I'm working in online advertising since 1998 and finally decided to start my own blog.

Got a great performing banner? Duplicate it! Let me tell you how.

Hello guys, this is yet another article that was inspired by questions posted on the STM Forum. Coming up with topics to write about on my own, is a surprisingly hard task. So instead of trying to guess what you might be interested in, I thought it would make sense to just write about what people ask for. And since I spend a ton of time over at STM, many of the question will be sourced from there πŸ™‚

So let’s go back to the million dollar question :


We all know this situation, some banners simply work better than the rest. Doesn’t matter whether you made that banner, ripped it somewhere or your AM gave it to you (some really do this) … as soon as you got a banner like this, you SHOULD IMMEDIATELY duplicate it and make more variations from it.

Affiliate Marketing is a competitive industry, any advantage that you can get is like blessing, but you need to act fast. Having variations of well performing banners at your hands, ready to replace the old ones anytime you need that … this can be your competitive advantage and your ticket to Profitopolis!

There is one small problem thou, as usually, replicating winning banners is not easy. It’s actually quite a complicated task. You might wonder why, shouldn’t it be an easy task to clone something that already works? It’s not and there is a simple reason for that : you don’t know why exactly a particular banner works better than the rest. Figuring this out is the hard part, but I’m gonna tell you how to do it!


Every banner has some key elements that can have dramatic effect on it’s performance. Some are more important than the rest, but all of them can make or break a winner.

Let’s list them one by one, I believe there are 5 KEY elements that we need to look at :

1. IMAGE : every banner usually has some sort of main image. It can be a game screenshot when promoting game apps. Or a photo of some hot MILF when promoting adult dating. Or a photo of the product you are trying to sell.

2. AD COPY : this is the ANGLE you are using to sell something, so the text itself – the actual message you are trying to get through to the users.

3. CTA BUTTON : Call To Action button plays a big role too. It can come in various shapes or colors. It usually carriers a small text like : Try It!, Read More, See Photos …

4. LAYOUT : let’s call it the β€œtheme”, so the general appearance of the banner. You can have an image in the left half, adcopy in the right half, then CTA button at the bottom. Or the other way around, image in the right half, copy in the left, CTA on top … there are many ways to skin a cat.

5. COLOR SCHEME : You can choose mild or bright colors, use white or black background … sometimes it pays to use similar colors to the placements that you advertise on.

Based on my own experience, elements #1 and #2 are the most important ones, #4 can be a game changer too … #3 and #5 are rarely the ones responsible for exceptional performance, thou they can help it a bit too.

In order to be able to replicate a well performing banner, we need to identify which of the 5 elements is the one responsible for the exceptional performance first. Read on to find out how to do it.


The IMAGE used on the banner and the ANGLE (ad copy) are usually the ones that are responsible for it’s great performance. But it’s not universal, because there is a connection with the vertical you are promoting.

For example : Adult dating is a typical vertical where the image matters the most. But in case of utility app installs for example, the ad copy takes the lead. You will figure this out with experience, but for now, let’s just accept the fact that Image and Copy are the most important factors πŸ™‚

In order to figure out what element is the magical one, you need to run separate tests for every one of them.

Let’s start by testing #1 – the IMAGE. In this case, we are assuming that it’s the image that makes this banner stand out. So keep the image and make 5-10 new variations by changing the adcopy. We are leaving the rest as is. In case our assumption was right, you will end up with up to 10 banner variations that all work well. This way you have something to rotate or use as a replacement in case you need one.

If this first round failed and all the new variations sucked, it’s time for round #2. This time we will assume that the adcopy is our magic bullet. Keep the adcopy as is, prepare 10 different images and start testing your new batch of banners. Possible outcome is the same as in the first place, you will either end up with a batch of good banners or a batch of failures.

In case both of these tests failed, we have a problem πŸ™‚ The possible reasons are :

a) we picked very bad new angles
b) we picked very bad new images
c) the reason for success of the original banner was the combination of original image and copy

In case the problem was either a) or b), we can overcome this with more testing, so introducing more new images and more new angles. But if after another batch of 10 angles and images we still don’t see at least some new successful combinations, it’s time to give up and accept the fact that we cannot make new variations by changing the image or adcopy.

When this happens, we can still replicate the banner, by changing the other 3 elements. Use very different styles of CTA buttonsΒ : round vs cornered shape, static vs blinking … Change the layout too, swap the sides, move the image to the top or bottom, try larger font for the copy …

While testing the color scheme, compare dark vs light, mild vs strong colors. Mimic the color of the targeted placements if possible.

