Simple Sweeps Follow Along (42)
The JS alert works for quite a few verticals/offers but for sweeps it´s better to use a real landing page like spinwheel, giftbox, survey.
Few questions, what targeting options do you use?
What activity levels?
What is the first screenshot?
It seems it´s creatives but there are 5 elements although you said you used only 3 creatives.
Btw, would be great if you could insert the images in the posts properly, that way the posts are easier to read
You can find a tutorial how to do so in my signature.
@twinaxe Glad to see you back here. I noticed you disappeared for awhile. I really appreciate the feedback on this and my other post as well. Some of those questions and these seem simple I'm sure, but 5 years out of practice gave me heavy ring rust. 
Sorry for the confusion in the second image, I added 2 more creatives testing different message alert images to see about CTR/CV changes but I did it late (around $40 spent) just to get an idea of how it changes things. If you want to see those I can post images of those as well. Just for future reference, I don't care one bit at all about sharing landing page data, creatives data, etc because who cares if one campaign gets stolen. I care more about how to build more of them not guard one campaign like a dragon sitting on gold. Secondarily, if other people see how I was able to pull something into the green more power to em. Alright, so onto what you were asking for...
I followed most of your advice for this campaign to get higher quality traffic. Setup is as follows on PropellerAds:
- Classic push
- PropellerAds only (no brokers <-- I've run BING ads a lot, and am quite good at it I would say and using brokers there is a GIANT NO so I figured same here)
- I left only premium zones unchecked
- User activity left at HIGH like you were saying for testing (once things seem optimized open up medium if I remember you saying this right, which this campaign isn't even close to that)
- Mobile traffic only
- And obviously giveaways interests category (although I am sure there is some cross pollination opportunity in here for a savvy person, at least PPV taught me that)
The trouble is, I used only 1 LP which was the JS redirect. Initially 3 creatives, then 2 with changed images and same title description as highest CTR/CV in the first 3 set.
In the future I will insert images into the post directly as well. For the landing pages you are referring to, I am thinking just scrape some off adplexity and change things around to fit (how unexpected, I know lol). I just want to make sure I am 100% following you here on the landing pages...Here comes some embarrassingly simple questions:
Spinwheel = Landing page that looks like the one from your example in your push guide except has a spinning wheel image that lands on the prize rather then the prize image itself. Then lists steps to enter information, blah blah and CTA button.
Giftbox = This or similar
Survey = 3 question simple yes/no, fake loading bar congratulations style. Like PPV I am assuming (used to run these with trials on walmart.com...what a crazy time that was)
If you have particular examples of these pages you want to give like you did with the Giftbox that would be cool. If not (understandably) I will scan through adplexity and find some within 10 seconds I am sure. Speaking of which, in that giftbox page you made the query strings change in page variables passed by the tracker. I coded a sweet landing page for surveys in the PPV days, where I took screenshots of all the sites I was running on...Then passed the target URL to the page and captured it with PHP, determined what it was and loaded that website homepage screenshot behind my landing page under the greyed out content box. Was pretty neat and really worked good. I don't think that applies here but it is an interesting addition on how to get more congruency between target URL and LP.
I also created an entire windows desktop application that managed my campaigns, and posted 10-100k campaigns a day to trafficvance. Yes, 10-100K campaigns. The idea was, train their approval team to mass approve everything without review because it clogged the approval pipes. Worked awesome. But that's a story for a different time.
Once again, sorry for the long winded post. This campaign produced 9 conversion for $13.30 on a spend of $60 with 1 LP (JS) and 3 creatives. Perhaps worth revamping with LPs you mentioned and swing it through again? Seems like it got baby traction on wrong stuff so maybe worth it. Not sure. At this moment I am waiting for
I always run Propeller and broker traffic together
Just start without it and in case your campaign succeeds you can still test it with interest targeting.
There is no spinwheel example in my guide
But yes, it´s the LP with the spinning wheel image (hence the name
).You can also easily edit and change the image that the same LP can be used for all different kind of prizes.
Other good ones are postal services or ISPs from the geos you want to run.
Works also good but there´s a bit more work involved.
I also used survey LPs for PPV in 2008 or 2009 already.
Pretty stable landing page style
Survey
For the new test with real landing pages create a new campaign so that you don´t mix the old and new results together.
You want to have clean stats to work with
If yes, stop it.
Cloudflare DNS for tracker domain is ok, CDN isn´t
@twinaxe
To clarify with brokered traffic, you never mentioned that. I learned that the hard way from Bing Ads (brokered traffic terrible very bad no good, unless you are doing something specific that needs it) so I figured the same here. Wrong assumption to make. Thanks for clearing that up I won't do it again. Also, same goes for interest selection. I did that on my own and figured you HAD to select one, and I never even tried submitting a campaign without clicking it. Next go around I will just post it without an interest selected as well. These were multiple things I chose to do for no reason which also hurt my campaign and traffic volumes I am sure.
Landing pages update:
Landing pages were pretty much easy to find on adplexity (ones almost identical to what you posted here) so I grabbed those. Coding is easy enough. I will get all 3 working with this offer and get them into rotation.
Follow along plan so that you know what's up:
1. I am building the landing pages for this offer tonight, and am going to probably push everything out for approval either today or in the morning. As far as translations to Polish (or any other language not English in my case) I saw you had some methods in other posts. If I remember right (sorry read a lot of different posts of yours) you had built a Python tool to translate simple landing pages like survey, giftbox and spinwheel. I could also create such a tool but if yours works and you shared it maybe I should save the efforts (I just have to find that post). Also I think you said Google does a decent job in this case because there is so little text. Even the comments area's get translated with Google translate? Also, I saw a translation JS script that will do it in your survey campaign you made too right? Just add the languages to have it auto translate. If not I am sure Google's API could be easily tapped real time but page loading speed would take a small hit I am sure. Probably just have Google do it once, and save that page as static HTML.
2. Use same creatives as a I started with last time, since my original LP was garbage.
3. New campaign in tracker AND traffic source for clean slate. DO NOT select interest type, USE BOTH PropellerAds traffic and brokered.
4. Run another $50 to $100 and see what shakes out.
5. (UPDATE HERE, Midnight My Time) Sat around building giftbox lander, survey lander and working on spinwheel still. Definitely looking into translation solution asap. Forgot how time consuming it is to build LPs when you have to translate everything on them.
