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

---

newbie stumbles upon a promising campaign... but struggles to move forward (21)

---

Hey everyone,

I recently started a bunch of new campaigns to help me on my quest of finding my first profitable one. One camp stood out so far but I'm not sure how to optimize it and make it profitable.

Here's the stats so far:




Traffic: Pop from Zeropark with Bid: $2.3
Offer payout: $2.3
Running time: 2 full days
Country: UK
Tried a duplicate of the campaign with different bid $6.1 but it did really bad.


I know (or I think I do) that at this point I should be optimizing my camp by blacklisting the bad placements/zones that has 2X the payout with no conversions.

But NONE of the placements has spent that much yet... - not even close since there are a loooot of placements coming from zeropark.

Here's the placement stats:



Any advice or recommendations on how to move forward with this campaign ?



Best,
Nicolaos

---

Hey Nicolas, for how many days has your camp been running?
Are the above stats from one day only or from a selected date range? Is $2.3 CPM your starting bid or you have increased it progressively?

One thing you should pay attention on targets is Landing page CTR.

---

Hey @platinum
The campaign has been running for 2 days now.

I did another campaign with CPM of 6.1 but didn't convert well. (only one conversion. Is that weird? I mean its higher payout shouldn't it be better quality)?

Here are my lander stats btw:


I guess the Lander1 and Lander 2 has good CTR. I have no real reference to benchmark that. So what do you think ?
As for Lander 3, it has shit CTR, but it converts. The reason is that its two steps (one to scan their PC and Second to claim their free download) so i guess every click has high intent of actually downloading and installing the offer.

Thanks!!

PS: I'm updating the main post with data from your questions.

---

One thing I would do is not to over-judge any clicks, as time goes by and you launch more campaigns you'll notice that a click won't tell much of the user intent rather than the placement's quality.

I wouldn't wait for a placement to reach 2x the offer payout before killing it as long as it has no LP clicks. By this I mean a placements with more than 300-500 visits that is not showing any sing of user interest of clicking on your landing page (without recording a single click) won't go far for you. This by assuming that you are using a tested lander.
If you hadn't cut any placements before duplicating your first campaign and running it with higher bid, you were just buying cheap traffic at a higher price, so when it comes to bidding as you may notice on other threads and guides as well, is to start bidding low than progressively rise the bid as traffic volume starts to lower down.

---

Yep, makes sense! So a better conversion on a particular landing page is only caused by a better quality placements and not because of the intent of the users due to the lander’s copy ?

Or is it both ?

Yep this strategy makes a lot of sense! Better than cutting at 2x the payout. But my question is...
if the average CTR of a landing page is 13%-15%
Wouldn’t it be enough proof to cut off a placement if it didn’t get 1 click within only 150-200 visits?

I mean it has no chance of reaching anywhere near that average CTR, why keep it ?

Maybe even use the binomial confidence calculator to make sure that the maximum predicted conversion of this placement couldn’t possible get higher than the average CTR.

Is that the right way of approaching this ?

Yep, looks like I did that mistake. So Next time ill start with low bid and as i cut placements ill duplicate the camp and increase bid. Right ?
Ill read more about this too. Any recommended posts or guides ?

Many thanks Platinum ! Very valuable feedback

---

Don't miss understand my point, if you didn't have a convincing angle on your lander the visitor most probably would have never converted. Your lander's copy is your best weapon you can have.
It would make sense, but it depends how aggressive you'd like to be when cutting placements. Considering that you have a high payout offer, you might be with 10%-20% ROI on that specific placement, meaning that is still profitable.

Try sending some more traffic to that campaign with the bid you had previously and check out the results. In the first days of your camp you will have to cut out lots of placements and ZP has enormous traffic volume.

---

Thanks ! I'm going to do this and update my results !

---

Here's a quick update ...

I sent more traffic to that Camp and here are my current results.




I've already spent almost 30X the campaign Payout of the offer and my ROI is -84%.

Basically I had NO CHANCE at optimizing this campaign.

