Header Image
  • The Beginner’s Journey: How a PR Manager Started Affiliate Marketing. Part 6. Tests, optimization and summing up
    Reading Time: 12 minutes

    Hi there!

    After a short break PR-fire Vi once again is rushing into a digital space of the AC Universe!

    No, my absence is not connected with the full disappointment with the results of my first campaign (yes, it is a spoiler, but the details will be revealed soon). 

    After looking through the final campaign statistics, I realized that I urgently need to improve my skills, which means turning to the illumined sages of affiliate marketing. I will not go deep into the details, because the most important lesson I learned there was to ignore failures, because they are a part of the learning process that all beginners go through.

    What is much more important is to follow my example: gain experience, take note of it, and start studying the topic in detail – both theoretically and practically.  After all, without studying (self-education counts too) and learning, it is impossible to succeed in this sphere. It’s important to be motivated, but it is not enough to master this niche.

    And now let’s move on to the results of my campaign, which has splitted into several similar ones during the optimization process. As promised, I will tell you how the optimization process went and where it eventually led. 

    Well, now we can sit comfortably in the armchair and dive into a story.
    In the last part  I made creatives, created a campaign in Keitaro tracking software, set up the postback URL, and launched the campaign.

    A/B testing: landing page / pre-landing page + landing page

    Once the campaign was launched, my manager Valeria from RichAds and I decided to check if it is possible to increase conversions if before switching to the landing page users will go to the pre-landing page. 

    In order to do this, I needed to duplicate the first campaign and add a link to it – landing page + pre-landing page [89257] Top Blog – Sex Up – AR. To choose a pre-landing page, I looked at the converting creative pairs on the offer page (for my landing page):

    Moreover, it’s stylized as a blog which is run by a young man, and where the secret to enhancement from a porn star is supposedly revealed. The pre-landing page contains a lot of explicit footage from adult videos. Usually blogs, especially with stories, show good results. Oh, and the text for my creatives was easy to adapt to it.

    First of all, it is necessary to copy a link from my AdCombo account. I go to My Campaigns and create a new campaign, which includes two links: one to the pre-landing page and another – to the landing page. If I change the link on the page of the offer, the first link to the landing page will be deleted, so I added it to the first campaign. That is why I need to create a separate campaign in AdCombo for testing.

    I choose the required landing and pre-landing pages and copy the formed link.

    Now it should be added to the tracker, to the new campaign, i.e it is necessary to repeat the process of adding the offer to the tracker.

    Next, I copy the link from the tracker to add it to the duplicated campaign.

    I copy the campaign and change the text of some creatives to make them more in line with the pre-landing page (I specified the same numbers that are on the pre-landing page):

    Remedio de estrellas de porno. +5cm en un mes. ¿7 veces por noche? ¡Fácil!

    Porn star remedy. +5cm in 1 month. 7 times a night? Easy!

    Chico encontró cómo tener + 5cm. ¡Ahora mujeres no le dejan descansar! 

    Guy discovered how to get +5cm Now women don’t give him a break!

    You can change settings in a new campaign. Changes in the second campaign, or vice versa in the first, will not affect the other. 

    Valeria advised me to raise the bid to get more traffic. 

    The bid was raised: $0.009 -> $0.011.

    I did not double-check the settings with the manager, although if any doubts exist, it is always possible to ask the manager to double-check the settings you have doubts about before launching. This is especially relevant for beginners who are just trying to launch their first campaigns and do not have much experience. Thus, getting the desired result from the campaign is more probable.
    To be sure that everything works fine, I clicked on the new link. The click appeared in the tracker. Great, everything is ok.

    A/B testing results and first campaign improvement

    The next day I noticed that one of the sublists was highlighted in red. In two campaigns, the second was created 1.5 days earlier than the first.

    Valeria suggested that if I have a postback set up and a targeted CPA campaign indicated, then expensive and ineffective sources will be highlighted in red. Since little time has passed from the date of campaign creation, I don’t need to do anything about it yet. Valeria advised to collect statistics for at least 7 days, sometimes it is better to collect statistics up to one month. This is the only way to analyze the results and make the right decision. 