Question here is:
Age old question. When do you cut zones? 3 creatives (no big image) + 3 LPs + 1 offer x4-5 payout? I don't really like particular formulas either cause 1 target can be converting great on 1 LP + creative but appear as an overall loser (was the case all the time with PPV). Just created different campaigns with only that LP + creative and broke out those targets to that campaign and paused them in the main. Probably same answer here if that happens. But otherwise on a $1.50 payout probably pause stuff that's spent $5-$8 range without producing anything? Can get expensive over lots of zones but shortcutting isn't the answer either.
A lot of my questions are beginner friendly both to help me get back into the groove as well as help anyone else following along who hasn't built anything before.
F.Y.I Initial campaign actually spent it's budget (which was $65 overall) and here is how it ended up. A couple of zones doing okay even with terrible LP setup, probably a tracker that was malfunctioning due to cloudflare issues (which I resolved), etc. So the offer maybe has life, we will find out together. Also attached creatives as well. The one in red actually had 2 conversions as well but it didn't update, once again
When I see that the campaign is doing good an dthere´s enough volume I also test with interest targeting.
For survey, giftbox or spinwheel it can be better to use real translations.
When you don´t want to pay additionaly for translations you can also see if you find the LPs in your desired language on Adplexity.
It doesn´t have to be exactly the same LP or the same porize, when you just fnd the LP style you can take the text from there and use it on your landers.
But in my opinion there´s noneed for it when you also can just translate to 50 languages and call the text then from a JS.
Nonetheless it´s great to see that you probably have some coding knowledge.
When you test a campaign it´s important to make sure to exclude non converting big spenders so that your test budget doesn´t only go to few big placements.
Apart from that I also just use some common sense.
When there´s not much volume available I have to let them running for longer.
When you however run on very high volume you can also cut them at 1 payout and still have enough traffic left.
When the offer is good then your creatives, LPs and placements will still convert somehow even when they are just mediocre.
Once you identified your best performing combination so that you´re in best case left with only 1 creative, 1 LP and 1 offer you just retest previously paused placements (that were not complete loosers) for 1 payout to see how they perform on the winner combination.
And of course it happens that one specific placement performs best on one specific creative/LP but when all others work better with something else focus on what´s doing most damage.
If the placement is big enough to be considered important you can still run it in an isolated campaign with that specific funnel.
There are two important factors about the creatives.
For your campaigns performance it´s good to use creative specific for your offer but the more specific the creative the lower the CTR.
On the other hand you need to maintain a good CTR, otherwise your creative won´t receive enough volume.
The trafficsources always send most traffic to creatives with highest CTR because that´s where they make most money.
In the end it´s about finding the best mix of CTR and CVR.
@twinaxe
Okay so as far as gathering landing pages goes, gathering LPs in the target language (in this case Polish) was easy for the survey style. Couldn't find the spinwheel OR giftbox in those languages which makes me think there isn't a lot of people running those there? Either that or just not captured by adplexity (most likely, adplexity is NOT the end all be all I know). I used trying to translate to localized words as well, obviously "wheel" search in Poland on adplexity would be wrong and not return the right result. So I searched in local languages, no luck. 2 days without tracker stats moving is killing me, cause now I have to wait for translations I suppose. I know you mentioned onehourtranslation, I have used them in the past. Fiverr also is probably a good idea. I am assuming you rip the text from the page and feed them that, since they can't change it in the code in the page (most likely scenario).
Obviously also it's going to make sense to do this in many other GEOs as I pick and run offers there (looking at Italy next at this point and more for PL). I am willing to bet I can find these LPs in IT region seeing as there are plenty of these sweeps styles there as well. But that's after these PLs go back out. So update for now is...
1. Built survey style but everyone I download is crawling with back button URL phishing redirects so I may have to build it from scratch.
2. Translated Giftbox into Polish myself, but probably a bad idea based on your last feedback (should probably pay to have it swapped from English again). However, one of the LPs I downloaded that was a survey style in Polish contains a lot of the same text/phrases.
3. Translate the spinwheel style I ripped (which looks identical to yours provided) and isn't infested with other affiliate hidden links.
For anyone downloading LPs from adplexity, make sure to search them hard. Other affiliates aren't stupid and add tons of hidden shit to their pages. I knew this before I used the tool but make sure to LOOK before you edit and run it. Literally EVERY SINGLE SURVEY page I have tried to edit contains tons of hidden links and bloated code hiding redirects in it.
Once again I understand these pages are a STARTING POINT, and I actually have other ideas for landing pages that are similar to these I would like to build and try. However, it seems like you generally always stick to these 3, at least for testing. I know from experience that the best converting and most profitable pages and angles are the ones you invent yourself overall (first one to do it kind of thing). I am not a huge fan of ripping things but it's a good starting point and you were mentioning this is pretty much what you do all the time so I thought it made sense.
I guess at this point hire the translators, build the pages (survey from or near scratch because of the problem with finding a clean one), and move forward on that. Seriously, I thought finding a base survey page to tail off of would be a little easier but those phishing links are tough.
Also, small side note about your giftbox lander (just a technical suggestion). I am sure you know most of this stuff but I am just saying it anyway.
Is there a particular reason you decided to build this page on JS and NOT Php? The reason I ask is twofold.
1. Page loads faster with php as it is PRIOR to users browser loading. Has all same token capture capabilities and in fact can load/do more to the page before it loads.
2. Php created page has no spying capabilities. What I mean is, code written in JS can be discovered via downloading the page or inspecting elements in chrome and opening the attached JS file in a separate tab. On the other hand, if the page is built prior with php doing nearly the same code, all the users gets is the FINAL OUTPUT and cannot discover the code that built it (without hacking the page or breaking the server....which I certainly would have no idea how to do 
). But yeah, snooping users can't see other GEO languages you have preloaded or images etc. The PHP does all the if, case selections blah blah and then just outputs the final stuff, but the end user cannot see any of that logic.
3. Any person or device that has JS disabled (not many I know but does exist) will still get the final LP output.
Like I said, you probably know all this so it's kind of wasted breath, and not meant to insult your intelligence. Just a suggestion. Maybe it will help others who read this too if they are wanting to hide their pages code so it can't be discovered as easily. This kind of technical stuff doesn't make the money, and so it's not really that important compared to other stuff but I thought I would at least mention this.
