Home > POP / PPV / Redirect > Follow-along Campaigns

---

Is dating still alive? (15)

---

Hey guys!
I'm not total noob in affiliate, but...
In 2016 I left affiliate marketing and took on other projects that were eating up 100% of my time. At that beautiful time I was working with dating and my main sources of traffic were POP networks.
Recently, I decided to get back into AM and see if the dating is alive now or not anymore .


My old experience tells me that in such verticals it's very important to test a large number of variant landing pages.
Luckily I have good technical skills, in particular I'm proficient in PHP. Thanks to that I made a MVT system, which helps me to replace images and texts on the landing pages with a couple of clicks and test different variants.
Now it is still raw, but can already help with the main task.
I also set up and prepared Binom tracker (which I have worked with in the past).


Now I need to find some good offers for testing and choose which traffic sources I will work with.
So far I am thinking about such sources as Exoclick or Plugrush.


To start with, I decided to take some offers from ClickDealer (I've worked with them a lot in the past too).


I've decided to keep this thread to share my results, and also to listen to your helpful advices.


Best regards,
Alex.

---

It does works, it will always work....because dating is a evergreen niche.

But remember one thing that, before you atleast 100 people have promoted the same dating offer in that pop traffic source, in same country, in front of same audience. So, it's already saturated.

That's why, the old technique that used to work 5 years ago, will not work now. You'll have to find a new angle. Without a unique angle, it's difficult to get a positive ROI.

Also, i prefer to stay away from SOI/DOI offers. Because their payout is mere $3-$4. I rather prefer to promote cc submit or CPS offers, as in that case payout is usually above $30 and ROI doesn't fluctuates so much as in case of dating offers. Also, the effort it takes to promote high payout offer is same as promoting dating offer.

---

I see it a bit different.

Let´s be honest, in 99% of all CPA dating campaigns the good ol´ rules lander/angle is used because it just works.

It´s more about testing images than testing angles.

There are also very stable converting dating offers that are promoted for many years already and that are still doing good.

For dating I would definitely stick with SOI/DOI first, payouts there can also be up to $20 or even more depending on geo and device.

Later it can be good to run rev share but I don´t know anyone who had success with CC submit dating offers.

---

Of course, my experience is already outdated as you said, but in general, I saw just about the same landing pages, which were before the world went crazy. In any case, I think that my experiment can be quite interesting.
At least have a laugh


The MVT idea is not new at all, but I decided to be an artist and make my own.
Without getting into the technicalities, I'll just tell you the basic idea:

1. Having a few basic templates, generate various landing pages with changes in text, pictures or even css-styles.
   
2. Set up rotation of variants in the tracker using a simple web-panel.
   
3. Send enough traffic to make decisions.
   
4. Disable inefficient variants and add new ones.
   
5. Repeat N times

So, here we go.

What technical arsenal is available for now:

1. A MVT-generator script with a simple control panel
   
2. Binom tracker
   
3. A script for a landing page, connected with the tracker and MVT
   
4. Several templates of landing pages
   
5. Several scenarios and several images

Current Reflections:
I’ve used conversion boosting scripts (alerts and backfixes) in the past, but today I've noticed that alerts don't behave like they used to. Backfix seems to work as before (at least on desktop, I haven't checked the mobile version yet).
With a bunch of variables at the start, I realize that my experiment could cost a decent amount of money, but I'm willing to spend a few thousand dollars.
I haven't tested the script and I'm not sure if it will work at all. But I consoled myself with the thought that there is an old Plugrush account that has a balance of ~$300, which is enough to test the performance of the system at all.

Current todo list:
1. Find offers, which in theory I can get some conversions (geo - Europe or Latam.)
2. Set up a traffic source and create several ad campaigns in binom
3. Generate and configure a dozen different variants of landing pages for this campaigns
4. Run traffic from Plugrush
5. Check out is it work or not

---

Guess who´s idea it was to implement it in Binom

What changed for you?
I didn´t see a difference in JS alerts.

---

How did I not know this was in Binom already?

Combine this with their upcoming auto-optimization feature with landers and offers and… I can go on vacation

---

I used LPG (Landing Page Genius) in 2011 or so.

This is also a MVT script that worked good with CPV Lab.