    Soon my manager from RichAds detected a couple of problems in my statistics.

     № 1. The chosen pre-landing page didn’t convert at all:

    The 8084 sublist that gave a lot of conversions for the first campaign, didn’t work at all for the second campaign with a pre-landing page. In my case specifically, the creative pair didn’t work, because everything depends on the components of the run. In general, you should test absolutely everything for yourself. Personally.

    Valeria suggested stopping the second campaign and concentrating more on the first one, which had better results. Despite the fact that the campaign with the pre-landing page had been running for a day or so, Valeria insisted that it should be turned off, because the first campaign with the same spend had brought 6 conversions, meanwhile the second had brought nothing. I deactivated the campaign in RichAds and the tracker.

    № 2. One of the creatives in the first campaign converted at the more expensive rate than the others: 

    $1.66 CPA, while the others had below $0.5 CPA. We noticed this a day before, but we agreed that I would keep a closer eye on it and report any changes in the stats. Within 24 hours there was 1 conversion, CPA decreased a bit, but not as much as we would have desired. Alas, it had to be stopped after all.

    It was not necessary to add a new creative, but in order to be sure that one landing page was converting better or worse, I suggested testing the second pre-landing page from the offer page. It already looked more like an article – [89256] Men’s Health Life – Sex Up – AR. There are less explicit photos in it, but a few can be spotted just in comments, “from users” section.

    Secret discovery will increase the size of your p*nis by 4cm a week!

    Valeria approved the idea. Now it is necessary to create a link again with the new pre-landing page and the previous landing page in AdCombo:

    Then you need to copy the link into a new campaign in tracker:

    I click on the formed link and see 1 click in the tracker:

    I duplicate the campaign with pre-landing page and change a bit the text of the creatives (once again I take the numbers from the pre-landing page):



    However, my manager Valeria from RichAds said that for now it is better to concentrate on the first campaign, because it has more conversions.

    So we raised the bid for it, but it did not really help to increase the volume, which is affected not only by bid, but also by CTR. The level of conversions of the campaign was good enough, but the highest CTR could be noticed in the first day of launch, then it started to decrease, that is even the increase of the bid have not influenced the volumes. 

    Although we decided to try another tactic: to duplicate the first campaign and add only new subscribers there. 

    New subscribers are the audience with the highest conversion rate, but you need a higher bid there, +20% to the current bid. There’s not much traffic there, so it’s better to run 2 campaigns at the same time: new subscribers + all sublists. You need to set different bids to maximize audience coverage in the campaigns.

    We decided to set this bid – 0.0132.

    I duplicate the first campaign and make changes to it:


    After 4 days I noticed that the balance is declining, I have clicks and leads, but only 2 sales. 

    Orders are made, but they quickly turn into holds or trash. 40 conversions brought only 2 sales. 

    Valeria said that we need to find out if there’s a chance that it is the norm for an offer – when only some of the leads convert, or it is not.

    If only a part of leads converts for an offer, and this is its own norm, then you can start cleaning sources with leads without actual purchases, so that you don’t keep spending your budget on useless, “empty” leads. If the offer doesn’t show this peculiarity in other sources, then you probably need to change the creative pair.

    Meanwhile, we found out that the first campaign with only the landing page had sales. So in my case, the best performance was with only the landing page. 

    The campaign with new pre-landing page had only 4 conversions during this time, and the campaign with new subscribers only had 4 conversions too. There were no sales.

    Finally, we decided to turn them off and leave only the 1st campaign with the landing page and all the sublists active. It was the one that made sales and may still do so. 

    I was wondering what was going on. This time I asked my manager at AdCombo – Artyom. 

    First of all, he asked me how long ago I chose this offer. I remembered that I didn’t choose the offer right before the launch, because after I had made it, I prepared for the launch by making notes and designing the post. So it took some time. Artem said that the offer acts better with a different type of traffic, so it’s better to try to change the source if I decide to continue working with it.