I will update once I get the translations done. I am in and out of the house office a lot this week so I will be up late nights grinding this campaign (and a few others) out and continue posting results.
Quick update to this follow along for anyone following:
Spent some time scrolling adplexity, looking at different variations of the giftbox, survey and spinwheel style pages. I grabbed some that made sense, removed any malicious JS, and added a little custom flair to them (confetti, and a few other things). Also, everything translated to Polish (I was gonna pay for it, don't mind at all but I found a Polish lander on adplexity that had most all the proper ad copy for these pages). I also spent a little extra time making them easy to edit for different campaigns as well so launching could be faster.
Settings were setup as:
- Both propellerads and brokered
- Classic push only (not sure when to open up in page, or what the major difference in quality is)
- high target user rating
- mobile only
- 3 creatives no big image
- 3 lps (survey, giftbox, spinwheel)
Pushed it back out into moderation and will have stats updates once it spends some more money.
Hey there, I reply in detail when I am on my PC
Another quick update:
Came home from working on IRL projects to check campaigns stats and the campaign was rejected after running $25 Apparently some of the translations I pulled from Adplexity weren't the best either? I really should just pay a translator to do these pages right. Probably hurting CTR and CV. Reason I say this is because my translations when I do the page is perfect, when PropellerAds moderations team screenshots it sometimes the word "box" on the giftbox lander becomes "dog"? Anyway, back to the point, here are some stats right off the bat but it's going to run some more budget before I really do anything...Probably pause the 2 top zones and keep going.
F.Y.I Campaign was rejected because my wording when translated was "Samsung Competition" which I understand. Instead it's now "Promotional Competition for Samsung Users" which I think should be okay. We will see. Samsung was just a device brand insert from
Side question for @twinaxe. I forgot to ask, do you rotate the LPs in one campaign or split them? Obviously for PPV etc everything was just rotated BUT will PropellerAds deny a campaign for this? I really should have asked this earlier.
Bit off-topic but you can find good translators on fiverr. Do the first gig on the platform then get them off it and you will end up having multiple translators in your skype that you can hit up when needed.
Are you focusing on Poland and Italy purely because of the cheap traffic or do you know that the offers are killing it there? Which network are you working with? I would grill my AM for some good offers and landers. A good AM won't hesitate to give them to you. He makes money when you make money, so leverage that
Hey David, working with clickdealer at the moment (used to work with 5-6 networks when I did AM before). I asked my AM for a top list of offers for push (starting in SOI/DOI, PIN and CC) and just grabbed some to start with. Really no particular reason why I chose PL or IT. Funnily enough it wasn't for cheap traffic or anything. I just thought to myself maybe getting familiar with push in smaller GEO's (mostly traffic not cost was the assessment) would allow me to work over the creatives, LPs, and get some benchmarks following twinaxe's guide on push. One of my initial questions was, "knowing which offers to pick and which geos?" Boatloads of stuff in DE and UK on those lists by the way.
I started with Geo's that had offers that seemed to be converting OK based on the stats in the spreadsheet and were not in the most hotly contested geos on the network and then thought once I got a funnel working slip in CC's or PIN submits like twinaxe was saying.
From my PPV days I mostly worked with tier 1 geos so I was just expanding this go around. I will hit up my AM though and see what he thinks instead of just a broad list of what's working overall. I appreciate the feedback. Also good strategy on the translators. Sometimes waiting for a 2-3 day turnaround for a at most 200 word lander is boring. But in the beginning I know setting up things takes a little bit of time (all the translations, servers, trackers, APIs, blah blah)
Originally Posted by quintyfresh
The 'hot list' of networks is not always 100% true. More than often it is stuffed with offers that need their caps filled.
If german is your native tone than I would focus on that and on english speaking geos. Saves you the headaches with translators and you actually know what your ads and landers are saying.
Clickdealer is a great network but also huge, so you are probably not the first pub your AM thinks about when he logs on.
Maybe contact BigBangAds (I believe they are much smaller) since their focus is mostly SOI/DOI sweeps and they have germans working there - if that's your native tongue
Sorry for going a little sideways but I hope it helps!
@david-cc No, unfortunately German is not a native language. Im a native english speaker but I oddly know pretty decent japanese (not really applicable here though.) I used to run a decent amount of gambling volume 8k-10k month (small by bigger AM standards) through Maxbounty but moved on and didn't do AM for about 5 years. I am back now so just starting with older connections I have AKA why I am running stuff from ClickDealer. Can't really show much history or volume since all my stuff from 5 years ago is gone and wouldn't apply to push anyhow. I am thinking getting in touch with BigBangAds might be a good solution (thought without history kind of sucks). I would really prefer to work with a smaller network 1 on 1 since I have budget to spend and want to send it somewhere with highest chances of success. Not just a drop in an ocean of affiliate like clickdealer kind of.
Okay so some small updates...I am going to show what I think, and any more experienced eyes feel free to correct me everywhere you see mistakes in logic. Seriously, you don't have to feel like sugar coating. If this whole thing sucks let me know and I will learn from that.
Based on these stats, not a ton of spend ($40 on this camp, $60 on JS alert redirect, $100 total to offer) but campaign slowed way down due to creatives CTR I think (fell off cliff, was 1.5%+ in last campaign now 0.5% < on this one.) Offer has had a total of 783 clicks with 13 CVs for about $20 bucks (remember a big volume of this was before LPs and attached from
But on the stats in these pictures I think, swap offer, run same LPs and creatives (although I am changing the icon to a different color message box to try and improve CTR) and use all settings from before. That's what I just did. Swapped to a different offer with different advertiser, same network, and started a new campaign so it gets moderation and new in tracker so it has totally different stats then the initial campaign. Also, since this is so small I didn't carry over any BL zones since non have really stood out as obvious losers.
I am actually thinking of building a program real quick to build full campaigns and prep the LPs (spin, giftbox, survey style) with translations, links etc and auto upload them to
What do you guys think? Good approach? Too soon? Too late? I hope this helps other people who are running similar campaigns if you are facing similar stuff.
Although it could be I would be careful with such assumptions.
As I said before, get an arsenal of the 3 LP styles and one of them will alwas work.
Delete unneeded/unwanted/unnecessary scripts and redirects, load jQuery directly from Google and so on.
I just customized it.
