How to get maximum profit from your traffic (10)
Introduction
I'm working now with a lot of different verticals, offers, networks. And sometimes for some GEOs I have too many offers, therefore it's becoming harder to do my tests, cause I need to create a lot of campaigns for different verticals, put offers/landers to each campaign. For some countries I was spending about one week to test all of these offers/creatives I have and to choose the best ones.
Then I started to optimize my campaigns, cut bad targets (OSs, Browsers, Carriers etc...), bad placements, and sometimes I had the situation when on the first day I had 100k visits, and on the 3rd day it was already mot more than 20k. Why it was happening? Because I was cutting so much targets/sites. Also my traffic became expensive, cause in some source the more narrow targeting you have, the more should you pay. In this article I will tell you about my new way of testing offers on POP traffic which can help me to buy a lot of cheap traffic, but with low cost.
Narrow Targeting
Okay, in the 1st tests of our campaigns we almost always use Wide Targeting. Then after our test we should take a look at our stat and see which targets should we turn off and which sites are showing us a bad ROI. Then our campaign is starting to get traffic only from "good" targets and zones. So your ROI is becoming better, and you see the profit. So it's a typical way of testing. Why should we pay for traffic which doesn't converts on our end? For sure we shouldn't pay!
Advantages:
- You buy only the traffic that works good for your current campaign.
- Your ROI can be really high (300%+) after you optimize it. So the difference between your spending and your revenue is big.
- You don't pay much money for the traffic, and you can continue with a little cashflow if your ROI is good.
- You don't have to pay big costs for the services. For example, for the tracker. Some trackers like
Voluum /Thrive can be not so expensive for you, if you have small campaigns with big ROI. - You don't need powerful server for your landers.
- You save your money, and you can test more different campaigns.
Disadvantages:
- You don't make big numbers, especially if the GEO is small.
- You have a very little audience.
- As you have the little audience, in some days your CR will be dropped, cause some people already saw your advertisement.
- You're not a VIP member in the sources/networks, cause you don't show big spend/revenue , so you can't have some special "things".
- You should change your creatives often, because it's burning so fast.
- Often these campaigns don't live long.
Wide Targeting
If we talk about this type of targeting we mean to buy all traffic for our campaigns. It can be all OSs, Browsers, Carriers, Platforms (WEB+WAP), publishers. At first sight it's much more easy, than narrow targeting, because we don't need to do a lot of changes in our campaigns, we have a lot of traffic, and we can work with big budgets. But it only works if you have a really good offer/LP. If so - all traffic will show you a green ROI. ALL! But it' happens only in 1% of all cases - when you find this "GOLD" offer. For mostly people this type of targeting doesn't work good.
Advantages:
- You buy a lot of traffic, haha you can really feel your influence.
- You have big discounts in sources, because you buy all traffic they have. In some sources there are absolutely different price for camps with wide targeting and with narrow. And sometimes it can be a very cheap price for your wide-campaign! For example, I'm buying traffic from GEOs like NO/SE/US/DE/AT/SK/UK/ZA/GR with $0.10 CPM! And it's also 3G traffic is there!
- You have better relationship with your affiliate network, cause you're generating big money for them. It means: Fast payments, bumps, private offers, secret tips etc...
- Every day you have many fresh users, who didn't see your LPs (or biggest part of them)
- You can work with one creative for a long time.
- You have a lot of data. And if you decide to run narrow targeting, it will be very easy to create another campaign with best targets/zones.
Disadvantages:
- As you're buying a lot traffic, you should have a big cashflow.
- Sometimes you don't have a big ROI, cause you still have some targets/pubs that show you bad results in your campaign
- You need a good server/CDN for that.
- You need to spend much more money for the tracker. In my biggest month I spent about $10k only for my tracker...
Algorithm
1. Choose a GEO
Of course, I would recommend to choose Tier 2 GEO. Because Tier-1 can be too expensive for you, also there will be a lot of competitors. If we talk about Tier-3 countries, we should understand, that there are a lot of traffic, but you can't find many good offers, or even just one good offer. There are some countries I potentially can buy a lot of cheap traffic ($0.05 CPM), but I just don't know what place should I send those users!
