Campaign Organization & Best Practices Help (8)
Hi Guys
First thing, i'd just like to thank you Vortex for your tutorials. I can see you put a lot of time into them and they have been a huge help to me. I have been lurking on this forum for about 5 months now and in the last 2 weeks I have just got around to start running campaigns / testing offers / and landing pages.
I have been having some moderate success with a couple campaigns. One I am averaging about -10% to +10% ROI daily that I am trying to optimize and improve. It seems like it has some potential. Another one I just started running actually had really good success and had about 42 dollars profit on 3 dollar spend for traffic (we will see if this continues though).
Anyways I had some questions about organization of campaigns and how to deal with timezone differences between the the ad source, and your affiliate network and just some general questions.
Timezone Differences Between Traffic Source / Tracker / Affiliate Network
The problem I am running into is that these campaigns have a conversion cap and there usually always being capped out everyday at some point. I am having to schedule my traffic when they reset (which is GMT+00) that is set by the affiliate network. I noticed that PropellerAds stats are in the EST timezone (GMT-5:00).
What is the best practice regarding this? Should I be setting my tracker to the timezone of the traffic source, or to the timezone of the affiliate network, or does it matter?
The other problem is, on PropellerAds, if i keep running the same campaign, my stats get bled into eachother because im starting them in the GMT+00 timezone, but propellerads is recording in the EST timezone. So for example if I need to update costs in the tracker, some of the stats are in the current day, and some of the stats are in the day before and If i have run this campaign for a few days already, i can't tell what the accurate costs are. Should I be duplicating my campaign in PropellerAds everytime I run it everyday to prevent this (even if I haven't made any changes to the campaign)?
I guess this question would only apply if you are having to work around conversion caps being maxed out.
Click tracker campaign duplication
Is there ever a reason to duplicate a campaign you have setup previously within your click tracker if you are continuing using same offer / same traffic source? What circumstances would you do this? Would you duplicate the campaign in the tracker if you are starting a new landing page split test? or is it best to always just keep running to your originally setup campaign so that all the stats combine together and don't get seperated across multiple campaigns in the click tracker?
Bot Testing
I have noticed that the bot testing tutorial has been completed yet, what is the best guide currently for doing that? I saw there is a guide from Caurmen from 2017, is this still applicable in 2019? Or maybe I am just having trouble finding recent information.
Anyways thanks so much in advance.
First of all, congratulations for taking action and on your first success! That ROI is nothing to sniff at!
Secondly: These are very valid questions that need to be asked! Thanks for asking them.
Regarding cost updating:
And gosh - it's definitely NOT necessary to start a new voluum campaign every day! That would drive you insane. Imagine if you had 50 camps running and had to do that for each one each day. You wouldn't have time for anything else!
As for when to duplicate campaigns: First of all, I would suggest to always have one voluum campaign per traffic source campaign. This will keep things less confusing. As for WHEN you'd want to duplicate a campaign - here are some examples:
-Testing bids - by duplicating the current campaigns several times and setting them to different bids.
-Start a whitelist camp based on current camp. For example if you're seeing that carrierX is converting considerably better than other carriers, you may want to duplicate the camp to only target that carrier.
You could also just change your original campaign to only target IOS. That would be fine as well. But by starting a separate whitelist campaign, you can use your original one to further test other traffic segments if you like, and/or new offers. Up to you really.
-Test any other targeting options: Such as frequency etc. It gets confusing if you use the same campaign to test different targeting, because then when you're looking at stats you may not remember what you changed and when - but then that's why it's important to keep a campaign journal!
(Let me finish replying in a bit - phone is dying and needs to be recharged.)
Amy
Sent from my SM-G930W8 using STM Forums mobile app
Hi Amy
Thank you for your reply, look forward to hearing the rest of your reply 
I had a couple other questions
The last two days ive been having some issues with conversions not tracking properly in my click tracker or on the traffic source. I am seeing more conversions in the affiliate network (clickdealer) but some are not being recorded in the click tracker (bemob).
Yesterday there was 3 unrecorded conversions in the click tracker, and today there are 4 unrecorded conversions. I am not sure where they are coming from since I am running from 3 different traffic sources. What would you do to troubleshoot this?
It seemed to be working fine previous to yesterday so I am not sure why the sudden change.
When you receive a conversion, the affiliate network postsback to your click tracker, and then your click tracker postsback to the traffic source, is that correct? Is it safe to assume the issue lies betweent he postback between my affiliate network and my click tracker?
Thank you
J
Regarding the issue above with missing conversions if anyone has this issue: It looks like it was related to conversion timestamping within the click tracker. It seems I received leads from people who visited the URL from the previous day in which it didn't record for today but for when the user visited.
Is it best to keep conversion tracking as : visit timestamp or postback timestamp?
Originally Posted by bitl4b
thank you matuloo for your response
Sorry for the delay - here's the rest of my reply.
Regarding Timezone Differences When Analyzing Stats:
When you schedule dayparting at Propeller, just schedule for traffic to run from 19:00 (7pm) EST (which is the same time as GMT +00). When checking stats in the tracker, there's no need to sync up tracker timezone with Propeller's timezone. The time range you'd want to set the stats to, will depend on what it is you're trying to optimize.
e.g. Say you're trying to cut placements, you'd probably want set the time range to the entire time you've had the campaign running, from the beginning until now.
