Home > >

---

Getting Traffic But NO Conversions AT ALL? Read This Now! (27)

---

While thinking about topics to write on, I realized many people face one problem when starting out : traffic is flowing, but the conversions are not. This made many people stop with AM sooner than they actually gave it a proper try.

The truth is, we have all the problems covered in various threads here on STM, but it's cluttered at times and harder to find … So let me address this issue in a separate thread, help you understand it better and give you a few tips to overcome this rather large obstacle.

WHAT IS NEEDED TO MAKE A CONVERSION

First of all, we need to understand there are several “factors” that need to walk hand in hand in order to get a conversion.

1. We need traffic and the traffic MUST be real! If we buy BOTs, we simply cannot convert. Simple as that.

The most simple, but still pretty effective BOT detection method is described here : http://stmforum.com/forum/showthread...n-Any-Campaign

2. We need a functional offer with a solid flow. If we pick a paused/capped offer or something that is poorly coded … we won't get any conversion.

Caurmen wrote a guide on how to spot awful offers : http://stmforum.com/forum/showthread...low-Testing-It

3. We also need solid creatives, be it banners or LPs, in case they suck … well, the conversion rates will suck too.

So basically, we need solid creatives, solid offer and we need to send real traffic to them … otherwise we cannot expect to see a conversion. This is very basic knowledge, but still, many newbies fail to understand that any of these 3 parts can take the whole campaign down … no matter how good the other ones are.

Now let's take a closer look at the possible problems and how to deal with them.

I'M GETTING TRAFFIC BUT DIDN'T GET A SINGLE CONVERSION YET! HELP!

We see this all the time here on STM, someone buys a couple 100s clicks but doesn't get a single lead or conversion. The world is doomed, all looks black, there is no light at the end of the tunnel … but hey, there is always hope!

This problem is usually tied directly to the traffic itself and that is also the first thing we need to take a closer look at. Remember how I mentioned that you need REAL traffic? That's what we are going to sort out first.

NOTE: All the traffic sources are competitive places, dozens of affiliates are competing against each other for the SAME traffic. Experienced people are bidding high for the good placements, pretty much all of them have black lists in place …

Guess what happens when you come to a traffic source as a fresh newbie and start a new campaign? You get the traffic that nobody else wants. Plus leftovers that the big guys left on the table. What you need to do now as a newbie, is avoid the traffic that nobody wants and target the leftovers. Let me explain more.

TRAFFIC THAT NOBODY WANTS? WHAT IS THAT?

There can be several reasons for a placement to become unwanted :

1. Placements with too high BOT %, you need to identify them and block first. Use a bot detection script and cut anything that has more than 80% bots for example. Trying to optimize BOTs is like inviting your vegan friend to an expensive steak restaurant … you're gonna spend money but it's not gonna end well

2. Placements with poorly placed banners, errors in code, incorrectly loading POP windows etc … translated to normal language, you will be billed for impressions, but the clicks will be low because the delivery process is flawed somewhere.

The users either don't see the ads at all, or see part of them … maybe they load in a broken format etc … one way or another, you don't want this traffic. The best way of detecting these placements is looking at engagement factors like banner CTR or LP CTR. If it's too low and thus giving you too high prices per click … cut it. It also works the other way around, if your LP CTR from a placement is 90%, while the others are at 30% … well, something smells here.

3. Some placements just send poor traffic and it won't convert for shit. This could be feeder traffic that they are purchasing to boost their numbers, it can be POP traffic mixed with banner clicks, it can be incentive traffic when it should be natural … there are many options here again.

Every network has a solid volume of unwanted traffic and they will happily sell it to you, even for the minimum bid Make sure you get rid of it as soon as possible, otherwise you will be left in the dark, wondering why you don't convert even when traffic seems to be flowing nicely.

Once you blocked the worst crap, you can focus on optimizing the real traffic. But don't aim high just yet, focus on the left overs from the big guys!

LEFTOVERS? FUCK NO, I WANT TO MAKE $X.XXX PER DAY!

Ok, but are you also willing to invest $x.xxx per day straight away to get there? Probably not, so sit tight and let me explain what I mean by the leftovers

Even when you detect a good placement, you don't want 100% of it's traffic. In order to fully utilize 100% of some spots traffic, you would have to rotate multiple LPs, offers and a ton of creatives. Otherwise, you'd serve the same stuff to the same surfers over and over, which would only take the effectiveness down. In many cases, you only need certain part of the traffic – some OS, carrier, device...

Experienced affiliates know this, that's why they use impression caps, daily budgets and all kinds of targeting … So even thou a placement is solid and targeted by multiple affiliates, there will always be some of it's traffic left for the taking for a rather low bid. This is what I meant by the “leftovers” and this is the kind of placements you need to target first. Solid traffic from real placements, that can serve ads in visible spots or deliver properly loading pops … for a decent price.