A good way is to choose some EU countries.
For example these ones are working so good now for me: GR/AT/DE/IT/CZ/HU/IE/MK/BG/SK/SI/LT/EE/PT/BA/PL/NO.
Also you can choose some of them: IQ/SA/IR/AE/KW/QA/HK/MY/TR/SG/CO/CL/MX/ZA/KZ/RU/BH/JO/NG.
2. Try to find all verticals in this GEO
We will be testing several verticals together. And we should have a full list of verticals.
Which verticals I'm always trying to find?
- Anti-viruses (We can also push "Download Content" offers with AV landers - if offer allows it!)
- Sweepstakes (It can be iPhone X, Samsung S9, Vouchers from some popular local shops). If we talk about flows, we can take: PINs, 1-click, SOI, CC-Submit.
- Installs (Make sure it's an app for the large audience, because we can't promote some highly specialized apps throw POP traffic. Browsers, Tools, Anti-Viruses are ok).
- Dating (As a big part of all POP traffic is Adult, why we shouldn't test it also? We should!). Try to find SOI dating instead of DOI. Dating apps are also good.
- Adult (We can also add some 1-click or PIN adult offers to the split test, sometimes it works good).
If you can't find some verticals, I recommend to create account in at least 3-5 networks, and you will find them! I recommend: Clickdealer,
3. Add offers for each vertical
I recommend to take for the test at least 3 offers from each vertical. But I recommend to run 5. If somewhere you have only 1-2 offers for the GEO+Vertical - it's absolutely OK. I have many countries where there is only 1 offer, but it converts like crazy. If it's only 1 offer there, it doesn't mean that this GEO is bad, and you shouldn't test it.
On the other hand if you've found 10-20-30-1000 offers, ask your AM to give you 3-5 best ones. You can ask him to give you, for example, 3 offers from TOP (With big Revenue) and 2 with a good EPC (But not big Rev). Sometimes offers not from the TOP work much better than TOP ones. Also the TOP can be not real
I know exactly, cause I used to work as AM in some networks.Important! If you see that some of your offers accept different carriers, but it's PIN/MO/Click2SMS flow - we can take all of them. Sometimes offers which have only 1 carriers work better than offers with all carriers.
4. Create Landing Pages for each vertical
You don't have to create your own LP. Just go to the Adplexity.com and see what people are running at the moment in the current GEO+Vertical. You will find all TOP landers (Usually it will be like 2-3 different landers, and 100500 absolutely the same ones, because people are lazy, and they don't want to create their own ones, but it's enough for the test, then we will create our own LPs).
Try to get at least 3-5 LPs for each GEO+Vertical. So what works ATM:
- AV: Google landers with timer and blue button. You can use it for Download Content offers and Installs as well. Just if you use the installs you should change your text from "Click the button and enter your phone number on the next page" to "Click the button and download the free app"
- Sweeps: FB Wheel, Google questions, green LPs with caskets, YouTube questions.
- Dating: Questions with photos of girls (not naked, just a bit sexy)
- Adult: Imitations of Tubes (A page with a lot of previews, and it's no difference when the user clicks, he will be redirected to the offer's page).
5. Create Campaign in your tracker
We should add all these offers and LPs to the tracker. Then we need to create a campaign for the source with all this stuff. We should create separated paths for each vertical and put there all offers and LPs. For example you will have sth like that:
- LP 1
- LP 2
- LP 3
- LP 4
- LP 5
- Offer 1
- Offer 2
- Offer 3
- Offer 4
- Offer 5
- LP 1
- LP 2
- LP 3
- Offer 1
- Offer 2
- LP 1
- LP 2
- LP 3
- LP 4
- Offer 1
- LP 1
- LP 2
- Offer 1
- Offer 2
- Offer 3
Be careful! If some offers allow only current OS (for example, installs), we should create a rule in the tracker and move other OSs from this path. The same with Carriers and WIFI, if we talk about 1-click offers
6. Create Campaign in the source(s)
For the 1st test I recommend to choose these sources:
- Plugrush
- Popads
- Popcash
- Propellerads
- Clickadu
Also you can try:
- HillTopAds
- Adsterra
- Exoclick
- WWWPROMOTER
- CitenkoMedia
- PopMyAds
- Redirect.com
- ZeroPark
- AdCash
- TONIC
We should create a campaign with large targeting (All OSs, Browsers, Carriers, Languages, Devices type - Smartphones&Tablets, 3G+WIFI).