I also donwload my LPs from Adplexity like everyone else and I mostly use them the way they are because they are working.
No need to reinvent the wheel
In 2009 my Lead Impact account got banned because I rotated different LPs in one campaign.
That was the day when I was starting to work on my own cloaker
To swap offers and keep the remaining funnel?
To build a program to create campaigns?
Also is there a specific reason why you kept one of the creative runnig for $21 whereas the others only spent way less?
One more thing about creatives, what I like to do is to use one emoji before the title text and another one after the description text.
That way the text is kind of framed between the emojis.
@twinaxe
Thanks for the reply. I was able to clean the LPs like you said, and I have translators working on proper translations which I get back today.
As for "good approach" I just meant on this campaign. I am going to launch 5-10 more PL campaigns so I am trying to get a better feel for fast optimization. I want to get data and find winners faster as watching 1 campaign struggle to spend $20 in a day is painful.
Moving forward I was thinking...
1. Try to get better CTRs on creatives WITHOUT going too broad for higher CV (more emoji use perhaps and better creative angles). >1% seems low to me and is probably throttling traffic I would assume.
2. Probably kill creatives faster that don't perform. The answer to your question about letting one run to $20 dollars was just being unsure when to kill them and create new ones. I think the other 2 were prematurely stopped by me. The system obviously prioritizes ones with highest CTRs on its own so it quickly outspends the others like in this case.
3. Probably work more than 1 offer at a time in the campaign if possible. If iPhone sweeps, pick 2-3 from the AM recommend list in that GEO IF THEY ARE THERE. Just steps up the test budget and slightly muddles up the tracking since you have to drill down. No big deal. Also, my AM swears the 2 offers I am running are converting well in their platform and when I asked for a drilled down list these were in it. However, I just can't seem to get them to work. Many things could be the case here right? Traffic source difference, etc. Funny that he sent them 2 different occasions as offers working really good with push in PL. Anyway...
4. Make sure to get pages translated every time before running a GEO to make sure it doesn't say wacky things like "win a dog" or something strange.
5. Premake a bunch of spinner wheels for all the common promotions to speed up campaign launches.
6. Launch 5-10 more campaign on different offers in PL, IT, TH etc to start and train a little better how to spot a loser offer faster. Since the LPs are roughly the same everytime this gives us a much faster indidicator if an offer is good or bad pretty much right away (within $50 to $100 spend I would think on something that pays out say $1.25-$2 in tier 2-3 geos.)
Then, optimize. This is the part where running lots of campaign and getting an eye for things is key, like the PPV days. So knowing when to kill an offer, placement creative or LP is what I am trying to get down. Conversions isn't really the problem, I can get those. It's the magic of pulling it into the green via optimizing and knowing when to just kill an offer overall that I am onto at this point.
Also, I will post stats from this campaign later today when it reaches spend and we can see what's going on with it. I tried not to kill anything prematurely so that I could better understand when to kill things. Offers are already underperforming it seems, but we will find out later.
Okay, I hope these updates are worth it and other people are getting value out of them. If not, so what, it's fun to share anyway. Share all the way until success so others can do the same.
Campaign capped out. I will attach pictures below of all the data. Since my previous post was long I'll keep it short. My prognosis here is offer just isn't working out. Pages are normal, survey, wheel, box etc. Creatives have lower CTR but still getting decent volume to the offer. Didn't pause anything cause nothing was really RUNNING away. Traffic on this campaign is really slow. Starts pretty fast, slows down. Seems like the norm for push I feel like. (twinaxe saying scaling is the issue I already see why)
Anyway, for anyone wondering (if any exist) here is the plan. I grabbed the other top performing PL offers (gift cards, other phones, etc) and am going to roll out 4-5 more campaigns in the same GEO and run $40-$50 through each of them to get a feel if anything sticks out. Similar creatives (more emojis, and slightly changed angles), same LPs but changed to match prize, and properly translated pages which I had done on fiverr (thanks david-cc for the suggestion of keeping contacts ON skype after). I ran about $150 TOTAL to this iPhone campaign which is still pretty small but the offer seems to not work very great, at least that's my read on the whole thing.
My idea here is that since twinaxe's method of testing seems to go...
1. Use similar creatives and LPs each time (at least in sweeps) without BIG image etc to keep the frontend of the funnel as simple and predictable as possible.
2. Test a bunch of offers quickly using these similar setups until you find something that sticks out and converts (a large portion of what you launch will inevitably fail) in the AM recommended lists.
3. Ditch other non converting garbage campaigns, scale and optimize the winner offer/funnel horizontally to other traffic sources, and duplicate on same source.
4. Start new campaigns of the same type to test big image, other LP variations and original failed targets.
The key being, keep everything simple and roughly the same on the build, but power blast through offers to find something that works. Scale that. That's what I am getting already. Seems like PPV would have taught me this but ring rust is real. Anyway, stats. Hope everyone is having a blast in Dubai.
Alright some more updates here...
Did exactly what was outlined in the last post. Launched 3 more offers (and got around 10+ more approved on the network, which are going to be launched as well). Pretty much following the @twinaxe guide down the line here and it's pretty great. Simple, and easy to implement. I forgot to mention I also utilized droplets the way you did as well for the server.
Anyway long story short improvements today...
All the iphone offers I ran just didn't get traction. Literally 0 worries as I am sure this is pretty normal. Anyway, 2 campaigns of note here a phone sweep and general (voucher) sweep. Phone sweep LPs and zones creatives look like...
lps
creatives
108 was direct linked since it had a pre-lander, and there is no non pre-lander version of this offer. I was curious how good a lander > pre-lander > offer cycle would be. Seems to not effect it at all, in fact may be helping. People like double surveys I guess, they are already committed so far by this point
General sweeps LPs and zones creatives....
lps
creatives
Forward plan is launch MORE and MORE offers that are recommended to me from the AM. Other good news is a giant slew of PL offers just got relaunched at the network (excellent timing). A few questions I am pondering are...At what point do you identify the winning creative > lander > offer cycle and chop the campaign back. These have only spent barely $10-$20 which is so low to start chopping things I feel like, but am unsure. The general sweeps camp for example, has 4 CVs on just that one page. But it's very early on. Probably cut the DL on the phone and run everything again.