It can be very helpful so I thought it would be a good match for Binom.

---

Alert used to be able to pull the tab to the top, but now I haven't noticed that.

Yes, that's about my goal.

At the initial stage, I have two templates and two CSS schemes for each of them. There are several prepared text scenarios, and a set of pictures with girls. Next, using the script I generate a set of landing pages that looks like this:


So, the landing pages are ready, and it's time to start the tests.
When I was working with dating in 2016, it converted better from phones than from the desktop, and the landing pages themselves are adapted specifically for mobile devices.
For offers I went to two affiliate networks I had worked with before - Clickdealer and Adtrafico. I decided not to spread myself too thin by geography and chose Germany and France to begin with. I knew these geos and had a good idea of what to expect.
Next I set up the first campaign.



I decided that it’s enough to add these (randomly selected) offers for the test.
I set up a campaign in the tracker and clicked through the link from my laptop and cell phone. Everything works as it should and you can run traffic.
Went to Plugrush and ran my first campaign.
Geo - Germany
Platform - cell phones
Browser language German
Orientation - Straight
Webview - Disallow

Estimator offered to set cpc $0.00152, I set it little higher - $0.002.

Then I selected the versions of landing pages that I want to see in the campaign (a total of 12 versions).
Yes, it turns out that I have a lot of variables at the start:
- no black and white lists for traffic.
- no tested offer that would burn
- no tested landing page that's sure to work
But the main expectation that I wanted to justify - is the average CTR of about 1-1.5% (I remember, before it was something like this) and ROI -70%

Then I made the same campaign in Exoclick, with about the same targeting settings, but with a smart Cpm strategy and maximum bid of $3.
The Plugrush campaign went through moderation almost instantly, and while I was setting up the campaign in Exoclick, I had already managed to get ~1000 visits from Plugrush. But what surprised me was that I didn't get a single LP click (only two my clicks). Then I thought about stopping the campaign and seeing what was wrong with my feed the next day.
Overnight the Exoclick campaign went through moderation and started traffic too. I got ~15000 hits with a CTR of only 0.19%. I have no idea what is the reason for that at the moment. The variant statistics came together without nicks and you can see that the traffic distribution is working. Of Course I have no conversions at all.



Graph (Unknown) - these are devices on which JS-scripts do not work and, accordingly, the tracker does not receive information about which version was selected (I will consider them bots).
The first two letters in the identifier of the version indicate the language, respectively, there are some clicks from the traffic source moderators, and the rest - in German, as it should be.
Need to figure out what's wrong with CTR. By the way, here is the link to the campaign I made especially for you to see how it works, maybe you can tell what's wrong with my webpages.
Here I made clone of one of my campaign for you, reader


Now the plan is also to make new templates from those interesting landings, which I found here on the forum.
By the way, when you download from somewhere landings, look them carefully and do not glimpse similar shoals - the title from the vertical sweeps, in the dating landing.


In addition, with this template encountered with the fact that the pictures are prescribed in css, and there are several at once. Yes, i can cut style snippets from CSS and paste them into HTML code, or add CSS file scanner to generator, but the fact that there are a lot of pictures - is more interesting.
Scenarios, which I have decided to use as a text component of my landings can be quite different lengths and senses, for example, the scenario can be short and simple:

• an "intro" block (mandatory for use in some templates)
• A couple of yes/no questions
• The final checkout block

(e.g. 4 scenes in total)

Another scenario may be several times longer

• an "intro" block (mandatory for use in some templates)
• four yes/no questions
• Three multiple choice questions
• Final checkout block

(e.g. total 9 scenes)

Accordingly, the pictures will need to either select the appropriate number, or make a static set of images (e.g. 5) and if the scene suddenly turns out to be more, rotate them in a circle.
Now it will be necessary to add such features to the generator, I think it will take some time.

---

Let me chime in too, since I'm promoting dating.

In my experience, the MOST important element is a good offer... LPs do play a role too, but the offer is what makes or breaks the campaign. As @twinaxe mentioned, the same LP styles keep on working year after year, so you need to put more focus on finding the right offers.

High payout is a must too and in order to get there, you need to send quality leads. In my experience, POP traffic can often fail when it comes to this and that's why I focus on banner traffic.

