am playing with at the moment. I have a couple of banners designed to blend into the site and others which will stand out to split test. At the moment I have gone with those which stand out just on the hunch they may work better but its really me guessing till I have tested it. I suspect it depends on the niche and the visitors overall.

I am happy to spend $5 each day spreading out to new sites and trying new ideas and new offers and just seeing what happens at the end of a few months. I'm also thinking about what will work well in the new year such as fitness, health, money related niche, etc as people take action in those first few weeks of January on their new years resolutions. So I may target around those starting now to get the ads in place with the knowledge they may not really pay off till next month.

I'll update my stats later but I am up to 3 sites now. One of the sites allowed me to place add space on their site for 1c for 30 days. Yes 1c. All of their traffic was from brazil though so I may need to target with some spanish? ads. But for 1c for 30 days I just had to buy it.

I'll be back in a few hours with the daily stats up, with the good news that I have been getting some clicks but at this moment no conversions yet.

You're doing awesomely! Love how you're looking ahead and planning, too.

Re the Spanish ads, yes use them for Brazil.

Re 1c for 30 days - wow! It will be especially exciting if you get some conversions.

 

[MrSmith]

UPDATE

Well first day down on my journey and its time to put up the stats for yesterday:

Site 1: 603 impressions 0 clicks. Cost was $2.80 for 30 days.
Site 2: 1746 impressions 1 click. Cost was $3.88 for 30 days.
Site 3: 35,260 impressions 4 clicks. Cost was $0.01 for 30 days. Yes 1 penny.

The CTR for the day was 0.01%. So much for hoping for 0.05%.

Including today there have been 6 clicks. 0 conversions.

LEARNINGS
The first two sites target US traffic as the main visitor with a bit of tier 2 traffic as well. The third site although had 35K impressions targets tier 3 visitors. So while the CTR is poor if I get even 1 conversion out of this I will be ecstatic. It will be a good testing ground.

My CTR is atrocious but that was to be expected. I will leave the banners as they are at the moment for about 3-4 days and work on a few other banners which might work.

THINGS STILL TO DO
Purchase some more ad space to get this test going across a wider range of visitors.
Work on new banners for use in the future.

Well I'll get back on later with some site research if I get time.

 

[game333]

Why don't you test if these people will get into action, what I mean is.
You're dumping your visitor to a sale page, and you wouldn't know nothing than the buyer bought a stuff, you got a sales.

So, you should try build a email capture page, to know that your visitors really did interact with your product rather than the blind stat of CTR.

Just my opinion, just some clear idea to give a clear picture of your stat.

 

[azgold]

Is there a lot of bot traffic coming to the 1 penny ad?

 

[Dispply]

Is there a lot of bot traffic coming to the 1 penny ad?

looks like this. I bet you should blacklist 3rd site. either it's bot or redirect traffic. Anyway no good.

Try playing with bids as well.

 

[azgold]

looks like this. I bet you should blacklist 3rd site. either it's bot or redirect traffic. Anyway no good.

Try playing with bids as well.

I suspect you're right about the traffic BUT this is a media buy, so no bidding. @MrSmith found a 1 cent for 30 days of advertising deal to practice and learn with.

 

[MrSmith]

Why don't you test if these people will get into action, what I mean is.
You're dumping your visitor to a sale page, and you wouldn't know nothing than the buyer bought a stuff, you got a sales.

So, you should try build a email capture page, to know that your visitors really did interact with your product rather than the blind stat of CTR.

Just my opinion, just some clear idea to give a clear picture of your stat.

That's a really good idea. Thanks for it.

 

[MrSmith]

looks like this. I bet you should blacklist 3rd site. either it's bot or redirect traffic. Anyway no good.

Try playing with bids as well.

Yes it could be some bot traffic. I wouldn't be surprised but I paid to advertise on the site as azgold said. It cost me a whole penny for 30 days. Its getting some clicks and I'll use game333 idea to confirm the traffic from the clicks. Considering its a media buy I basically have no other costs for that ad for 29 more days so there is no real incentive for them to click on the ads so I suspect the clicks are real people who are semi-interested in the offer. I just need to work on converting them.

 

[ppcnewbie]

Hi MrSmith (your name reminds me of the Brad Pitt & Angelina Jolie's movie "Mr. and Mrs. Smith"),

Interesting journey, I just added it to my Watch list.

I have a question and a suggestion.

The question: how did you buy the ad space? Directly or from a service like buysellads?

The suggestion: If you're direct linking your banners, indeed, you're losing all the info about your visitors and basically, you won't know anything about them, just that they converted (when they will).

Problem is, especially since it's a PPS offer, usually it's mandatory to have a landing page for pre-selling them.

