Continue Or Pause This Capaign?
- Thread starter endriul
- Start date
Graybeard
Well-Known Member
Maybe, where are the conversions in what GEOs and what percentage of all of that traffic.
the COD for nutra No --why is it COD only?
These offers cost the offer owner nothing in ad expense --you are paying that --exactly why should I bear that cost is the first question I ask myself.
Graybeard
Well-Known Member
i am not going to type all the math from an image --too much work for no gain.
I am going to assume this header is correct?
1,2,3, are the country ISO codes i am assuming
not enough volume to be meaningful (as a sample) conversions have a way of grouping in small volumes (volatility)
if each GEO had 5,000 visits you could make a statistically accurate 'projection'
calculate your epc (earnings per click) and get all the GEOs to a sample of at least or greater that 500 then calculate your epc again then decide. the epc should be 1.2 -- 3x the cost per click to pick some winners.
In overview:
I am going to assume this header is correct?
1,2,3, are the country ISO codes i am assuming
not enough volume to be meaningful (as a sample) conversions have a way of grouping in small volumes (volatility)
if each GEO had 5,000 visits you could make a statistically accurate 'projection'
calculate your epc (earnings per click) and get all the GEOs to a sample of at least or greater that 500 then calculate your epc again then decide. the epc should be 1.2 -- 3x the cost per click to pick some winners.
In overview:
Graybeard
Well-Known Member
I would make a clone campaign --then change the URL location to the new landing pages being tested before submitting the campaign.
This way you could compare the results direct v. a bridge page that a click it to the actual landing page of the offer.
IMHO, I don't believe you can sort the value of a siteID by using your own Bridge Page (pre-Lander --same thing).
I sort SiteIDs by the reject rate of the bots and dinosaur browsers used. Any real people can buy. Maybe with very high volume 30,000+? visits you may be able to trend by the siteID.
When I was beginning my selling career, almost 50 years ago now, I remember getting some very large sales I made at some really crappy looking factories. Funny thing is that many people/businesses keep a low profile but have a lot of money in the bank to spend --you just have to know how to sell them
You can leave a lot of conversions/sales on the table by prejudging your opportunity too.
My own rule of thumb is: if the reject rate of a site (reject rate being of the bots and dinosaur browsers used) is > 33% I blacklist
--Unless the volume of traffic from that siteID is heavily weighted in the whole of the traffic
--then 40% or more might be the cutoff depending on the cost per conversion --the EPC.
But it's all relative:
That is not the usual. Your epc *mileage* may vary with the pay-out return of what you are selling. My returns are CPS rev-share on the new signups --it's a different animal altogether than a CPA one shot deal.
This way you could compare the results direct v. a bridge page that a click it to the actual landing page of the offer.
IMHO, I don't believe you can sort the value of a siteID by using your own Bridge Page (pre-Lander --same thing).
I sort SiteIDs by the reject rate of the bots and dinosaur browsers used. Any real people can buy. Maybe with very high volume 30,000+? visits you may be able to trend by the siteID.
When I was beginning my selling career, almost 50 years ago now, I remember getting some very large sales I made at some really crappy looking factories. Funny thing is that many people/businesses keep a low profile but have a lot of money in the bank to spend --you just have to know how to sell them
You can leave a lot of conversions/sales on the table by prejudging your opportunity too.My own rule of thumb is: if the reject rate of a site (reject rate being of the bots and dinosaur browsers used) is > 33% I blacklist
--Unless the volume of traffic from that siteID is heavily weighted in the whole of the traffic
--then 40% or more might be the cutoff depending on the cost per conversion --the EPC.
But it's all relative:
- my class1 premium signup is worth perhaps $30 to $60; and,
- my class2 freemium signup might be worth $2 to $8.
That is not the usual. Your epc *mileage* may vary with the pay-out return of what you are selling. My returns are CPS rev-share on the new signups --it's a different animal altogether than a CPA one shot deal.
Graybeard
Well-Known Member
I was a field salesman for years face-to-face
You are best off to use a filtering service adscore.com clickease.com
Let me know when you can administrate a server in ssh --that takes a long time to learn and really know what you are doing however
endriul
New Member
I've updated my campaigns, all related to adult content. Currently, I'm running three offers.