Once you have this sorted out, we can move on.

NOTE: Identifying these leftover placements is the beginning of your journey, not the final destination. The main objective is to make sure we are not attempting to optimize something that it is not possible to optimize.

The good thing is, you can identify majority of the poor placements while running at small bid, because people usually don't bid high on those … and at the same time, low bids will give you access to the leftovers. Once you have the whole funnel sorted out, you will uncover the better placements by raising the bid.

OK, GOT THE TRAFFIC PROBLEM SORTED OUT, BUT STILL DIDN'T GET A CONVERSION.

TIP: Make sure the traffic you are buying is also matching the targeting requirements of the offers you are promoting. This is very important in case of carrier billing offers for example, if they want Orange traffic and you're sending O2 … well, it's not gonna convert obviously.

Now when we have the traffic problem sorted out and we are absolutely positive, that we're not buying BOT's or some utter crap, we can move on. In case you still didn't get a conversion, there are 2 more possible conflict zones.

You're either using the wrong offer or the wrong creatives.

WRONG OFFER? MY AM TOLD ME THIS ONE HAS KILLER EPC, SO IT MUST BE GOOD!

Well, what's killing it for one person, doesn't have to work for you. They might have different banners or LPs, or a special traffic that you don't have access to. When asking for offer recommendations, don't settle just with the EPC! Ask for volume instead, higher volume is a better indicator than high EPC.

It only takes one super affiliate to reach good looking EPC, but to do some serious volume, the offer has to work for many affiliates.

Do some market research too, with the spytools that exist these days, this is a very easy task actually. Spy around, see what others are promoting and ask for the same or similar offers at your affiliate network.

TIP: Gone are the days of scoring with the first offer you choose to promote! Prepare to test a dozen of them, or several dozens … that's how AM works now, sometimes you need to test a large amount of offers to find one that works well with your creatives.

We have one more factor to take a look at, the creatives!

MY CREATIVES ROCK, I RIPPED THEM FROM A SPYTOOL!

Well, right, and so did 99 other affiliates. Are you sure you copied an affiliate who is turning profit, or was it another newbie who is loosing money by using poor creatives?

Spytools can help you a lot, just make sure learn how to use them properly. Don't think that you can just rip a bunch of banners or LPs and turn profit. It's the same as with offers, you need to test a lot of them to find something that works with your funnel.

TIP: Some LPs work well with a certain type of offers, but will totally blow with a different offer, even thou they might both be in the same vertical. There needs to be some kind of connection between them. Usually it's the general feel and it's hard to describe why it is so. Just keep this in mind and don't be surprised when a proven lander bombs with a new offer.

OK, I UNDERSTAND I NEED SOLID TRAFFIC, OFFER AND CREATIVES … BUT HOW TO BLEND IT TOGETHER?

As I mentioned at the beginning, any of the three core parts of a campaign can kill it. And identifying which one it is in your case, is the hard part.

From my experience, you can divide these into 2 sub-parts : 1. traffic, 2. offer+creatives.

I always start by filtering the traffic first, then I rotate creatives and offers until I find something that works.

TIP: Once you manage to get a solid performing campaign, take the funnel and make it your benchmark funnel for the future. When you are launching a new campaign, you can use this benchmark to test things faster, because it has already proven itself once. It won't work 100%, but can definitely speed up your testing process and save you money. When working with the same traffic source, you can also utilize placements blacklist this way.

LET'S SUM IT UP A BIT, SHALL WE?

Starting a campaign and not getting single conversion sucks, but it has become quite the norm for the newbies these days.

Before you drop the gun and run away from AM from good, stop for a while and take a look at what you have done and whether you actually took the necessary steps.

1. Are you sure you bought real traffic? Bots won't convert! If your ads don't load, they won't convert. If there problems with the delivery … you can't convert. Just because you spent $50, it doesn't mean the traffic was worth it. Take your time and filter out the CRAP!

2. Did you test a dozen offers, or just one? What if I tell you, many people test several dozens before they find one that works? Doesn't matter that your AM tells you, this one offer is the hot shit … you need to test many offers.

3. How many banners or LPs have you tried? Or did you send 500 clicks through a directlink and wondering why there are no conversions? No no no, doesn't work like this anymore. Get a spytool, tweak the most popular landers and test test test. Spytools help, but they are not Almighty, learn how to use them and count with the possibility that you could have copied another newbie!

