Home > Adult Traffic (NSFW) > Follow-along Campaigns

---

Another TJ follow along... (8)

---

Hello STMers,

I'm running a campaign on TrafficJunky and am having some trouble understanding the process. So I'm starting this follow along to get some tips from the more knowledgeable folks on the forum.

Here's what I'm working with...

TrafficSource: TrafficJunky - Desktop
Offers: 2 SOI Adult dating offers ($3.25 / $.3.50)
Geo: USA
Ads: 3
LPs: 2
Placements: 1 / NTV-A
Sites: 1 - Top 5 traffic site on TJ
Freq Cap: 1/24
Daily Budget: $100

Since I'm still trying to figure out the campaign process, I've decided to start off by testing small. Only doing 1 site / 1 placement & testing a couple creatives at a time.

The campaign is not new, it has been running for 6 days. Here are the total stats so far...

Cost: $201.45
Revenue: $130.75
ROI: -35%

Started off by testing 5 ripped banners. 2 of them had decent ROI so I made 3 variations of each, then ran them again, had a total of 8 (included original 2), only changed the image. Yesterday night I started the culling process, I followed Vortex's guide on cutting banners to decide on which banners to get rid of. Was left with 3 banners, which are currently running. I've also uploaded 3 new banners (single angle) and am waiting for approval.

For the LP's, I started off with 2 ripped ones. Did some spying on high volume sites and found two landers that were being run my lots of affiliates so I decided to give them a try. 1 LP was doing poorly so I paused it and ran a new one. So far this new one is doing a bit better.

As for the offers, both of them have been converting evenly. So I can't really cut one or the other. I'm going to continue to run them both.

--

For now I will continue running the 3 banners ( will be 6 once the new banners get approved) to gather more data, I will update tonight with the campaign stats. I will write down what I think should be done next and if I'm going in the wrong direction, please call me out on it. Also, if there's anymore details I should be sharing about the campaign that will help you guys help me better, please let me know.

Thanks for reading, sorry for the long message.

Have nice day / night.

- V

---

Hey STM,

Hope everyone's having a good one!

I'm going straight into this, here are the stats of the campaign for today...

11/2/2016 -
Spend: $31.30
Revenue: $6.50 - 2 conversions
Profit: -24.80
ROI: -79%

Total -
Spend: $233.05
Revenue: $140.50
Profit: -$92.55
ROI: -39%

As you can see, I'm in the shit with this campaign.

Those 3 banners that I left running have done poorly so far for the day. Here are the stats of those banners...

Banner stats on TJ:


Banner stats on Voluum:


Those banners have been running since Monday. They had a decent ROI up until Tuesday night so I decided to keep them running. But now they're not doing so good I'm contemplating just slaughtering them and saving myself some coin, but I'm sticking to Vortex's method on when to cut banners. I ran the numbers through the PeakConversion calculator and none of the banners have reached the 10% approx probability of sucking. So I will continue to let them run, I will recheck them in the calculator tomorrow morning.

As for the 3 other banners that I submitted this morning, they haven't done that well either. Here are their stats...

Banner stats on TJ:


Banner stats on Voluum:


Gonna let them reach 3x offer payout in spend before I run them through the calculator.

----
Both 2 offers & 2 landing pages that I'm split testing are converting evenly, I think I will continue to let them run.
----

Been reading some follow along's here in the forum and I noticed that a lot of peeps like to run a batch of 10+ banners per campaign, even when they have a small budget like myself. They run the banners, then cut the non-performers and the good ones they save for further testing. Then they test another batch of 10+ banners and find the best ones from that batch, then they rotate the performing banners from each batch on a day to day basis to decrease the possibility of banner blindness.

I think this is something I should be doing. What do you guys think, is that a good idea? I think it's def worth a try.

Quick Q's -

>Am I doing the right thing in letting the first 3 banners continue to run?
>Is there anything else I should be doing to turn this campaign in the right direction?
>Should I be testing more banners?

Thanks in advance for your help.

