$2000 Spend - 1 conversion - Loss $1950 - TABOOLA - help me please (9)
$2000 Spend - 1 conversion - Loss $1950 - TABOOLA - help me please
I test product in ecom.
Payout is $45
i run taboola.
I block push placement on taboola, and i start with device break apart - 1 desktop, 1 iphone, 1 android phone campaign.
First, I test 4 landing pages (everyone different angle), I determine highest landing page CTR .. total spend $300 ($100/campaign).. my avg landing page ctr was now around 23-25%
then I run winner, and test 3 more headlines on it.. after 50 clicks to headline, i pick one with highest ctr, which was 28-33%
people tell me I need data, so I let i increase budget to $300/campaign
I get only 1 conversion, $45 spend around $2000
when I say, I think offer is bad - they tell me I need more data...
It makes no sense, if i blow $10,000 and i get 3 total conversions what good is that? I am not made of unlimited money
so how to make native work?
Originally Posted by Atnakov S
A good offer will work even if everything else is bad - it may not be profitable from the start but that's why you make the choice to improve and focus on it once it shows some potential. 1 conversion with $2000 spent is terrible.
Which network is this who said you need more data? How many clicks to offer on the network, what's the offer EPC?
Plus, keeping landers based on ctr is a bad idea. You need to look for conversions and profit. CTR is a good indicator people maybe interested, but interest without taking action is useless.
Creative ctr is obviously important, you want the best from those. But landers can be profitable with <10% ctr but more conversions than a lander with 25% ctr.
Originally Posted by anthonyl
How to use it?
Originally Posted by Atnakov S
That is rough man... sorry for your loss...
I'm not sure what advice you got, but there is NEVER a time where you would need to purposefully go in the hole that bad...
If you spent 6000$ and got 4000$ back that might be okay....
Or if you spent 400$ with just one conversion that might be okay...
But there's no reason you should ever have to lose 2000$ to make 45$...
So yeah, there must be something off with the campaign... either the targeting is wrong, or the offer is no good, or there is a broken link, or the ads are wrong, or something else..
It's hard to tell what went wrong without more info... is this your own product or an affiliate product? What geo are you targeting? What was your bid and average cpc?
If you want to post more info would be happy to share any thoughts..
Anyway though, don't feel bad or be too hard on yourself, it's really tough to know when to cut things early on... I did the exact same thing on MGID when I started and spent 700$ with no conversions on an offer that in retrospect could never have worked in a million years, so you're not the first to have that happen to you
The thing with native is that each network is a unique "beast" and takes a ton of learning... the trick is to "learn" the platform as cheaply as possible by focusing on Tier-3 geo's, being conservative, cutting losers quickly, etc.
Next time I'd say pick an established offer that's definitely succeeding based on your spying on Adplexity, then run it in New Zealand or Denmark or Australia or Switzerland or somewhere with less volume but better margins, and then don't hesitate to cut it once you go a couple hundred in the hole and move on to the next experiment... Again that presumes you focusing on affiliate offers rather than running your own stuff though, so if that is the case I guess things are a bit different.
Note: If you're serious about native I'd also definitely consider the James Van Elswyck courses on it - they are pricey but well-worth it in my opinion, given how expensive it is to learn the traffic sources anyway (far more than the cost of the courses).
But yeah, happy to look at any additional info you want to throw out to try to diagnose what the problem was.
Originally Posted by jack_l
That is rough man... sorry for your loss...
I'm not sure what advice you got, but there is NEVER a time where you would need to purposefully go in the hole that bad...
If you spent 6000$ and got 4000$ back that might be okay....
Or if you spent 400$ with just one conversion that might be okay...
But there's no reason you should ever have to lose 2000$ to make 45$...
So yeah, there must be something off with the campaign... either the targeting is wrong, or the offer is no good, or there is a broken link, or the ads are wrong, or something else..
It's hard to tell what went wrong without more info... is this your own product or an affiliate product? What geo are you targeting? What was your bid and average cpc?
If you want to post more info would be happy to share any thoughts..