All I did was able to do is cut out 9 placements around the time where I had spent around 20X payout.

Here are my placement stats: (I highlighted the ones that I was able to cut out because of very low CTR)

All rows below this screenshot have very little data to help me make any decision.

In total, I have over 5300 Placements for this campaign (but most of them onlyhas 1 or 2 visits).

I think its very little to cut out only 9 Placements based on what I've been reading on STM. But I cannot cut out other placements because I do not have enough data to make more cutting decisions.

What do I do in scenarios like that ??? It is something I'm facing with every campaign I ran so far.

1. I cannot cut out a good amount of placements because I have no enough data.
2. I have already spent 30X the payout and my ROI is -84% (understandably bad because I have no optimization whatsoever)
3. All my spending was split between 5000 placements giving me no real data that enables me to make cutting decisions.


Thanks in advance !

- Nicolaos

---

This is one of the issues with higher payout offers. At $2.3 a conversion it's going to require a lot more spend just to test the waters and see if there is potential. Spending $60 to get the data of 4 conversions is quite dear; in theory, if the offer was $0,10 payout you would have had nearly 100 conversions and be able to make much more sound decisions etc.

I cannot comment on how to take it forward from here as I'm still very new myself.

---

Yea maybe not enough data!
Hopefully someone experienced with this can clarify or recommend an optimisation strategy for such a campaign with 30x payout already spent.
Also, I’m not sure if we can really consider 2.3$ a high payout offer. In Vortex’s newbies tutorial she recommended payouts of 1.5$! Im assuming 2.3 isn’t that far from that range !
Thanks for your input craig88

---

Yeah I guess it's not high in the grand scheme of things but I just enjoy the lower payout offers to get as much data as possible. I can cut my losses quickly if I realise the offer isn't working and I can test it on 2 different traffic sources at the same time without blowing through too much budget.

But it is frustrating when my revenue only climbs $,1 for each conversion haha

---

Haha ! Yea exactly, the reward for each conversion doesn’t release that much dopamine when its only 0.1$

You’d have to get a whole lot of conversions to make good money and there might still be the risk that there isn’t enough traffic coming from the placements that were profitable for your camp !

---

Just to check, did you use spy tools like Adplexity as a basis for landers to use in this campaign?

---

Yep. My LP's are from Adplexity but tweaked the copy to be better.

They're performing at 15% and 16% atm.

---

Nice FA! First of all, apologies for the extremely late reply. Been playing catch up on work and sleep since I got back from my trip to China (where I lost my laptop power cord and had trouble finding a suitable replacement...)

Enough excuses - here are my suggestions:

1)UK is an expensive geo with lots of traffic and therefore a ton of placements. In order to conquer this geo, you'll need to be willing to invest quite heavily into testing offers and landers, and also into identifying placements to cut. If you're not experienced, it's recommended that you pick another geo to start.

2)Zeropark is one of the best pop sources in that it has nice traffic volume, but because the quality is mixed (since they're brokering from a wide variety of networks - which is where the volume is coming from), it's not the best source for testing. Try to test landers+offers+targeting at a source like PopAds that has good quality but limited volume first, then scale the best stuff to Zeropark and focus on testing bids and cutting placements there.

3)As mentioned - test more landers and offers! The following post contains a rough approach:

https://stmforum.com/forum/showthrea...l=1#post324392

4)Even while testing offers and landers, you can start cutting placements. Traffic volume isn't your current issue - it's having a ton of placements draining your budget. I would suggest to cut placements that are at 1x payout in loss or even 0.5x payout in loss. Once you've found a promising offer+lander+targeting combination, you can always retest the placements you've cut.

5)Cutting based on CTR is inaccurate, but if it's really different from the average CTR then it's justified (plus the fact that you're dealing with so many placements and need to cut fast). I definitely agree with your cutting decisions as per your screenshot. No need to use stats in this case - using stats to cut pop camp placements seems overkill to me since there are so many of them. Just exercise your best judgement.