NOTE: The less important 3 elements are rarely responsible for exceptional performance, so I don’t spend as much time on them as on the first 2. You can use less variations and don’t be afraid to mix it up a bit too.


This is pretty rare, usually I can come up with at least a different color variation… but indeed, this happens too. Sometimes you just cannot make a duplicate from a well performing banner. In this rare situations, we were lucky enough to score a special type of banner, where all the elements simply clicked and cannot be changed.

When this happens, you just have to accept it and keep on looking for the next big hit. Testing new banners is something that you should never stop doing! You never know when your old ones die and you will have to replace them. Be prepared πŸ™‚

TIP: It’s not so obvious at first sight, but in some cases, certain banners perform good only with certain LPs or offers. In case you happen to have a banner like that, you won’t be able to come up with duplicates, when using the same funnel.

This doesn’t mean that all the new variations are bad thou! You just need to test them with different landing pages and/or offers. So in case you are really trying hard and cannot come up with a good variation of a killer banner, consider testing them with a new set of LPs and offers.


You know I love real life examples, so let’s use one here. I’m gonna show you a banner that was literally killing it for me almost 3 years ago. It was also my most ripped banner of all time, I think literally everyone ripped it from me πŸ™‚ I’m pretty sure it’s still running somewhere.

It was for a BBW Adult dating offer, targeting english speaking GEOs. I came up with the angle and also made the banner. It’s a very simple one, but you know how it works, sometimes the simple things work just better.

Here is the banner :

Wanna try to guess why it worked so well?

Was it the image, the adcopy … or maybe the layout or colors? πŸ™‚


Think some more!

Do you think you know the answer? πŸ™‚

Ok, I guess you want to know the answer. It was a combination of the image and the color I used. This woman literally pulled all BBW lovers like crazy. She is chubby, but not too much, so you can still see she is actually a pretty hot young lady. This girl was actually an adult model, so I was able to get more of her pics and make plenty of great performing banners with her.

I also mentioned the use of colors – did you notice how I used the same color as her dress had? That’s not a coincidence, I did it on purpose and it really helped the CTR of that banner. This way I was getting cheap traffic that converted well … ideal combination!

I hope I inspired some of you guys! Happy testing.

BTW: In case you liked this article, share it on the social media that you are active at, I’d appreciate it.

Thanks for reading!

[grwebform url=”https://app.getresponse.com/view_webform_v2.js?u=Sym6E&webforms_id=8162402″ css=”on” center=”off” center_margin=”200″/]

Written by


  1. Matuloo my STM brother. you are dropping so much gold on your site man! I really appreciate the straight talk and cant-find-anywhere-else knowledge. They need to add you on affbuzz. hahaha Happy Holidays and Merry Christmas! Time to get the word our that you are really providing the most value in our industry. Bless up

    1. Thanks man!

      I try to write in a way that I would like to read myself πŸ™‚

      Feel free to contact affbuzz, I’m sure if more people ask for it, they will add my blog – they have a feedback plugin on the site.

  2. Hi Matej, thank you for your writing. Your topics and insights are really interesting and really something mew and fresh in affiliate space.

    As you told us you need more topics to write about, I would like to know what are the affiliate marketing fundamentals. Not fundamental skillset for affiliate marketing, but fundamental actionable steps that if one do every day, he is destined to produce winning campaigns.

    I was reading a lot of articles about Kobe and everyone says that his everyday work was always boring for an outsider because he was focusing on fundamentals to excel every day. That pattern is repeated amongst a lot of great and successful people in every field.
    That’s why I’m interested in what would it look like, if Kobe’s practicing fundamental steps and shots were translated into the affiliate marketing.

    Thank you

    1. I’m afraid I’m not familiar with the Kobe articles, can you link me to some?

      The idea for a topic is very good thou, I will try to put something together for you! πŸ™‚

      1. And I’m just curious can something like this be applied to affiliate marketing. What does the guys like you and other super affiliates do to be the consistent.
        I know that every affiliate have it’s own systems and of course you don’t want to share it in the details, but I’m just courious about the patterns in general.
        What are those fundamentals that if someone follows on a daily basis he is more likely to succeed.