When it comes to mobile VS desktop, somehow I'm always doing better with desktop, but the truth is, bulk of the volume is in mobile now.

I'm really looking forward to see the results of your experiments, keep us posted!

---

Hey @matuloo,

Perhaps this question has already been covered more than once, but still... How to get the right offer? Thank you.

---

After getting such sad results with CTR, I thought that I have some problems with the landing page, which I won’t notice. In order to check this out, I also decided to use a static landing page downloaded here on the forum. After adding the landing page to the split and sending some more traffic (about 1000 visits) - I’ve got only one LP click, by the way exactly on the static landing page.



At this point, I reconfigured the dynamic landing page by turning on JS-alert in hopes of getting more user attention.
With a little more traffic I’ve got the following results:



And then I had a thought about caching and Cloudflare's SSL certificate, which could interfere with the tracker's performance.
I went to Cloudflare and switched Proxy Status to DNS only mode and re-issued the SSL certificate through the tracker console.
After disabling caching on domain (I use the same domain for tracker and landing page) I`ve got performance degradation, and quite tangible.

Without caching, on the same domain:


Then I moved the landing to another domain, and enabled caching on it, and left the tracker domain in DNS Only status.

With caching, on two different domains:



Then, I tried to check the speed of the landing page itself, if I go to it directly, without the tracker`s redirect. In this case I`ve got the following speed results:

Such a big difference in performance, of course, prompted me to use ClickApi after all.
When trying to connect ClickApi, I encountered the fact that it only works when using the primary tracker domain. If multiple domains are linked and not the primary domain is used for clickapi, then LP clicks are not fixed. Weird of course.

So, performance test using ClickApi:


Yes, it works faster than a tracking link (which is understandable, there is no redirect). But still, after passing the questionnaire's landing is a tracking link, which tracks LP-click and sends traffic to the offer. I’ve noticed that it does not work very fast and takes about one second.
Of course it's possible to get the link from the landing page to the offer immediately (it's possible to get it through ClickApi library, but in that case the click won't be registered in the tracker, which is not good). I didn't want to make an additional page that will redirect to the offer itself - it may result in double clicks, which I don't need. So far I will not bother with this.
Today I`m launching new campaigns, but I`ll change geos too. It is possible that Eastern Europe is working for me in a different way. My current goal is to get a CTR above 1.3%.

Yes, as legendary matuloo said, the offer is probably the most important aspect of a profitable campaign, but without clicks to the offer, even with the most stinging offer you won't succeed in getting conversions.

---

You're working with POP traffic, right? In this case, you need to get the overall loading speed below 1 second, including the tracking redirect. Otherwise, the traffic loss will be significant and affect the overall performance of your campaigns.

Different GEO will definitely perform in a different way and it's ok to switch GEOs when one doesn't work... just don't jump from one GEO to the next without giving it a proper try first.

One word of warning regarding LP CTR... this parameter is important as it tells you a lot about the LPs, but it can be very tricky. Quite often, the ones with low CTR convert better than the ones with a high CTR. The only thing that really matters is the final CVR and ROI.

---

There is just one way van graaf... by testing. You need to enroll with good networks, ask for their best offers and then test them. The truth is, every network will say they have the best offers or at least "close to the best ones" but obviously, this isn't always the case. So you need to find out the hard way

---

How much wud you guys spend on testing 1 dating offer? 3-4 times the CPA amount or depends on your gut feeling?

---

Depends on the initial results.

Since I have plenty of proven LPs and banner at my hands, I know that a solid offer cannot fail completely. So for an initial test, I would spend roughly 5x the offer payout and if there are no leads at all I know it's a bad offer and stop. If there are some leads, I keep on testing until maybe 15x spend, max 20x and then look at the results. The next step depends on how many leads I got and how deep I'm in the negative. If I'm close to profit or at least not too negative, I keep on testing for several days... this is the phase where the gut feeling comes into play. I'm not going to beat a dead horse, so if after several rounds of tweaks and optimizations I'm still not in profit, I drop the offer.

There are other factors that make me stop sooner... when the CVR is erratic for example or when changing ads or LPs don't seem to make any difference.

---

Home > POP / PPV / Redirect > Follow-along Campaigns