Don't think too much about cutting or not cutting the smaller placements. For bigger placements, either run more traffic before cutting (so you're more certain they're bad) or if you're not sure, cut and then retest later when you have better lander+offer+targeting.

6)As for bids: I would suggest to start with average bid to make sure you're not getting junk traffic. This is important when you're testing unproven landers and/or offers. If you're not confident in the traffic quality, when you don't get results you won't know whether it's your landers/offers not being good enough, or if it's because your traffic quality is shite.

So then what's average bid? Some traffic sources have estimators that will tell you. If you're running on a source that doesn't have an estimator, try $1.33 for wifi and $2.33 for carrier. (The ".33" is to avoid the heavy competition of newbies trying to bid at whole dollar amounts, and also other affiliates that are ALSO trying to avoid that competition by bidding .11 and .22. We may need to start bidding at .44 soon!)

You can and should test higher and lower bids once you've identified a promising offer+lander+targeting combo.

7)User intent has been mentioned. I wouldn't worry about that, as pop traffic is basically general, broad-intent traffic that consists of anyone and everyone. You can't really segment this traffic aside from maybe targeting different categories (and only if your traffic source has that option), but even those are normally far from accurate. So, when designing landers etc., it would make a lot more sense to go broad rather than niche, i.e. cater to the masses. Checking google hot trends for the target geo for example may be a good approach.

One example: When I was dabbling with sweeps way back when Paul Walker had just passed away, I used the surveys lander to ask questions about Paul and the Fast and Furious movies and got good results.



Now that I'm all caught up on your FA posts - looking forward to seeing the next update!




Amy

---

I just saw that as I was traveling and wow ! That was exactly what I was looking for.
I read that post over 4 times as it clarifies soo many thing.

Thanks a lot Amy for these great points.

Here's my action plan:
I'm going to change Geo, since im New and UK is not newbie friendly.
I'm going to stop testing offers on zero park, and start testing on Popads and only bring the winning offers to zeropark and test out different bids.
I'm going to test out much more landers and offers.
If placements are very high in number and draining my budget, i'll cut them even at 1x or even 0.5x payout loss. And if it makes sense i'd cut placements doing very poorly in terms of CTR relative to the average.

One last thing...

When testing different bids, you mentioned...

Is that 1.33 for wifi and 2.33 for carrier a generally good average to start with with ALL geos ? or its only suitable for tier 2 or tier 3 ?

Thanks again, that was superb.

---

Glad you found that useful!

These are averages across all geos.

Meaning, $1.33 may be too high for cheaper geos and too low for more expensive geos, but bidding this amount usually won't give you trash.

Of course these values are just a last resort. It's better to check the traffic estimator on the specific traffic source you're running on, or failing that, check the traffic estimator on another pop source and go with that.



Amy

---

Superb, Noted !
Thanks Amy !

---

One additional point I'd note: if an offer's not converting after a reasonable multiplier of its payout (I usually use 4x for desktop and 7x for mobile, but you can fine-tune by traffic source) I'd recommend stopping the campaign, regardless of how many placements you have or haven't cut by then.

If your clicks are spread over a wide variety of placements, and the offer/lander combo still isn't converting, that's statistically really strong data that there's something wrong with either the lander or the offer, because the usual confounding variable of "single large placement that ate all your cash" isn't there.

Always remember this rule: most offers won't work out for whatever reason.

Also, in a situation like this I'd really strongly recommend bot-testing the traffic source before you start running your campaign in earnest. There's a good chance you can cut out a lot of time-waster placements right off the bat that way.

---

Wise words as always!

This is why it's almost always better to test lots of offers on a low'ish budget per offer, and pick the gems to test further - as opposed to using every trick in the book to make a particular offer WORK.

Applying all those tricks can be very valuable after finding a good offer.



Amy

---

Thanks for the awesome points @caurmen
Im very guilty of trying to force an offer to work by over optimising instead of moving on and acknowledging that many wont work and i need to test different offers!
Ill stick to the 4X and 7X rules when cutting placements + I just saw you recently released a bot testing tutorial, I’m implementing that asap !
Thanks again

---

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