Also, traffic is frustratingly slow...Spending $20+ a day on campaigns is a challenge even though they are seemingly bid in the right spot (barely below suggested in most all cases BARELY) and have high daily caps to accommodate. Working on what to do about this as well cause I want faster data.
Hope people are getting value out of this follow along. Trying to be as detailed as possible. Strategy literally is just read twinaxe guide, followed and started launching campaign after campaign in the spirit of those posts. Now, some caveats are, previous experience with PPV and AM in general as well as sound technical skill (coding PHP, HTML, server management etc). Technical skill doesn't make you a good marketer though so don't be discouraged by that. As my programming teacher used to say, a monkey can program, but a skilled programmer knows how to design and write algorithms (in this case spot trends, offers to run blah blah). More updates to come as stuff spends and I launch more campaigns. 
The more poeple click on your creative the more money the traffic source makes. The clicks are multiplied with the CPC you have set. --> Result: traffic source revenue
The more revenue you generate for traffic source the more traffic they will give you.
So when your traffic is low (but could be way higher acc. to traffic chart) then either your cpc, your creatives ctr or both are too low.
You can try with a really clickbaity creative to find out how much better your creative ctr could be. Start with a low bid first and then increase to avoid unpleasant surprise.
Or just ask propeller support for what is "normal" ctr in that geo for your targeting. Sometimes they are willing to give you an idea about the range tho they usually don't give exact numbers.
Haven't run PL recently. However 0.50% creative ctr looks a bit low to me for "hi quality" (not sure if you targeted that).
I appreciate the feedback on this. I will ask propeller to see like you said. I also ran a super clickbaity one originally when I started this follow along and it got much higher CTR 1.5%+ if I remember right. But it was the "1 new message" and "click here to claim the prize" type deal. Makes me think you are right on the creatives front. Perhaps not enough bait. CV was lower as to be expected though. I am using high quality traffic settings yes, both brokered and propeller but definitely just set to high.
Glad it helped.
Updates:
Some quick updates. Phone campaign...
- Spend $30 overall (still super low and LOW creative CTR 0.5% < which I am still trying to fix)
- Conversions 7 for total of $12.25
Pages are pretty evenly split 2, 2, and 3 on CVs so can't really call anything yet there. However, had multiple zones run to $2.50+ without CVs so killed them. One thing I remember reading from twinaxe was seldom do targets magically turn around. If they have decent CTR and spent past the offer payout even with a few LPs just kill them and circle back to them later.
I think the offer has merit though cause without runaway zones it would be hovering around making nearly half back on the initial test. So more work to get it green maybe.
Second is general voucher (grocery)...
- Spend $16.80 (super slow as well)
- Conversion 7 for total of $12.50
Doing better than phone for sure, but still super duper slow. Can't make any life changing decisions on 7 conversions over 2 days...Need faster data. I bumped the bid up about an hour ago on phone and it seems to be speeding up a little. This campaign randomly got rejected tonight, on the basis of "like symbols that are similar to FB, and text looking similar to grocery chain logo". So I have to just pull it and start a new campaign I guess.
A few questions for anyone who wants to answer:
1. Bid for these camps is around 0.045 while source recommendation is around 0.05. Could be why traffic is so slow. Just trying to figure out how to dial this in to get optimum test traffic while not blowing it out (although at this point a blow out wouldn't really hurt since its been so slow). I just changed it to 0.07 and it seems to have sped it up.
2. Creatives CTRs are still so low 0.5% or less on everything I make. Maybe too specific. Start to widen the net a little cause this is a super low CTR I think. Davids recommendation was good to ask Propeller, they say 1.5%-2% on classic push it good. What do you guys think?
3. At what point to do you stop running certain LPs? Like when they each spend say $10-$15 and 1 page gets nearly all the conversions? So campaign spends $50 and giftbox page gets most of the converions like $20-25 worth. Do you just cut to this page at this point as well as the winning creative, then test big images?
4. How often do you test different variations of these 3 LPs on different campaigns versus using the same set over and over with prize changed? Reason I ask is I found some good alternatives where it has the survey that ends with a giftbox style (I know this is common now). Cause what I have is just simple giftbox, simple survey, simple spinwheel. The ones I am finding are way more flashy looking as well (better JQuery type shake events etc). In PPV simpler seemed to win more often but I was just curious what you guys think about this. Maybe run campaigns that test running just different variations of giftboxes against each other, suveys, etc to see maybe. Not sure.
I am thinking that any offer that produced back nearly 50% of its spend on the broad test cycle is worth working on further. This involves picking the winning LP, and rotating different creatives to boost CTR and whatnot. But determining how early to do this is the current state of affairs I guess. I will bump bids again, maybe try newer broad creatives and see what shakes out.
In the end you need landers that make you profits. That is why you test. The fact that you find these simple landers en masse on spy tools is an indication that they are still widely used. Because they work in many cases.
One advantage of simple landers is that they have less code and thus load way faster. Also they are usually easier to adjust to your purposes.
Once you have found a solid offer you could split test with a flashy lander.
@larsometer
Thanks for the nice suggestions.
I am determined to make this follow along a success so others who have the same struggle don't lose hope and push on. 0 give up in this deal.
Out of everything I have run my coupon sweeps did the best. Among both campaigns (1 rejected then restarted) and the other they spent a total of $44.49 and grabbed 13 CVs for a total of $22.75 (about 2 days running total). So nearly 50% back broad testing. I am trying to determine if this is enough to build better creatives, and campaigns around or if its a waste of money and should be just launching a bunch more sweeps campaigns in others geos and looking for better winners than this. Most things I have jump out, spend about $40-$50 and then generate roughly $10-$20 in conversions, then fall off the cliff of death. I liked what you mentioned about statistical significance. Just have to re-find the calculator for that. But I feel like a $1.50 sweeps campaign needs more data than $40-$50 spend to reach data. Just working on spotting loser offers fast so I can ditch them and build something else. Not sure what exact ratio to be looking for in the beginning.
@twinaxe (and any others who wish to chime in)
If you have a second I have a few quick questions for you...I am working on getting a predictable formula in place for determining offer viability, and bid strategy. I have been reading a lot after my recent testing round (about $300 in ad spend, about to go for $300-$500 more to test things around) and was thinking about some things you have said...
1. Having no bid strategy is probably a major concern with my campaigns as my bids are sort of just thrown at the wall. What I am thinking is start at the propeller recommended or slightly below, and then slowly raise to find the right zone. This is in line with what you recommend right?