Anyway though, don't feel bad or be too hard on yourself, it's really tough to know when to cut things early on... I did the exact same thing on MGID when I started and spent 700$ with no conversions on an offer that in retrospect could never have worked in a million years, so you're not the first to have that happen to you
The thing with native is that each network is a unique "beast" and takes a ton of learning... the trick is to "learn" the platform as cheaply as possible by focusing on Tier-3 geo's, being conservative, cutting losers quickly, etc.
Next time I'd say pick an established offer that's definitely succeeding based on your spying on Adplexity, then run it in New Zealand or Denmark or Australia or Switzerland or somewhere with less volume but better margins, and then don't hesitate to cut it once you go a couple hundred in the hole and move on to the next experiment... Again that presumes you focusing on affiliate offers rather than running your own stuff though, so if that is the case I guess things are a bit different.
Note: If you're serious about native I'd also definitely consider the James Van Elswyck courses on it - they are pricey but well-worth it in my opinion, given how expensive it is to learn the traffic sources anyway (far more than the cost of the courses).
But yeah, happy to look at any additional info you want to throw out to try to diagnose what the problem was.
it was unproven offer yes.
about JVE course, does he give blacklists? seems that is most important here
Really sorry to hear your awful experience in this test!
Some affiliate managers will most likely tell you that offer X and offer Y is killing it, even when in reality these offers suck.
On the other hand, there are some other affiliate managers that really want their affiliates to succeed and provide them good offers and information about the offers.
A good approach to avoid such situations like the one you experienced, it is good to always do some research about the offer, before starting to send traffic to it. Here are a few of the,.
1. Ask your affiliate manager to provide you some stats about the offer. Network EPC, number of clicks, conversion rate and if possible the networks where these offers are being promoted (native, social, search, etc..)
2. Check what's going on with the offer on spy tools like Adplexity Native. Search by Advertiser > Offer Page URL. A good well-tested offer will show plenty of results as well as give you a rough idea of the best creatives and networks it works best.
3. Check out the overall traffic volumes of the offers. A simple search on the free version of Similar Web will be just enough to create an idea if the offer is getting enough traffic or not. If you get results like 5k-10k visits over the last 3 months, most probably the offer isn't that good.
Based on the information you get about the offer from your affiliate manager, you will be able to quickly estimate the amount of clicks you need to send to the offer in order to make it work. A simple calculation will tell you how much traffic you need to buy and send to the offers.
then I run winner, and test 3 more headlines on it.. after 50 clicks to headline, i pick one with highest ctr, which was 28-33%
Keep present that as you add more variables to your testing, you will need to allocate more budget into it. So keep it simple. A max of 3 landing pages for the initial test is more than enough.
In the initial phase of the campaign you simply need to test the offer in order to see if it shows any promise or not. A single lucky conversion is not enough to run heavy optimizations.
Also consider this:
Let's assume you are killing sites and ads based on your first version of the landing page. Among these sites that you have blocked, there are also some well-known top sites.
Later on, you introduce some impactful changes to the landing page that can double the landing page click-through rate and conversion rate. Once you have introduced these changes, odds are you might want to re-test a few of these top placements, because you were optimizing based on a lower landing page ctr and cvr and not an optimized lander that can turn these placements profitable.
1 - Offer might be no good. What's the offer?
2 - What are you doing optimizing towards LP ctr goal? Your goal for LPs is conversions only!
3 - Set hard blocks on campaigns - you should have a threshold of ROI & spend to pause a campaign outright, say under -50% ROI at $500 spend.
Well, I would suggest do not wast 2k as test budget in one shot, for $45 payout stuff.
Until your campaign starts to perform, you need to process with different AB testing of material, banners, landers etc.
If your cvr sucks (provided your ctr etc is within kpis) and you are running on a whitelist, it means the offer is bad. The first indicator that an offer works provided those kpi mentioned earlier are in place, is the number of sales. If you spend 2k and get 1 sale, then the offer sucks
Home > Paid Traffic Sources > Native