Too Many Banners? (14)
I have a questions about the amount of banners and the amount of impressions.
Is creating too many banners at one time for a campaign a bad idea?
Also, if I were to increase my daily budget on my traffic source would that increase my impression per banner?
First you need to mention how many impressions you are getting. Secondly you need to mention the traffic source. Thirdly you need to mention the payout of the offer you are looking to promote.
I would search for statistical significance in the search bar. That will point you in the right direction.
Thanks for your reply Mr Green.
Last night I researched the statistical significant on Part 7: Statistical Significance And Picking The Winning Ads. I had cut all my banners that were lower than 1% and now my banners are around .1% to .4%.
All my banner Impressions are now about 10k - 25k, I used the spread sheet calculator on Part 10. Choosing Ads That Make Money With A Stats Calculator and my Minimum Viable Conversion was about .03% for a 120% potential ROI. I'm bad at math, but I think I understand the process so far. I cut all banners that had a maximum Conversion Rate below the Minimum Viable Conversion.
Now I have 84 active ads running at the moment. I've been testing for about 13 days worth of data now and spending at least $100-$150 per day pausing during 1AM-8AM. All my traffic is coming from Traffic Junky. My payout for my campaign is $2.8.
Is having too many banners bad for my campaign? Some of the ads past 10k impressions with no conversions passed the Stats Calculator, is it still safe to continue to test the banner? I want to test faster with more impressions, would increasing my budget to $200-$250 per day help?
Increasing your budget will definitely help you test faster, this is always the case. You need data and it costs $$, therefore more $$ = more data.
How are your ads doing so far?
Opening up the flood gates on testing is a good way to plow through banners, optimise placements, landers, etc. but if you are constantly losing money on most of the ads it may be a counter-productive approach.
Ideally you would spend as much per day as you can while still proactively optimising, improving and developing your campaign, while still keeping a level head about the overall campaign potential and not overstretching your cashflow.
There's no point spending $2000 on an unprofitable campaign when you could have determined that it was unprofitable at $500 spend.
Take a step back and look at the campaign over time - is ROI improving? Are you successfully moving toward +ve ROI? Or is it currently profitable?
Thanks for the reply Zeno.
I've gone through almost 13 days trying to optimize now, but honestly kind of blindly. I started out with 20 Ads and continued to add more each day. I removed the Statistical Significance of low CTR Ads and keeping all Ads above the .1% CTR, but I wasn't sure which ones to cut. I think I was pausing ads that had no conversions after 10k Impressions, which I have no clue if that was a good decision to make or not.
I finally put my head down yesterday night and understood Part 10 of the Step by Step Guide of Choosing ads that Make Money, but I think its late within the Ads Impressions and I lost money by leaving unpotential ads running and potential ads paused.
I'm using the Stats Calculator for Ads with Impressions of 10k, 15k, 20k, 30k (After Impressions were built up and haven't done Stats Calculation, because I just found out how to use it), to determine my ideal ROI of 120%. I was able to cut a lot of ads after.
Here's a quick screen show of my Ads and its stats from TJ Analytic + CPVLabs on a spreadsheet. The Ads are sorted by ROI and after cutting Ads, the overall Campaign ROI is at -67.96%
How do I determine if my Campaign is still good to move forward? Since I finally made correct optimization (I think), should I give it more time and see if the ROI improves?
Where did you get this ideal 120% ROI from? I think that's an unrealistic aim. Would you not be happy with an ad that you spend $100 on which makes $150?
Looking at your ad list shows quite clearly that you really needed to cut ads earlier...
Your payout is $2.80. On many ads you've spent $20 and made $2.80 - its not wise to let them run that long! It's virtually burning money.
Set a cap for cutting based on costs vs revenue. E.g. If you have spent $5 with no conversions, cut. If you have spent $10 with fewer than 3 conversions, cut.
Your top two ads show a lot more potential but note that their CTR is no better than the average. Why is that? Is there anything unique about those ads? Take what they are and modify them, build on them.
These ads are nearly profitable and only have 0.2% CTR, so use that as a cut-off threshold.
You don't need to go to 10k impressions to be confident of an ad performing at below 0.2% CTR - at 10k impressions that's 20 clicks.
Cut this in half and cut at say 5k impressions if ads have 7 or fewer clicks, provided the advert hasn't converted. Your CPM looks to be ~$1 so this should be about $5 spent on the ad. If the ad has a decent CTR consider running it for a bit more but cut it if it hits e.g. $7 without any conversions.
It's up to you to come up with some sensible rules that are statistically sound and avoid you wasting money.
How many days have you run the campaign after your big cut on ROI? If it's been running for a few days after that cut and it's still at -67%, I'd say cut it. Otherwise, let it gather a bit more data post-ROI cut.
Good work - don't worry about having started out slightly wrong, you're learning fast and that gets you closer to profit.
Thanks for the help guys.
I end up dropping the campaign after investigating all the Ads. Almost 99% of my Ads had a huge lost of from the Cost, I guess I have no clue. I'm a bit slow at this, but I'm never going to give up, I am learning so much.
I'll be starting a new campaign. I should be able to keep check on Ads early and manage the next campaign a lot better.
Thank you again for all your helps guys! 
Nice - looking forward to hearing how you get on!
And don't worry - AM's complicated enough that almost everyone feels like they have no clue at some point. I remember the first three days of trying to understand tracking - I felt like I'd dropped into an alternate universe where everything I knew about the Internet no longer applied...
Thanks Maynzie, I still have many questions before I want to post a follow along.
I did set up a new campaign but I'm confused still confused about somethings...
I've been tracking and analyzing my Traffic Junky campaigns with CPVLabs and the settings is set on CPV. Before cutting ads do I use CPVLabs "Cost tab" to determine when to cut Cost vs Revenue or do I determine this in inside Traffic Junky Analytic "Spent tab"?
After some thinking I would think using the TJ Analytic Spent tab since its CPM, is this what I should be doing?
Are there any other tools other than CPVLab that actually tracks CPM?
Compare the cost of the banner at the traffic source to its revenue.
In some cases you can upload cost data to CPV Lab.
In general, tracking systems cannot track costs at the traffic source. Most traffic sources also do not pass click cost data, i.e. with any tokens.
So, the workaround is exporting this data from the traffic source and importing it into CPV Lab, though I can't help with this as I don't use CPV Lab at all.
Thanks Zeno! Makes sense. So blind and now I see
You should be getting your spend data from your traffic source, yes. That ensures, or at least makes it more likely, that it's accurate.
Home > Questions and Answers > General Questions