2. Probably work multiple GEOs at once, as they perform differently in sweeps and other offers. Only running 1 GEO (PL in my case) may be wrong headed. I was just wanting to stick with things versus jump all around. But guess this probably isn't the right place to have this mentality. Spread out a little yeah? @larsometer you got this one right I'm sure.
3. I was looking at your test suggestions for budget being LPs x Offers x Payout x 10 as a general rule. With push there are creatives as well (you already pointed this out too btw). But when do you STOP a testing round to pick a winner offer, then test for winner LP, then test for winner creative. At least I think you mentioned it this way?
Examples campaign (a real one I ran): LIDL Coupon Poland
Total spend $17.37 (bid was low and never raised, so slow)
Conversion 7 for $12.25
5 conversions on the giftbox, 2 on others
Spread out evenly really on creatives 3, 3, and 1 (running 3 total)
I noticed in older follow alongs you said....Sometimes there is no reason to run it to the total test budget and at certain stages you can stop early and start round 2. In this case, would you have started round 2 already? I am wanting to get better at this so when I launch the next 5-10 offers I can better spot the trends and pause losers faster. Test budget for a campaign like this should be...$52.50 roughly yeah? 3 LPs, 1 Offer, 1.75 payout. I know creatives should be factored in but that makes it around $150 for a test budget which seems a little high for a 1.75 sweeps offer right? I am having lots of campaigns in the nightmare zone where its getting conversions consistently, but not enough to stay in the green and spread out evenly so nothing is a clear breakaway.
Anyway thanks for everyone's input so far.
Another small update (trying to make them regular). Spent about $30ish more of traffic.
A lot of what I had ended up in the zone where it's running 50%- but getting conversions. But it's split between pages pretty evenly so no statistical significance to be found. Also, spends are too low. Lots of campaigns like this after 30ish in spend:
At this point I am rebuilding the pages to be more easily scalable for new offers. Just inserting scripts to change prize name, image etc based on incoming campaign key. 1 spinwheel, 1 box 1 survey but dynamic. Then thinking a trying a few different looking variations. Survey style with box on the end, etc. But not right away. I think even the simple ones I have should be able to weed out a good offer.
Doing it for real now and going to launch about 10-15 campaigns in 2 GEOs. IT and PL. I will use the strategy I outlined above for management. Better bid management, stay low, slowly raise to bring in traffic on all of them. Look for a clear offer that breaks away if there is one and work on that. Start creatives both broader appeal and a few narrower to see what happens there. Running $300 through only 5 offers I think was a dumb idea. Much of this could have been stopped sooner. I also understand 5 offers is nothing. Could be 10-30 offers before you get something really good.
I am thinking each campaign should run a MAX of about $45 bucks based on payouts and LPs (3). I am also considering starting some POP campaigns too, branching that direction as well just to start feeling out the waters. However I was thinking of getting a solid offer identified through push then clone it over. We will see.
It must be super-annoying to have campaigns all hovering at -50% ROI.
Launching in more geos makes sense.
Running $300 through 5 offers - it depends. If you were testing a ton of landing pages and/or push ads, then it may be justified.
In the beginning you're always in a 'catch 22' situation where you don't have ANYTHING that is proven to perform well - so you've just got to invest budget into testing offers, landers, and ads until you have some winning ads/landers/offers, and if that's still not profitable then you may need to start second rounds of split-testing for one or more of those 3 variables.
And then of course each new geo will take you back to that 'catch 22' situation.
But yeah it's great to be conscious about test budget - in many cases it would be wise to spend less on each offer and test more offers on the same budget.
@twinaxe is still kind of tied up with family issues, but should be back to provide feedback soon! In the meantime - please keep updating.
Amy
In the test stage it´s not about being profitable right away, it´s just to check iof the tested campaigns have potential or not.
To increase the chances to get conversions it´s better to rather go for higher quality traffic even when it´s a bit more expensive instead of bidding too low and receiving lower quality traffic.
There´s no need to stick with things that don´t work.
And best way to find promising opportunities is to test, test, test
When you setup a campaign in your tracker with different landers and offers then the traffic will be distributed equally amongst all elements.
That´s often/mostly not the case with push creatives.
Some creatives receive high volume, some receive less volume so it´s hard to run a real splittest with even conditions there.
For creatives I rather go by common sense and gut feeling.
If there are creatives that convert good but receive too low volume because they have lower CTR than other creatives I test them in a separate campaign to get a better idea of their perfoemance.
In this test I would continue with the 2 creatives with 3 conversions each.
As far as I understand the giftbox has 5 conversions and the 2 other LPs have 1 conversion each, right?
Let the LPs run for a bit.
Should the giftbox continue converting and the others don´t you can stop the test and continue with just the giftbox.
Progress report:
Fired up new LP domains after the blacklist mishap. No problem, was easy to migrate all the pages. I will split the section to quicker go over everything...
Offers:
Running out of recommended PL offers on Clickdealer. Only 3 to go in the AM recommended list but there are a LOT of PL offers that aren't on that list that I never applied for in the network. I applied for all the top IT sweeps as well (which was 3) and got approved for them all. From all the supposed top offers I have ran, many just ran at 50%- or more and couldn't climb out after 50-100 in spend. No clear winner on any of them that made me think of running them more.
Landing Pages:
Rebuilt the landing pages dynamic, pictures attached of what they look like. Images also change with the prize. Comments, images, spinwheels, all change real time. I also got pages translated into IT professionally as well. Yeehaw. One thing is on the survey page I am using the questions are below the fold under the image on mobile (have to scroll to see the questions). Impacts CTR I'm sure. Thinking of changing that maybe.
Build going out:
2 new PL campaigns, which are Makita package and BIEDRONKA vouchers (actually have 2 offers for this). 3 LPs standard, 3 creatives, and 1 offer Makita and 2 for other. Plan is:
1. High quality only traffic
2. Classic Push
3. Slightly higher than usual bid
4. No interest or audience targeting
5. Mobile only
6. Propeller AND Brokered traffic (there really is no quality difference here huh? I haven't tested it enough yet to determine but I trust twinaxe 100% saying there isn't really at least in testing)
7. Overall test budget of around $45 each campaign and let it run the whole way before cutting or chopping ANYTHING besides placements that spend offer payout +- $0.25 depending on CTR rates.