So, depending on the offer, use your creativity (or spy) to come up with 1-2 landers to test.

If you want to get their emails, as suggested, even better. If not, you still have the option to add a Google Analytics script to track the stats, eventually use custom events for tracking different actions on the landing page. Plus, if you plan on running a lot of traffic later, you'll get the option to retarget these visitors on FB by building a custom audience by simply placing a 5-line script on your page and you're done. After gathering lots of data, in FB Audience Insights you'll be able to see different stats about the visitors. Even if you don't plan to promote this offer on FB, it's still good that you get some demographic data.

Later edit:

There's some guy here on the forum, forgot his name, that mentions in lots of his posts that it's important for marketers to try to understand their audience also, not just analyze some numbers. And it makes sense, because it's important to know who's your ideal customer and with what types of messages you could make him convert.

This is even more important if it's a Cost Per Sale offer. Here, obviously the conversion rate will be much lower compared to SOI offers, so obviously more traffic to be sent, and again, makes sense to try to find our as much as you can about your visitors/audience.

 

[azgold]

Hi MrSmith (your name reminds me of the Brad Pitt & Angelina Jolie's movie "Mr. and Mrs. Smith"),

Interesting journey, I just added it to my Watch list.

I have a question and a suggestion.

The question: how did you buy the ad space? Directly or from a service like buysellads?

The suggestion: If you're direct linking your banners, indeed, you're losing all the info about your visitors and basically, you won't know anything about them, just that they converted (when they will).

Problem is, especially since it's a PPS offer, usually it's mandatory to have a landing page for pre-selling them.

So, depending on the offer, use your creativity (or spy) to come up with 1-2 landers to test.

If you want to get their emails, as suggested, even better. If not, you still have the option to add a Google Analytics script to track the stats, eventually use custom events for tracking different actions on the landing page. Plus, if you plan on running a lot of traffic later, you'll get the option to retarget these visitors on FB by building a custom audience by simply placing a 5-line script on your page and you're done. After gathering lots of data, in FB Audience Insights you'll be able to see different stats about the visitors. Even if you don't plan to promote this offer on FB, it's still good that you get some demographic data.

Later edit:

There's some guy here on the forum, forgot his name, that mentions in lots of his posts that it's important for marketers to try to understand their audience also, not just analyze some numbers. And it makes sense, because it's important to know who's your ideal customer and with what types of messages you could make him convert.

This is even more important if it's a Cost Per Sale offer. Here, obviously the conversion rate will be much lower compared to SOI offers, so obviously more traffic to be sent, and again, makes sense to try to find our as much as you can about your visitors/audience.

I believe @MrSmith has tracking installed already to collect data.

And yes, understanding your audience and getting them onto your list is tremendously helpful. MrSmith said he's using this buy to learn and that's what he's doing.

Buying a month's worth of advertising for a penny so you can get your feet wet, is awesome IMO. Can't get much (any) cheaper than that!

 

[MrSmith]

Hi MrSmith (your name reminds me of the Brad Pitt & Angelina Jolie's movie "Mr. and Mrs. Smith"),

Interesting journey, I just added it to my Watch list.

I have a question and a suggestion.

The question: how did you buy the ad space? Directly or from a service like buysellads?

The suggestion: If you're direct linking your banners, indeed, you're losing all the info about your visitors and basically, you won't know anything about them, just that they converted (when they will).

Problem is, especially since it's a PPS offer, usually it's mandatory to have a landing page for pre-selling them.

So, depending on the offer, use your creativity (or spy) to come up with 1-2 landers to test.

If you want to get their emails, as suggested, even better. If not, you still have the option to add a Google Analytics script to track the stats, eventually use custom events for tracking different actions on the landing page. Plus, if you plan on running a lot of traffic later, you'll get the option to retarget these visitors on FB by building a custom audience by simply placing a 5-line script on your page and you're done. After gathering lots of data, in FB Audience Insights you'll be able to see different stats about the visitors. Even if you don't plan to promote this offer on FB, it's still good that you get some demographic data.

Later edit:

There's some guy here on the forum, forgot his name, that mentions in lots of his posts that it's important for marketers to try to understand their audience also, not just analyze some numbers. And it makes sense, because it's important to know who's your ideal customer and with what types of messages you could make him convert.

This is even more important if it's a Cost Per Sale offer. Here, obviously the conversion rate will be much lower compared to SOI offers, so obviously more traffic to be sent, and again, makes sense to try to find our as much as you can about your visitors/audience.