I've divided the campaign for two of the offers:
For optimization, I've focused on:
Btw: Fist campain is direct link then i put lander (i know I know this is huge mistake)
Let me know what you think guys
I've divided the campaign for two of the offers:
- Using 2 traffic sources
- Targeting 2 devices: Mobile and Desktop
For optimization, I've focused on:
- Creative CTR: If an ad has more than 10,000 impressions and a poor CTR, I pause it.
- Landing Pages: I retain the winners based on the highest conversion rate (CR) and click-through rate (CTR).
Btw: Fist campain is direct link then i put lander (i know I know this is huge mistake)
Let me know what you think guys
Attachments
Graybeard
Well-Known Member
what offers what payouts?
I think you are scraping bottom (profit margin) so far --either the traffic source is awful or the offer is unappealing.
Hate to be negative but something is wrong.
1. all offers are not the same
2. there are peculiarities to adult type traffic also.
Cut any 100% loss `line/row` off when it reaches 700 impression with no return
this is what I got 2024-02-14 to 2024-03-10:
; 4/120000
~0.0000333 `referrals pops -direct` gross count
(very deceptive registrations/traffic network stats --what a tracker might show without any filtering)
about 98,400 filtered click out (my server filters known datacenters mainly)
24000 / 98400
~0.2439024 `ALL sessions (GA4)` Google-Analytics filters i show session 0s,1s ... so most traffic (0.7560) is bots or blocked (somehow?)
4932 /4
1:1233 /`engaged sessions (GA4)` /registrations
real crap --first of all my ad sources (4) were very poor
I linked to pages on my white label webcam site. I though they were pretty good pages and about 20% of the real people traffic found the traffic worthy of engagement. But based on the statistics that I remember from 10 years ago
4932*.02
98.64 <the free registrations should have been (expected to be) near 98 or so, 70 -100 maybe, so there is a general problem
Next month I will try PPC or maybe HTML5 and landing pages and test the differences (if any).
endriul
New Member
I run SOI (Single Opt-In) offers with payouts ranging from $1 to $3 in tier 2 countries. The traffic source is well-known adult network banner ads.
When it comes to the offer, I use spy tools to track what others are doing. I create different funnels, meaning I design various creatives leading to landing pages. I don't copy.
I'm confused about what to do next. If I create another campaign, the results will probably be the same :/
When it comes to the offer, I use spy tools to track what others are doing. I create different funnels, meaning I design various creatives leading to landing pages. I don't copy.
How? Sometimes I receive 10,000 impressions, which cost me around $0.05, and I achieve a 0.01% click-through rate (CTR).
I'm confused about what to do next. If I create another campaign, the results will probably be the same :/
Graybeard
Well-Known Member
are you playing the lottery here? banners should be >0.15% that is that is what 1.5/1000 1.5/CPM
2000 views 3 clicks?
An internet search was a joke -- I haven't used banners in years but according to AI "should be >0.15%" seems very realistic as an estimate'
Graybeard
Well-Known Member
modify to:
- Cut any 100% loss `campaign (banners)` off when it reaches 10000 impression with no return <<< seems ok
we are talking about banners now --not ctr from landing pages - Cut any 100% loss `landing page (campaign)` off when it reaches 700 impression with no return.
that might get expensive
endriul
New Member
That's interesting. What are You using?
Okay, let me explain it this way: for example, I have 10 banners. I give each of them 10,000 impressions. Let's assume that 5 have less than 0.05% CTR, and the rest have CTRs ranging from 0.06% to 0.09%.
I keep the next 5 and give them another 10,000 impressions or more, discarding the underperformers and adding new banners similar to those with the highest click-through rates 0.15%?. I'm working on achieving this, but usually, I end up with a maximum of 0.08%. Regarding the landing page, I typically get around 300 clicks because the traffic is relatively cheap. However, I can increase the budget to 700 and see what the results will be. Do these campaigns seem profitable to you, or do you, as an expert, immediately see them as failures?
Graybeard
Well-Known Member
i read that again --LESS
.0005 or 0.05%
micro percentages
**deleted
what is the current conversion epc at the offer? --below os missing a cr rate
; (((10000 * 0.0005) / 1.5) / 0.70)/10
~ $0.4762 CPM for banners /cr at offer so a lot less
Last edited:
Similar threads