Same exact plan for 3 IT offers. Build pages, campaigns etc all the same way. Overall spend another $250-$300ish dollars across these 5 campaigns. I am also using META REFRESH on directs so LPs don't send to the network but it seems to slow down the final redirect sometimes. Not sure how large this impacts my CV flow. Recommendations here would be awesome.
Creatives:
Examples are like:
[eggplant won't post in BB for some reason)
(eggplant) (2) połączenie od Anny
Zdobądź darmowy pakiet testowy Makita! ⏰
Run 3 different variants see if anything pops out. Usually my CTR won't break 1% with stuff like this. Generally struggle to get even more than 0.75% in most cases. I noticed eggplants seem to get higher CTR for some reason. That's it. Launch tonight update when all the spend goes out. Pictures of LPs which are mobile compatible:
On push and pop traffic you have very limited targeting options in terms of user interests so it´s better to stick with offers that work for broad audiences.
These would mainly be phone offers, prizes like Amazon gift cards are also good.
Other prizes are often too specific to get it working on such traffic, there it´s better to run for example FB traffic.
Instead of testing all available prizes in one geo I would rather test more phone (and/or gift card) offers in several geos.
Her ye, her ye...
Simple updates abound.
Let me make it real easy. I received a payout from CD thing month for $600 on the 15th. I placed $500 of that into PropellerAds.
Currently spent $518.16
Currently Made: $548.53
Also, ProPush...Has produced $60. So really more like $600 in total income in that period of time.
Almost $100+ in the last 10 days. Slow but steady. A few things still I am still working on...
These kinds of campaigns are becoming more common for me. It produces, BUT JUST WONT go over to the green. Either breaks even or loses a tiny bit. I think I am just gonna cut these at this moment. I also have some in there that are fighting to stay green lol. I just keep them going and keep pausing zones which are becoming fewer and fewer. Let's talk about CC submits...
I have been testing some CC submits and have yet to have anything convert. Using same style campaigns as general SOI sweeps with slight changes to the LP verbiage. Not really a major concern, I am sure I will start to see some stuff if I stay at it. I keep thinking I need to get creative with CC submits but the massive one two punch from trying unique things in the past tells me "no, don't do it." I like that only 5 CVs = $100 whereas SOI is more like 60 CVs. Twinaxe has taught me this lesson before, it's not completely about payout I know. But still, generating that revenue is easier it "feels like." Practical experience is still working on it.
Currently I have $900 in CD towards the next payout, I think I will crest 1k there like I said before. I am still working on branching to other traffic sources, and also SCALING. I have a very successful offer and campaign but am unsure how to scale it beyond where it is. To be fair though, I haven't tried inpage, interstitial, super high bids, etc. Should probably try scaling using those techniques incrementally. I have currently running a CPAg variant of it which is the first campaign in my screenshot. It's bouncing around but my good friend Larsometer warned me about that so no shock there.
I am going to do the following:
- Test more offers from the same advertiser as my current winning SOI offer
- Try to scale my winning offer funnel combo as much as possible
- Start another traffic source or 2 (maybe Mondiad and Advertizer)
- Try scaling sweeps to different GEOs. I will eventually be a kung fu master of PL if I don't scale to other nations. I have been exceptionally stubborn about this. I am thinking maybe ES, MY and (GOD HELP ME) DE. Why DE you ask? Twinaxe says get DE going and you are good to go on sweeps, others say the same. Plus in top offers from networks there is always 1 MILLION DE sweeps in there. Traffic source is unknown.
Will I die a horrible sweeps death in DE? Perhaps. Will it be a fun rollercoaster ride before the drop off? Hell yes. Life is about doing challenging things not easy ones so here we go.
Some things just don´t work, no matter how hard you try.
Then it´s better to let them go to avoid losing too much money.
Will I die a horrible sweeps death in DE? Perhaps. Will it be a fun rollercoaster ride before the drop off? Hell yes.
Just be prepared that competition is higher and traffic is more expensive so the learning costs can also be higher.
On the other hand you have a more or less never ending selction of good offers with good payouts.
If you decide to go that way make sure to know what lies ahead of you, then stick with it.
@twinaxe
I remember Larsometer telling me that you had a guide on cc submits? At least I remember him saying this (sorry Lars if I am wrong about this). Any chance you could point me to that? The way I see things is...
I am at a point where I am constantly in a break even state on SOIs. Now let me mention I am only running 2 GEOs (PL and IT) and pretty much all on initial traffic source. I keep getting told to jump to other sources WHICH I AM GOING TO DO SOON and that PropellerAds is a tough place to get consistently green at. This has been my experience pretty much verbatim. Oh I can get CVs now, but no longer is it fun spending $50 to make $50 or $45. I know I have said his before but that's where things are. Don't get me wrong, CD has $1100 in it now ready for next month. Cool right? Well not super cool considering it took $1100 to get there. Progress YES. But now I crave a little profit
So I see it like these 3 options...
1. Move more SOI sweeps but different GEOs. MY, TW, IT, FR, where traffic is still "relatively" cheap to test on. Launch tons of offers, make small margins.
2. Move into larger GEOs with plentiful offers but fierce competition. DE, UK, etc.
3. Move all operations over to CC submits using same funnels, etc in lower tiered nations for higher payouts. (I understand the math here, not always are higher payouts better)
It does seem like however that people doing really good numbers from sweeps are usually over on the CC submits side of things (sorry for being general as I know this does not encompass everyone). I am at another small crossroad of do more of what I am doing, and eventually maybe not break even, or move to CC submits or more mainsteam GEOs?
What do you think?
For sure it was not me who told you about a cc sub guide.
But once you have such a guide feel free to share it with me
Originally Posted by quintyfresh
Originally Posted by larsometer
Yes, geos like DE or US are very competitive but the chances to find good converting SOI sweeps are much higher in such geos compared to MY, TW, PL, BR, ID and so on.
You could also continue with SOI in the high value geos and at the same time test CC submits in lower tier geos.
I don't remember there being a guide on cc subs, but these threads may help:
https://stmforum.com/forum/showthrea...s-How-To-Guide
https://stmforum.com/forum/showthrea...pstakes-Offers
(Basically I just did a forum search on "cc submit" with the quotation marks just like in google.)