As for price, we can set $1.1 CPM for all EU geos and for rich Asian/Latam/Arab/African geos like: SG/AE/ZA/HK etc... For all others we can set $0.71 CPM. Why not $0.7? To keep a better place in the auction
What budget should you set for the campaign to get enough data for optimization? I recommend to spent $25 for each vertical. So If you have 4 verticals, it will be 25x4=$100.
But keep in mind, if you have offers with $7+ payouts, it will be hard to test them for $25. In this case I recommend to set (average payout)*10.
For example if you have 3 offers: $4, $5, $2.
(4+5+7)/3 = $3.6
$3.6*10 = $36.
Then we can calculate recommended cost for each vertical and just count how much we should put to our campaign's budget.
Optimization.
After our test budget is spent we should start our optimization. But in this method we won't cut targets, we will just move it to the best Path (the best vertical).
The first our thing - is to cut all bad offers/LPs which didn't give us any leads or very bad ROI. Let's keep 1-2 best offers/LPs for each vertical.
Then what should we do? Let's look at all targets all check which Path showed as the best results. We will leave it for this target. For example first of all let's check Device types. If can be the situation when Dating showed good result with Tablets, and all other verticals not. Let's create a Rule in the tracker and send all Tablet traffic to this Path.
Another situation... Some versions of Android can work good with AV Path, and at this time some others with Sweeps. Let's create a rule again and separate this traffic.
So after all optimizations you should have sth like that:
Path #1 - AV:
Android 5.0, 6.0
Chrome
ex. Tablets
Path #2 - Sweeps
All android (ex. 5.0&6.0)
Android Browser
ex. Tablets
Path #3 - Dating
iOS
Safari
Tablets
Path #4 - Installs - For example, it's not converting. We will turn it off.
So I think you understand my logics.
We should create rules for each targeting option and send this traffic to the best place = the best Path. And then we can continue to run the traffic, and our tracker will automatically analyse the user and send him to the best place when he will convert with max probability.
What should we do if we don't have a good Path for the current Slice (Combination of targets)? We can just send this traffic to some smartlink networks (Monetizer). And this traffic can convert there.
Conclusions.
This method works very good for people who have good cashflow for tests and have access to at least 2-3 networks (to be able to find as much verticals and offers as you can). The main reason why I'm doing it now - I pay less and get more. You will see that sources can do discounts for you, but even without it you can buy very cheap traffic. Your camps will be very stable, cause you're buying all slices for many verticals. And if an offer stops, you will just send this traffic to the different offer or Path, so your campaign won't be stopped.
If we talk about publishers, of course, you can block absolutely bot-sites, but for all others you can find them an offers. If not, you can just send this traffic to Monetizer.
You're final profit will be much more better than you run narrow-campaign with one vertical. And also it's better to keep all verticals in 1 campaign and not to create a lot of camps in one source.
Be careful! See that your rules do not contradict each other.
Good luck
OMG - another knowledge bomb from MrBraun!!
To recap your approach:
1)Target very broad - basically the entire geo. This will enable you to get traffic for cheap, plus some networks will place your campaign in higher priority and assign more traffic to you.
2)Compile a list of offers in various verticals. (Here, it would be beneficial to do research to identify which verticals and even specific offers are working well in that geo at the moment - Adplexity should help.) Prepare landers for all verticals.