Keep in mind that all the traffic network use the auction model, in case they are able to sell the traffic, it means someone is bidding for it! These people are not doing this to throw money out of the window. So if it's not working for you, there is some problem with one of the 3 core factors I mentioned above. FIND IT!

Thanks for reading

---

Great sum up Matej!

---

So to find good placements would you recommend look at longest running pages on Adplexity, find which publishers they are using and target those to take their leftovers?

---

Yup, this is correct thinking - if someone keeps on buying from certain placements again and again, chances are the publisher will be solid.

---

How many placements would you target with this method? And how thoroughly would you research a placement? For example would you go on Adplexity to view the LP's running on each placement and make sure they aren't saturated with your vertical or even exact same LP? Or is that not worth the time?

---

As many as you can, whatever is solid, you want to run there I personally don't research placements on adplexity much, I prefer testing all on my own directly at the sources - that's faster, thou it costs some. For me, adplexity is more of a source of inspiration when it comes to banners/lps/offers.

---

Thanks for this.. Is rotating the same offer between different affiliate networks recommended?

Also, what if there are no similar offers in the geo for you to split test?

---

Yes, split testing the same offer from various networks is one of the recommended steps to take.

Well, that's a problem then. You can try what is available, but if nothing works and the problem seems to be on the offer side, it's time to move on.

---

Really nice guide and very helpful indeed.

One to print out for my "essential info" folder I think.

---

I'm glad you find it useful simant!

---

Thanks Matuloo, really helpful post!
One question I had was, even if you're able to find solid publishers on Adplexity, how would you then go and bid on them on your traffic source? Aren't the domains usually anonymised with some kind of number code instead of the actual domain name? (I'm mostly familiar with MGID so maybe it works differently on other platforms?)

---

Indeed, many sources hide the domains ... but in some cases it's not that hot. For example adult sources are usually pretty open when it comes to this. Another trick that you can use - the networks have to identify the placements somehow, so the IDs are often to be found in the source code of the sites

---

Interesting - sounds like I need to do some more digging. Thanks for the response.

---

Matuloo thanks, but i do not undstand, after download LP, how install?

---

Wow I've got so much to learn, great post! as a newbie I can spot the mistakes i've made!

---

LP is just a simple "html" (or php or whatever) page, so you just upload it to a server, like you would with any other website that you want to put online. In case you don't understand the basics of publishing sites on the net, you might want to read some tutorials on it.

---

Hello Matuloo,

Thanks for the insightful tutorial. I am a little but confused here.

For example, if there was an Affiliate A targeting Orange traffic and that there is a total of 50 people on a particular source that matches the requirement, why would Affiliate A not want to get all 50 of them? Surely the affiliate still has to put impression caps, etc to maintain banner's effectiveness.

Hence, the ultimate question is why would the affiliate want to leave traffic (a.k.a profit) on the table?

Thank you

Cheers,

Cynthia

---

You answered it yourself partially : there is traffic and there are unique users, so while you might want to show your ads to every unique user, you don't want all the traffic, so to show it to them on every impression they generate. So you limit it with CAP.

There are more reasons to this : if you want all 50 users, you need to bid higher than everyone else, in order to get the first impression ... this is something you might not want too.

Part of the 50 users will return the next day, or the one after ... if you show ads to all 50 users every day, you will end up with showing the same ads to the same person again ... this overlap will be lower when you don't take 100% rotation.

The ads will also burn out faster when targeting all 50 users, so you will pause the campaign sooner, which will limit the exposure to new users, who'd come in the following days.

There are more reasons for this, but I'm sure you understand where i'm going

Matej.

---

This may be a horrible question and show how much of a newbie I am but the one thing that I am struggling with the most is creating landing pages, I use spy tools like adplexity but have no clue what to take out of the code and where I should place my links. All that being said when I am working with an affiliate network, say Affiliaxe and they allow me to see creatives are those the actual offer pages or would those be examples of how I should design the landing page? I hope that makes sense as to what I am asking. I appreciate all the help that is offered on this site.

---

Hello,

here is a good thread to read about how to fix ripped landing pages, start there : https://stmforum.com/forum/showthrea...Ripped-Landers

---

Thank You...I will go through this and see what I can learn.

---

Thanks for a great tutorial! I've read it few times and lots of good info here. However, this part about cleaning up traffic is still a bit confusing. Can you please tell me if I understood it correctly?

So, for example, at the moment I am running 1 campaign on Ero-Advertising. My banners were displayed in about 120 domains, and about 240 ad spots. Got zero conversions, unfortunately. I am running carrier traffic only so bot traffic shouldn't be a bit issue (also, bot detect script also doesn't show anything bad).