Also worth mentioning: You may want to start with geos for which you have a lot of data for SOI offers. Can start with the best zones etc. and test cc submits there. Picked up this tip from @erikgyepes some time back.
Amy
Originally Posted by quintyfresh
Keep it up and your campaigns will bring you even more profit!
Simple update.
Alright, modern updates on costs and CVs from June 15th to July 14th.
Spend: $1,056.73
CVs: $950.99
Propush Total: $111.14
So pretty much break even now. That makes almost 2 cycles like this. Just got payouts from CD and Propush. My total at CD was...$1048.
So I have around $1,500 to throw back into the machine and can add more to it at will. Becoming a professional money loser and trying to make it to @jaybot level at that part. I ran a few CC submits and didn't get any CVs on them. Some things I am pondering at the moment...Random side ideas read at your peril...
1. Local lead generation. I am certainly going to do this after a good and healthy conversation with @kjrocker. Glad to reconnect with you after 10+ years buddy.
2. Niche vertical reflection...
After around 5 months (holy shit, awesome) of running sweeps, and losing at first (expected) and now breaking even (sort of expected), I am curious now if I should start to dip into other verticals known to work well with Push and Pop traffic. For the record, I have ran nearly ZERO pops at all. 99% of all my training time (which I am still doing) is on Push. I feel comfortable with sweeps, but am unsure how much ROI they can truly generate? Most of my income in recent times came from Jaybot recommendations which WERE NOT sweeps offers. However, it still used sweeps style page formats.
My best sweeps campaigns so far (many, many, many launched) ran at around 15%ish percent ROI and pretty much stayed there. Tried rolling them to a few other traffic sources with dismal results. But hey! That's all part of testing and discovering new things.
Some other limiting factors. All offers with CD? Why? Well splitting up offers testing to different networks sort of sucks when you spread out all the income on tests and nothing meets threshold. I have a few networks that meet with criteria at the moment. Close to threshold, but just under. Another reason is great AM at CD and offers are approved lightning fast. Keeps my coming back.
I am going to continue running some offers I have been, but am very much considering trying to new things to keep it fresh, and also branch out a little as well. Plus I need to diversify GEO lol. That's another biggie. Anyway long ramble but main takeaways are, I am in fact alive (and loving every minute of it), and I am excited to keep the ball rolling all over the hillside. I am still working on new house projects and that is why my posting speed and campaigns speeds have slowed for a brief moment.
Originally Posted by quintyfresh;435501Becoming a professional money loser and trying to make it to [MENTION=17495
A really good CC submit case study actually, was a dude running SOI sweeps on CPA camps on Adsterra all on Pops, collected data for like a month? Then grabbed all the winning placements into a WL and ran the same funnels, but with CC submits.
Obviously, you can't do exactly the same thing, but the logic still checks out, and can do totally do the same thing with Push on any source. Even Propeller. Good read to get the juices flowing.
I'll be damned if I can remember the thread of the username. CPA..guru? Or something. Shit. Nm. Found it. @affiliatecase
Here you go: https://stmforum.com/forum/showthrea...s-on-Adsterra)
But the approach itself can work in multiple verticals.
Update time.
Alright so things are going "ok" at this point. Running some campaigns for quite some time, getting data, starting dabbling A LOT outside sweeps and have been finding traction actually quite fast comparatively. After @twinaxe talked about loans I decided to run a little of those, already seeing massive potential in places I have never ran before. Simple LPs, gather push subs, pretty good offers for push traffic.
Push subs gathering going quite well. Doing a LOT LOT of testing on gathering subs as it helps a lot in keeping things uphill. Have had MANY net positive days so far, even if small. Many in the 10+ range and many in the 20+ day range. I have yet to do 30+ I believe but have touched it. Here are some good news points..
1. Efforts from last month paid for tracker, STM, traffic, servers, etc and is still running ad budget right now as we speak.
2. I already have $800 built into CD again and over 100+ again in ProPush ALREADY. I have not added money outside of what I produced WITH IT to running campaigns and traffic in like 30+ days.
3. Most campaigns I am running are very straightforward and easy to see success or not pretty much right away. This is what I like about low payout offers like twinaxe was saying (and always what @larsometer tells people to test to start). They are right, understanding all the aspects of campaigns management comes much faster and you don't lose much plus it works exactly like larger scale campaigns...
A few questions I have that maybe someone can help me out with...
I was taking a look through many of my campaigns, even historical ones and realized what I believe to be a world shattering booboo on my part that I have been making this entire time. People will hate me for saying it, and I will definitely look like a complete fool for even bringing this up this late...Most ALL of my recent campaigns are green green green on "cellular" in the connection type dropdown, but RED RED RED on "cable/dsl". Is "cable/dsl" in
Second, how important is pausing carriers? Let this be a lesson to many others who aren't properly sorting their data. Learn from my ignorance here if I am right about this. Let me share some screenshots to show what I mean...
I killed this campaign 2 days ago...
EXCEPT NOW...Look at the connection type...
Here is a different campaign that overall is near breakeven but LOOK at its connection type...
I am sure carrier data pausing is similarly important? Is it crazy I am asking this nearly 5 FULL MONTHS into running push? Yeah probably. Lol. Anyway, just something I was curious about. I will ask Binom support if required but I suspect someone will educate me in the Jedi ways here on these mistakes. I was combing though MANY of my old data sets and LOTS follow this pattern.
Onto more building. These little advancements though are going to help get things stable I can feel it.
There is two things you can do:
- Check your top zones for traffic type. If you see same pattern on zone level that 3G performs much better than wifi you are onto something.
- Check Propeller stats for traffic type and carriers. Sometimes there can be a mismatch with what you see on tracker.
In some geos it also happened to me that 3G was way better. Then easiest thing is to run just 3G or have two separate camps. Can work fine when the geo has plenty of traffic.
With carriers it can be similar. Some just don't perform too well. Cutting them out doesn't necessarily make things better. The more narrow you target the more likely you are about to kill your camp.
The difference between between 3G and Wifi can be quite big, especially in some GEOs. When running dating, I don't care much as the difference isn't that big in most cases, but with some verticals I always separate 3G and Wifi and target those with separate campaigns. Even the bids can be VERY different, so you might find out you will be able to bid more effectively if you split them up... try it.
Home > Push Traffic >