- V

---

Hello all,

I'll get right to it, don't want to waste anyone's time.

11/3/2016
Spend: $55.89
Revenue: $26.50
Profit: -$29.39
ROI: -52%

Total:
Spend: $301.68
Revenue: $167.00
Profit: -$134.68
ROI: -44%

The ads (3) that I had running did poorly overnight so I paused them this morning and replaced them with 8 entirely new ones. So far these are the stats of the new banners...

TJ:


Voluum


7 of the banners have converted and I've only ran them at 1.4x the payout in spend. I'm very tempted to pause banner 5341 as it hasn't converted yet. But I'll let it run until morning and see how it does. I'm following the method of spending 3x the offers payout in spend to test each banner, is that a good method of testing banners?

The campaign is still running with the same settings as when I started, 1/24 freq cap, $100/day budget, same site/placement/bid.

These are the stats on the landers I'm running with...



As you can see, conversion per landers doesn't really have a big difference. I'm not entirely sure about the CTR & CR metrics. I know I'm suppose to pay attention to these metrics but I don't know what is considered a good or bad CTR - CR. What is the average CTR / CR a lander should have?

As for the offers, offer A has almost double the conversions as offer B.



I'm following Vortex's guide on cutting offers with different payouts. The guide states that I should run both offers until one of them reaches below the 10% range. So far offer B has not reached that 10% so I should continue to run it.

---

Also, I started a new campaign on a different site on TJ. I dug into the data from this campaign and found that 4 of the banners that had decent ROI (before they all went to shit!) converted through the same funnel. So I decided to test out those 4 banners with only the 1 lander / offer that they converted with. Why not, it's worth a try! I did some manual spying on several sites and found one where dating offer competition was low and ran the campaign there. Here are the stats on that campaign so far...

Cost: $4.37
Revenue: $9.75
Profit: $5.38
ROI: +123% << Not entirely sure if this percentage is right, lol, please correct me if I'm wrong.

I'm not so sure on what to do, do I just let it run or should I upload new banners, try and increase the landers CTR / CR? Please let me know what my best course of action should be.

Shit! I'm writing too much on these posts, I'm sorry. (this is probably the reason no one's replying) I just want to give ample information about the campaign to get some help from you guys. Please let me know if I'm doing something wrong with these campaigns and steer me in the right direction.

My main goal with these campaigns is not really to make a profit (would be nice tho!), my main goal is to understand the campaign process, the optimization, the banner/lander/angle creation, the scaling and also to get familiar with TrafficJunky. Really just to trial & error my way into profit.

Thanks in advance for any help.

- V

---

Hello Victor!

Somehow your thread slipped under my radar but finally noticed it

So here we go, I'm not gonna address each of your posts separately, but rather give you some tips that should help you

1. Why don't you put the traffic cost into Voluum? The screenshots would be way easier to read I know it's not possible to reach accurate numbers, but still, it makes it easier to spot problems at once.

2. When waiting for enough traffic in order to cut some banners or pick the winners - in case all the banners suck, there is no point in waiting for more data. So if you happen to come up with a batch of banners that all seem like loosers after a solid amount of traffic, do not hesitate to kill all of them and start testing new ones.

3. I assume you are using the NTV spots, based on my experience with TJ it almost never works when your CTR is below 0.1%, the traffic is simply too expensive that way. Aim for 0.13%+, over 0.2% on some sites. In rare cases, even lower CTR can bring positive ROI, but usually not.

4. Banner and LPs - some banners work better with certain LPs and offers - as you could see with the funnel you isolated and made a campaign for. Quite often, these isolated funnels tend to die with more volume, but in some cases they can hold. Especially when there is some variety to them - so not just 1banner/1lp/1offer url. The funnels are easier to identify with more data obviously, see the next point :

5. In adult, "random" conversions happen pretty often. It's not unusual to see a lead or 2 to come all of a sudden and totally screw up the data. Never make decisions based on 1 or 2 leads. Take a look at your screenshot with those 7 banners - 2 more leads and it's all even. Winners must stand out clearly and the same goes for loosers. You want numbers like these to be able to safely cut something : 9,7,2,5,1,0,2 - Here I would cut 4 banners. This is just an example, but you get the idea. In case the differences are smaller, you need to wait for more data.