    The interaction and effectiveness of offers and sources are very dynamic. In addition, the relevance of an offer, approval rate, quantity of goods, – everything changes very quickly. In order to maximize results, always check the relevance of the selected offer for a particular source and other details. To note the offer in advance, or to hear about it once and run it immediately or remember about it after some time is a definite risk. It is always better to ask the manager, which will definitely help you to make the right choice.

    Artyom looked at my stats and told me that I had a 58% of thrash. Hmm, and also holds…Why so many trash and holds? It was mostly wrong numbers. Only 1 customer, whose order later went to trash, refused to buy over the phone because he wanted to know more about the ingredients of the product.

    Requests were on hold because customers had given possibly correct numbers, but eventually did not pick up the phone to confirm the order. It usually takes the Control Center 3-4 days to get through to those who didn’t pick up the phone right away. But as each day went by, I saw holds turn into trash. Buyers never picked up the phone. 

    There is another assumption why the audience was not interested: some of the creatives needed to be finished or replaced. It is not confirmed but there is a chance, even taking into account all the conditions mentioned above.

    But I found out that this particular type of traffic does not fit the offer, even if there were successful cases before. It works differently in each particular case. With my campaign components, it was not suitable.

    In general, it was possible to test the other creatives, but I decided not to waste my budget any longer, because the best solution was to change either the offer for the current traffic source or the current traffic source for this offer.

    I wrote Valeria the next day, because the number of sales still hadn’t increased. We decided to stop the first campaign too.

    Statistics and outcomes

    In total, my campaign lasted a week before it became clear that it was better to stop it, because it was unlikely that anything would change for the better, I would only spend the remaining budget. 

    Just in case, I asked Valeria if it’s too early to stop the campaign, because it exists only 1 week? 

    She replied that it’s not a matter of days, but of a formula that easily determines whether a campaign has a chance to become profitable or not.

    If spending on the campaign amounted to ~10 values of conversions, and it did not yield results, it is worth changing the offer. The formula is: Target CPA *10.

    This will help to give a general idea of what to do next with the campaign. 

    If my first campaign had not converted by $8, it would have been stopped.

    Now let’s look at the final statistics.

    Statistics from RichAds 

    Period: 23.01 – 31.01

    Spending: $534.34 – $83.08 = $451.26 (balance)

    Income: $11 x 2 sales = $22 

    (Anti) Profit: -$61

    ROI: -73.4%

    Creatives that converted in the first campaign. It was the one that brought two sales:

    The idea of being an alpha male motivated the male audience the most. 

    Traffic from mobile devices was good:

    Android 11,12 and 8 had the most conversions.

    Statistics from AdCombo 

    Total spending on the campaign (including spending from the past posts) was $624.16.

    Making mistakes is a normal process that all beginners go through. You should not be afraid to make a mistake, but on the contrary – mistakes should be appreciated, as well as analysis of the experience gained and could be applied in the future. One should not forget that a professional is the one who has made the most mistakes (c). 

    After the launch of the campaign you need to keep the pulse on and be receptive to any changes in order to take action in time and not to waste the budget. Be sure to keep in contact with your managers, both the affiliate manager and the ad network manager. Remember – they are your reliable partners, because, roughly speaking, your profit is their profit too. 

    Always test everything and anything. There are no universal tips. There are recommendations, but to understand exactly whether they work for you – you need to test them. Imagine that your entire campaign is one big puzzle and you need to collect many pieces to get a beautiful picture.

    I hope my article will help you to master the niche, inspire with new ideas or just turn out to be a fun and useful way of spending your time. 

    The whole series can be found by tagging #affiliatemarketingforbeginners on the AdCombo channel.

    I wish you to make useful discoveries and make as many mistakes as possible, which will give you invaluable experience and giant ROI in the future!

0 Comment

Leave your comment