Another example: Say you're running a 2nd round of split-testing on landers, then you'd set the time range from the beginning of the current split-test, up to now. Important: In this case you'd need to know which day and hour, and in which timezone, this round of lander split-testing started. This is where keeping a campaign journal and logging all major campaign changes will save you, and it doesn't matter which timezone you record your time in as long as you know which timezone it is. Personally I always leave my tracker timezone at EST because I live in Toronto, so every record of time/hour in my campaign journal would be in EST consistently. So let's say my campaign journal said that I started a new round of split-testing on April 6th, 2pm, and I'm wanting to check stats now to see if I could cut some landers. I would set the time range from April 6th @ 2pm, to the present time. It doesn't matter which timezone Propeller is at - won't affect stats analysis.
There's also no need to sync timeszones between tracker and affiliate network, as the conversions that are posted back to the tracker, will be associated with whatever timezone you have the tracker set to currently. If you change tracker timezone, the conversions will change accordingly - I mean the time each conversion was made would just correct itself.
So for example say my tracker timezone is EST, and I have a new campaign that has only made one conversion so far at 23:00 April 8th. If I set the stats time range to show stats on April 8th (from April 8th to April 8th), I'll see one conversion. But if I change the tracker timezone to GMT+00, then set the stats time to the same April 8th, I won't see that conversion - because 23:00 April 8th EST is the same time as 04:00 April 9th GMT+00. I would have to set stats time range to April 9th to see that conversion. And this will happen no matter which timezone the aff network is in.
When you're analyzing stats to make optimization decisions, you only mostly care that all the stats you're looking at are from the same timezone, which they ALWAYS are anyway. And most of the time you won't care which timezone the aff network or traffic network are in. Manual cost update is one exception, because you're essentially "transferring" cost data from traffic network to tracker.
The jist of timezone-syncing: No need to sync tracker timezone to traffic network or aff network (although you can if you feel it would help you somehow). But know which time range you need to specify in the tracker when checking stats - which will depend on what it is you're trying to optimize.
Regarding Offers with Caps:
Your issue with cost-updating I've already covered - if you have any questions please feel free to ask!
Even in this case though, there shouldn't be any need to sync up timezones.
The tricky thing about capped offers - and I'm assuming you're referring to NETWORK caps where there's a cap for the total of leads from ALL affiliates - is that as more and more affiliates run that offer, the cap would be reached earlier and earlier in the day.
So every now and then (e.g. every day / 2nd day / few days) you'll need to check your stats by hour for the previous day / few days, you'd need to go to your tracker and drill into hourly stats for the previous day / few days, to see which hour conversions were stopping at. Then you'd adjust your dayparting setting at the traffic source (to end traffic earlier in the day).
This is when you need to, again, figure out the timezone difference. Let's say my tracker was set to GMT+00, and when I'm drilling into hourly stats, I'm finding that the offer stopped converting around 11:00. Propeller is in EST - 11:00 GMT+00 converted into EST would be 06:00 EST. So then I would go into Propeller and target the hour range 00:00 to 05:00 (an hour before the conversions stop coming in).
Regarding Bot Testing:
Yes I still need to finish that off don't I?
Yes caurmen's method can still work, but
The best thread on how to do that is here:
https://stmforum.com/forum/showthrea...oluum-Honeypot
If you're not using
Yesterday there was 3 unrecorded conversions in the click tracker, and today there are 4 unrecorded conversions. I am not sure where they are coming from since I am running from 3 different traffic sources. What would you do to troubleshoot this?
It seemed to be working fine previous to yesterday so I am not sure why the sudden change.
When you receive a conversion, the affiliate network postsback to your click tracker, and then your click tracker postsback to the traffic source, is that correct? Is it safe to assume the issue lies betweent he postback between my affiliate network and my click tracker?
Regarding the issue above with missing conversions if anyone has this issue: It looks like it was related to conversion timestamping within the click tracker. It seems I received leads from people who visited the URL from the previous day in which it didn't record for today but for when the user visited.
Is it best to keep conversion tracking as : visit timestamp or postback timestamp?
Good troubleshooting on your part!
As for whether to go by visit timestamp or postback timestamp: I'm with matuloo - it would be simpler to go by visit timestamp. One very important thing to keep in mind: If the offer you're running tends NOT to convert right away, where you're seeing quite a bit of late-arrivals in terms of conversions - then you'll need to factor this into your stats analysis process. For example, if you're finding that quite a few conversions can still come in on the 2nd and 3rd day after the original click, then when you're looking at stats, consider excluding stats from yesterday and the day before - to make sure that the stats you ARE looking at, already includes the late-arrival conversions.
Or, if you're looking at stats from the previous 2 days, keep in mind that breakeven stats may become profitable with late-arrivals - so you don't make cutting decisions prematurely.
Alternatively, if there aren't a lot of these late-arrivals, then don't consider them at all - just analyze stats and optimize the campaign as usual, and count any late-arrivals as a "bonus".
I THINK I got all your questions! Please don't hesitate to ask more. All this sounds so complicated when we discuss it in detail, but really, everyone does things a bit differently, so don't be afraid to try anything that makes sense to you. As you gain experience, you'll no doubt refine your approach.
Amy
Thank you for your reply Amy, definitely a bigger help in understand a better process going forward. Thanks for all the time you put into this and answering everyones questions, very awesome of you!
Home > > Newbie Follow-Alongs