6. Banners die fast in adult, make sure to test a lot of them. Some banners will only last a day, some can last for months. Please read this to understand the process better : http://stmforum.com/forum/showthread...-down-and-down

7. LPs - for US, the LP ctr should be between 25%-40%, it's a big span I know, but those are percentages I have personally been successful with. I would also suggest to test more LPs. The ones you have now are similar in performance, so test more. Sometimes all it takes it to swap the images on the LPs or to change the text.

8. Offers - the ones you have don't have the best payouts. I have access to some that pay 4-4.50, bit bumped tho, that's true. You might need to send some volume to get such payouts, but it would also be a good idea to look for more offers.

You're not doing bad man, you're pretty close.

Now you just need to test more stuff - banners/LPs/Offers ... search for the profitable funnels, pair the right banners with the right LPs, test various offers URLs.

Keep me posted and good luck!

---

Hey Matuloo,

It's great to see you here, thanks so much for stopping by!

Tbh I haven't really thought about that. Also, I'm not entirely sure how I would go about doing that. I'll try and figure it out asap.

I'm still trying to figure out how much "enough data" looks like. Is there a baseline I should follow for when I should begin to cut banners? I'm currently following the "spend 3x offers payout" method per banner. Would you say that's a good way of getting enough data per banner? I just want to be sure that I'm gathering enough data to make statistically correct decisions and not waste money.

Yes, I'm testing on NTV-A. Thank you for this tip, I've been wondering what an average banner CTR percentage should be. Do you think I should consider banner CTR when cutting banners? Or is that not an important metric I should check?

Would you say it's best if I test entirely new landers, or should I split test my current landers and try to increase their CTR/CR? I'm not sure which method to follow first. I've thought about doing this, replicating one of my current landers and making subtle changes to the replica and sending traffic to both and see if the CTR / CR increases. Please let me know what you think my best course of action should be.

I will, thank you!

I really do appreciate your help, Matuloo. You're very generous.

---

Just turn on cost tracking for those campaigns and enter the average you pay per click. It's not gonna be 100% accurate, but better than nothing

3x payout is enough, in many times I don't even wait that long. In adult, banners die very fast at times, so it's not even viable to wait that long on smaller placements. The good thing is that good banners start to stand out pretty fast too. It takes some practice, but you can learn how to spot the worst ones pretty fast.

It is important to certain extent. When the CTR falls below certain level it's not possible to turn profit with such banner. After some time, you will know what are realistic expectations in terms of earnings per click, so if a banner gives you way more expensive clicks, you can cut it. CTR can also tell you a lot about your campaigns, I wrote a thread about it : http://stmforum.com/forum/showthread...ttention-to-it

Keep the original ones too, or at least one of them, so you can compare the new ones with it head to head in a proper split test.

You're welcome, hopeful it will help you !

---

Hey Matuloo,

Thanks so much for your replies to my questions. You're awesome!

Unfortunately, I won't be continuing this follow along any more. I paused both campaign this weekend since they started losing money and I had no clue on how to optimize them.

I honestly don't understand the optimization process. This is what I'm struggling with right now. I'm gonna stop running campaigns and dedicate some time to try and figure out how the hell to optimize a campaign the right way, because at the moment I'm just spending money and don't have single clue on what to do once I have some data flowing in.

Caurmen has written some good guides on optimization and campaign metrics that I hadn't seen before so I'm gonna study those guides, learn them, then continue to test new campaigns.

Thanks so much for your help though! You've given me some valuable tips that I will definitely take into consideration in my future campaigns.

---

In case you are not certain about some part of the process, definitely pause for a while and do some more reading. But do not read too much, real life experience is better to learn from Alert me via a PM when you start posting here again please.

---

Home > Adult Traffic (NSFW) > Follow-along Campaigns