  3. Hey Matuloo, arrived here from an STM post you made. I wanted to get your opinion. I just had 2 days of +ROI and all of a sudden the conversions tanked. I haven’t changed anything and my AM’s told me the offer’s still performing. I’m guessing it’s the banners but 2 days seems a little quick for burnout isn’t it?

    Thanks in advanced!

    1. Yes, 2 days sounds a bit too fast. But in some cases it can happen. This can be the case with impulse niches like Adult for example, or when working with small sources/placements.

      Another option is that your creatives attracted a very specific audience that responded very well but only represented a small % of the visitors on the source you were using.

      You can also try to pause the campaign for a while and restart it later – some days/hours work better than the rest.

  4. Merry Christmas Matuloo
    Great post, as usual.

    On the subject of images, could you give us some insight on sourcing and use of images?

    I have never paid for any images, nor have I had any problems by using images from the web.

    I use images only on banners and LP, should I be worried?

    What are the best sources and practices?


  5. Hello Matuloo,

    I’ve been trying to make adult dating work for me for a few months now, but still not much success. I have no problem making campaign work for a few days, but after that it dies. Banner burnout is my biggest problem and I am still struggling to solve it. Was hoping maybe you could help a bit?

    After discovering this article about a week ago I tried implementing strategies described here. I took 2 of my best performing banners and I tried different copies/angles on them. Some new angles performed significantly better than others. Overall the campaign was in small profit (+15% or so). After few days it looked like couple angles performed better then others. So I took those banners and duplicated them with new images, while trying to keep similar style of models, like in those original images. I ended up with about 25 banners.

    I ran them for 3 days, killed the worst ones. Ran 2 more days, killed even more, ended up with 8 banners who were outperforming the others. And that’s it, the next day conversions stopped. It just seems so weird that all the banners would burn out so fast.

    Could it be so, that the reason those banners were working well was because they were part of the group of 25? And once I killed all the worst banners, the good ones quickly burned out? If that’s the case, maybe I just need to run many banners (like 20-30) all the time and just replace the worst ones all the time? Do you usually run lots of banners in your adult campaigns?

    Also, I somehow thought that speed of banner burnout also related to the number of banners you have in the rotation. So for example, if one good banner can last about 4 days, so if you have 8 banners, you could run the campaigns for about 32 days. But it seems like in this case this wasn’t true… It kinda seems bit weird to me, if that’s true. Maybe then I should do so, that once I find a good banner (or good image) I should only run that one for few days? And once it dies out, just replace it with new one?

    I know this is an old article, but hopefully you will still reply. Really appreciate it.

    1. Hello Justas πŸ™‚

      Usually when this happens, the banners are the whole funnel is not strong enough. It can also be a bidding problem.

      There is also the chance that was some problem with the offer…

      Fundamentally, you did it all right, testing variations, cutting the bad performers … that’s the way to go about it.

      You are also right about the burnout being related to the amount of banners in rotation, the more banners the longer it can last … but I wouldnt use simple math like you suggested. It depends on the rotation, audience size etc … but yes, using more banners should help pretty much always.

      Hard to give you specific advice, but I think the funnel wasn’t strong enough, that’s why it died quickly. You need to find out where it needs to be improved.


      1. Thank you a lot for your answer! Really appreciate it.

        I kinda think there were multiple problems with that campaign. First of all, the offer was paying low, even if it had very good conversion rate. But I had to keep it at at least 10% conversion rates (counting on hits to landing page, not the offer page) to keep it profitable. And not even good profitable, like 20% or so. I know offers convert at very different rates, but since 10-11% is so much more than industry average, I thought that was one of the problem. I mean it’s really hard to keep such rate. I might be wrong here though. I later changed to better DOI offer, which pays normal amount and didn’t have to keep such huge conversion rate to be profitable.

        Another issue was landing page, I think. I was using some downloaded lander, which wasn’t changed at all. As long as I made some changes to lander (added new models, country flags) it worked better.

        As for banner burnout, I still have couple questions. I guess I am a bit overthinking the whole process. If you can, tell me if I am thinking correct here. Basically once I find something that works, let’s say 2 banners. Then I will run campaign with those banners with, let’s say $100 per day. Then I should dedicate some budget, let’s say in this case $20 per day, to testing new banners. And of course, highest chance of finding new good banners is buy using logic described in this article. And once I find good banner, just add it to rotation with the other ones, and cut out the ones that do not perform anymore. Is this how pros do it? Or there is something else fundamental to it?

        Thank you again for you help!

Leave a Reply

Your email address will not be published. Required fields are marked *