I have a list of media buy services such as buysellads which I have compiled over the last few days. Its just a matter of going through them to find ad space which I am happy to use for a reasonable price. I find it takes a while to go through and do the research on the sites and it can be hard to find some bargains. I have also sent out some emails to other websites asking for a price. Some have replied and we are in negotiation on the price. In fact the ones which have replied are those where I have identified areas they could add banners but they are currently not doing that. This is opening a lot of doors as they are more open to negotiation as I have helped them.

I am not direct linking but I wasn't collecting emails. I still haven't set this up but hope to do that over the weekend.

In regard to the 5 line script you are talking about for retargeting. Where can I get that script ? or some information about it ?

The advice on knowing your audience is a good one and is basic common sense as well. You need to know who you are targeting in order to have the best chance to sell to them. I have a target audience for my offers and the ad spaces I am buying are aimed at those audiences but is aimed world wide instead of just US visitors. I may narrow this down as I get more data.

Thanks for the great reply.

 

[MrSmith]

Just a few thoughts going around my head

The concepts behind IM is the same as any other business whether it be online or offline. You need to have something to sell. You need people to sell it to. Getting the product to sell is easy enough but finding the people to buy it from you is the hard part.

If I look at this like an offline merchant. I could get a product to sell (lets assume I get a commission for each sale) and I could go the "free" route knocking on peoples doors and trying to find someone who might like it. While this is free it takes a lot of time. The same thing happens with online which is where I could build a free website and generate free traffic but it takes time and effort. The alternative to this is to advertise my product on the tv, radio, with flyers, etc. These all cost money but I can reach a much larger audience and have people come to me to buy the product. I can do something similar in IM online with a RON campaign. Another option would be to purchase advertising billboards in certain parts of the city where I target people with the incomes to purchase to my product. Trying to target the advertising is fairly obvious online and most people have an idea how to do this already (I am using media buys).

In the end though it all comes down to 2 things.

• Cost
• Income

These two numbers turn into either a profit or a loss.

Cost can be controlled by the amount of money we spend on bids, ad spaces, etc. After enough data is collected we can start to make much better predictions around the CTR, CPM, CPC, CR and profitability of an ad space. Initially we have no idea and this is why people say the first part is buying data. You need the data to be able to determine what is possible.

Once we have data we can try and modify and optimise the CTR and CR in order to increase the income. We can try and minimise our costs. If we can't do either then there is no chance of turning a losing campaign into a profitable campaign.

How I think this affects my journey here

I need to keep my costs to a minimum in the learning stage as I do not have the data to make accurate judgements on the income I can produce with the offers I am promoting. I am doing this by targeting niche sites with media buys based on what I think should work. Perhaps right here is a problem. What I think may work may not be what works. I'll find this out over the next few weeks. A quick solution to this problem is to maybe do a quick low budget RON for a few products to see what is converting into sales.

Anyway just a quick rambling today to get what I am thinking out of my head and onto the forum for others to read/comment and perhaps see what they think.

At this point still no conversions from my ads but the clicks are coming through so hopefully one of them converts eventually.

 

[MrSmith]

When do you have enough test data?

So I was just trying to determine my cutoff point for testing offers. I had a good think about this and I think I’ll be doing it using the following method.

• Offer 1: Conversion rate of 9%
• Offer 2: Conversion rate of 20%.

For Offer 1 I have a 91% chance that I will not make a sale off a click. Using this logic I can work out the following after 5 clicks.

• After click 1 : 91% chance of no sale
• After click 2: 82% chance of no sale
• After click 3: 72% chance of no sale
• After click 4: 68% chance of no sale
• After click 5: 56% chance of no sale

I can work out each percentage with simple maths so 5 clicks is 0.91*0.91*0.91*0.91*0.91 = 56%.

Now at what percentage do I say the offer is a complete failure. It wouldn’t be at click 5. I think it needs to be at < %10 chance of sale or even < 1% of sale. This would 24 or click be click 49.

For offer 2 it would be completely different as it has a much higher conversion rate with only an 80% chance of no sale. So the same table is below for it for 5 clicks:

• After click 1 : 80% chance of no sale
• After click 2: 64% chance of no sale
• After click 3: 51% chance of no sale
• After click 4: 41% chance of no sale
• After click 5: 33% chance of no sale

For this product the 10% chance of no sale (or 90% chance of a sale) would be after 11 clicks. The 1% chance of no sale or 99% chance of a sale would be after 21 clicks.

Generally you will see products with a higher conversion rate are much cheaper and therefore rightly need less clicks to get a sale as the profit per sale is much less.

 

[azgold]

@MrSmith your data analysis should also tell you about your audience - where most of them are from, what time of day they do the most click throughs, what devices the majority of them use, etc. (whatever variables you track for a campaign). All information to help you tweak your campaigns, not only for this run but it can give you a starting point if you move that or a similar campaign to a different platform in future.