I am running quite a low bid, of course. So if I understand correctly, I should be getting the "leftover" traffic here. So is my next step would be starting to cut domains/ad spots (this is what you call "placements" in your post, right?) that are performing much worse, than others? And in this case, since there are no conversions still, the only performance indicator would be CTR. And I would be cutting those, that have it way lower. Am I thinking right here?

If I am right, after I clean worst placements, I should increase my bid, to get better quality traffic? Hopefully, this would get some conversions, so I could filter out some more placements. Please correct me if I am thinking wrong here. And this actually takes me to the second part of my question.

Do you have knowledge, how the auction systems work in adult traffic sources? Does paying more gives you better quality visitors? If that's true, I am a bit confused, how is that working? Like for example on Facebook it's much more clear. FB has so much info about every user, that someone, who, for example, already bought something via FB ads, let's say, 20 times, is very highly ranked user, and showing your ad to such user would be much more expensive than to some random one, who hasn't bought anything even once. Does bidding work similar in these networks as well? If I outbid all others, I get the "best" user (based on info traffic network has about that user, that is he much more likely to convert) which is the most likely to convert with my offers? Am I thinking correctly here? And if not, could you correct me please?

As always, I would immensely appreciate answers to these questions. I know understanding more how traffic networks work, would help me A LOT in finally making my first green campaign (that last at least few days ). Just like 1 post from STM about how FB ads auction works (that a friend shared with me, before I was even a member) helped me a lot to improve my FB ad performance

---

Getting NO conversions is not the best situation, not sure how much traffic you got, but maybe you are testing a bad offer ... in this case you won't be able to do much with the campaign.

How low is it, the absolute minimum or somewhere in the middle? Try not to go for the bare minimum. The best way to find leftovers from SOLID traffic, is to bid a bid above the minimum and watch the amount of traffic you receive. What you want to see is small amount of traffic compared to the overall volume, even with a bit higher bid ... that indicates that people are bidding actively for this spot. If a spot sends a lot traffic even with the minimum bid, it means the demand is low. Cutting by CTR is no the best way, but when there are no conversions at all, it's the next best possibility ... do not go to aggressively at it, just try to get rid of the worst spots/placements.

Yup, once you got rid of the worst ones, you can try playing with the bid to get more volume and get some conversions rolling in.

As for the auction system in adult, most networks use a simple system ... higher bids get served first. This assures higher volumes and higher quality. The higher quality comes naturally ... people who just start browsing are easier to convert, so the banners served earlier during their browsing session, will yield better results. A typical surfer, who has been on some site for hours, is not likely to click on banners anymore, so the "late" impressions also result in poor CTR ... so the first few impressions per surfer are the most valuable.

Another reason is the experience level of surfers, the seasoned porn consumers know all the tricks, they know what the ads are etc ... while a new surfer comes fresh to a site and clicks on everything, including ads ...

So it's not like the networks know much about their surfers, it's a natural "selection" so to speak. The less experienced surfers click on ads more frequently, and they do it within the first few banner impressions of their surfing session the most.

Cheers,
Matej.

---

I still only sent very small amount of traffic, so gonna run for a while. Since I am not on those massive networks, and trying to play with smaller GEOs, it takes a bit of time to get enough traffic. So will still run for a while. But yes, offers or landers can be an issue. However, I wanna take care of traffic first. And, as I said, having the understanding how things work decreases the chance of making mistakes and wasting money.


Nope, it's not the minimum, I guess it's in the middle somewhere. Or at least higher than minimum.

That is really useful info! Honestly, would never have thought about this in such way myself. Thank you! BTW, what does it mean "a lot" of traffic? Like much higher, compared to other placements?

Would have never figured out myself! Thank you!

---

Great thread Matuloo. I've implemented bot traps on landers to see which placements end up with visitors clicking on hidden links. I struggle with when to know to cut the placement.

How many clicks would you advise evaluating before deciding on whether cut cut out a widget (running native) based on bot trap clicks?
And once that threshold is reached, is there a certain percentage of bot clicks you allow?

Don't want to cut too soon, but don't want to waste spend on the ones that feel like bot clicks with 5 LP clicks and all 5 click both the real offer and the bot trap.

---

When testing a new source or a vertical etc ... I usually let it run long enough to start seeing certain patterns. It's not possible to give you exact number of clicks, but what I'm looking for is anomalies. So if 80% of the placements has like 40-60% bots and then there is maybe 10% of placements with 80% and more, I know these 20% need to go. You will always see some % that will be the "average" values and some will be way higher and some way lower ... these are usually the ones with most BOTs.

In theory, any placement should have a bid level that reflects its traffic quality, in a bidding environment. But in some cases, the bot % are to high to make optimization possible, so it doesn't make sense to mess with those.

---

Great thread thanks

---

Home > >