3)Set up paths in
4)Start traffic, and analyze stats to identify which verticals/offers are converting the best and worst for each traffic segment, and set up tracker rules to direct traffic away from the worst and toward the best.
5)For traffic segments where no vertical/offer is looking promising, send the traffic to a traffic monetization service (such as Monetizer/Afflow).
This approach will allow affiliates to completely take over and own a geo more effectively than by testing one vertical or carrier etc. However, newbies may not have the heart to spend the money required for this type of testing. But I would encourage everyone to give this approach a try!
Amy
I just read the post again - there's only one thing I have a question about:
Cause if you really can get traffic for that cheap, I'd love to know how you're doing it!
Amy
Originally Posted by vortex
Is this a typo that should be fixed? I've never seen traffic THAT cheap anywhere! Did you mean $0.50 CPM maybe?
Cause if you really can get traffic for that cheap, I'd love to know how you're doing it!
Amy
It's really $0.05 CPM.
There is no secret. Just if you're doing a good spend in the source, they can offer you a good CPM for some GEOs, especially not popular. I can give only one tip - Try to keep good relationship with them
Originally Posted by mrbraun
It's really $0.05 CPM.
There is no secret. Just if you're doing a good spend in the source, they can offer you a good CPM for some GEOs, especially not popular. I can give only one tip - Try to keep good relationship with them
I've been offered exclusive deals from traffic sources before - but never as low as $0.05 CPM! Guess I gotta step up my game!
To the rest of the community: Mrbraun and I just had a chat on skype. The secret to getting such good CPMs is 1)spending a lot of money with a traffic source, and 2)nurture good relationships with their reps.
Some tips on nurturing good relationships that I've heard other people do (which is not really my style but it works) are: Sending gifts to reps to show appreciation (got to be careful about this one though - you don't want to get them in trouble for accepting bribes), and visiting them at their office. Basically when you give someone special treatment, chances are they would want to return the same.
You hear time and again that in this industry it's not just WHAT you know, its WHO you know. This is so true.
To me, this is the most valuable takeaway from mrbraun's post. Without the ability to get low rates, testing so extensively can really cost you.
Amy
WOW, This is a very clever approach MrBraun!
Thank You for the GEOs and Network tips as well 
Thank you so much for great tips.
One question: Can we use multiple geos in one campaign in
Second Question: Can I use Multi-Offer Click URL in my all landing pages instead of click URL?
This strategy is indeed very interesting. Mind sharing the numbers (well not the profits of course)? I would really like to know daily spendings and traffic volumes
And how do you keep up blocking bot traffic from that amount of traffic sources?
mrbraun, you are super man, tks vortex for summarising for newbie such as me. But i do not understand how to test 5 landers & 3 offers at the same time. How traffic will run into each landers & offers?
Ex: I have 100k trafffic source & have 5 Landers (A,B,C,D,E,F) , 3 offer (a,b,c) .How we combine each Lander for earch offer? Each Lander point one offers?
Can you explain clearly how to combine Landers and offers, and how traffic source will run into each ?
Thank you .
Sorry for my bad english
Originally Posted by thang1003
For example: All offers are iphone x sweeps, and all landers are designed to promote the same.
Then, you would add the landers and offers to your tracker.
Then, you would create a new campaign in your tracker, and specify those landers and offers in the campaign settings.
Finally, you would save the campaign, take the campaign url and set up a campaign on a traffic network to start sending traffic to the url.
What would happen is each visitor will get a random lander and a random offer. And the tracker will rotate evenly through all landers and offers you specified.
For example let's say you're running 3 landers - LanderA LanderB LanderC - and 3 offers - OfferX OfferY OfferZ. The 1st visitor may get LanderA + OfferX, 2nd visitor may get LanderB + OfferY, 3rd visitor may get LanderC + OfferZ, 4th visitor may get LanderA + OfferY, 5th visitor may get LanderB + OfferX....
Hope you get the idea!
Amy
Sent from my SM-G930W8 using STM Forums mobile app
Home > Paid Traffic Sources > Mobile