Home > Mobile > Follow-along Campaigns

---

Getting started in mobile (23)

---

Comparing the 2 offers you promoted above: It's clear that the sweeps camps have a lot more potential than the ones for Castle Clash!

For the sweeps camps:

-Although statistical significance has not been reached, it's looking like the mobile camps are outperforming the wifi ones. If you want to test more cheaply, you may want to consider pausing the wifi camps and just using the mobile camps to test and optimize your creatives. I would wait for stat sig before doing this though. And then later when you have better creatives you could always try wifi again.

-I'm surprised the kangaroo angle bombed. I think it's cute personally. You're on the right track. Observation I want to share with you: Based on all the spying and my own stats, what was working for sweeps since 2011 has not changed that much over the years! Meaning, much of what was working then is STILL working today! So, I would suggest that you dig up all the best threads on how to do sweeps here on STM, and make lots of notes. Also do a lot of spying to see how other people are doing it, and take notes as well. Take your time with this because sweeps are here to stay! It's not like other mobile offers where you have to act quickly and bank before the offer is pulled. Test, test, test and you will be rewarded.

As for sources, I can vouch for go2mobi being a good network, so I wouldn't recommend jumping around at this point unless you feel you're not getting enough traffic (which for a geo like AU is not likely). For placements, I would suggest only cutting the very worst ones (such as 3x payout without conversions). I would focus on testing creatives more than anything else. It'd be difficult trying to decide which placements to cut without good creatives, plus you need the traffic to do your testing, so I definitely wouldn't cut placements rigorously at this point - else you'd be unfair to the placements and may be left with not enough traffic to test your stuff.

Good luck!

Amy

---

Amy,

Thanks for the feedback! How many landers should I be testing at the same time? I'm currently doing 3 with very different designs. It's just that with 6 conversions there is not enough statistical significance to really understand which creatives are good and which ones are bad.

I'll try to find sweepstakes threads and read them. Do you think it makes sense for me to try sweepstakes on PPV platforms? It seems they are a bit more relaxed on what kind of landers you can run and there is a lot more volume from what I understand.

---

Well there you have your answer! Wait for stat sig first. Once you have a winner, see how close the winner is to breaking even. If it's not even close, add more landers to the test that also have very different designs. Keep doing this until you have a winner that is at least close to breaking even, then start a separate campaign (on the same traffic network, or another traffic network if you need more traffic to do your testing) to test elements of that lander to optimize it further.

In this new campaign, only test one element at a time so you'd know what works and what doesn't. For example, first round of testing can consist of 2 landers, the original, and the original + entry popup alert. Wait for a winner, then split-test audio vs. no audio. Wait for a winner, then split-test 4 different headlines. Wait for a winner, then split-test 5 different images. (Etc. etc.)

Of course, you can continue to run the first campaign to test very different designs, and whenever a design looks promising, you can spin it off into another new campaign and optimize it further like you did the other one.

Now that you have a system, it's only a matter of time before you'll find something that works.

Agree with you 200%! Definitely, definitely a good idea! Another benefit with PPV is it removes one of the 2 big variables: you won't need to test banners any more!

The beauty is that once you make PPV work, you can THEN take your best lander and scale back to display traffic, and just mass test banners. As opposed to right now when you're trying to test banners and landers at the same time, with neither a good banner nor a good lander, so you're having a hard time deciding what the heck to cut! Whereas if you have a proven lander, any banner you test that convert badly will be the "sole fault" of the banner because you know your lander works! In that case it would just be a matter of tailoring your banners to the lander and cutting anything that doesn't play well with the superstar.

Keep testing my friend!


Amy

---

Amy, great advice, thanks!

I've tried another offer on Decisive and it looked promising at first, but then completely flopped. Lost another $120 or so. I'm not sure if it's saturation of those placements or something else, but it's possible that my landers and angles are not very strong at the moment. Can landers and angles really make that much of a difference? Like same offer, same placement, can lander/angle really make it or break it? Is it more of a design/layout factor or copy factor?

Also, what is the good ballpark CTR for landers? I average around 4-6% for RON traffic on Decisive, I think this is really poor performance.

I also paused the Australian sweepstakes offer, because it's just draining cash with only like 2-3 conversion per day.

I've decided to take a small break from launching new campaigns and read some of the books being recommended on this forum next week to brush up on my copywriting skills. Afterwards, I'll probably try PPV to identify good performing landers/copy and then move them over to display traffic, that seems like a workable solution at the moment. I feel like I'm just throwing money away at Decisive and that I'm doing something fundamentally wrong. I'm picking out top performing offers on networks and if other guys are able to make it work, then I should evaluate what I'm doing and make adjustments, but it's rather hard for me to understand what exactly the problem area is.

One issue I'm concerned about is that PPV will really increase my Voluum bill, but I guess I'll have to give it a try anyway. Another alternative, I suppose, would be to just go all-in and launch like 5-10 offers per day on Decisive and see if anything sticks, is that a viable approach? I'm just concerned that I may burn through thousands in adspend with nothing to show for it and I'm not exactly swimming in cash at the moment.

---

Often, right after a campaign goes live, conversion rates will appear especially promising, and then go down to a more "stable" level. I don't have enough experience to explain the reason behind this myself (other than speculations). I wouldn't worry about temporary fluctuations in performance though, because that will always happen anyways - e.g. CR will vary over different hours everyday, and for different days of the week or time of month. Just focus instead on continuously improving your creatives and your overall ROI WILL improve!


The answer to that question is a resounding HELL YES))))

Sorry for yelling, but angle and lander design ARE primarily what will make or break your campaign. Of the 3 crucial elements of every campaign - offer, creatives, and traffic - the first 2 are the most important IMO, since networks that have quality traffic abound. For the "offer" aspect - you can and should test similar offers to find one that converts well, but other than that there's not much you can do to optimize an offer. Which leaves the "creatives" aspect as the one you have the most control over in optimizing your campaign ROI.


Lander CTR is not your main concern because it's not a high CTR that will make you money - it's a high CR that WILL. I'm not saying it wouldn't be beneficial to increase CTR, but doing so would only be good IF you can increase the CR as a result. I would just concentrate on testing landers and comparing and cutting using the split-test calculator - this calculator already takes conversion rate as the prime consideration.


Brushing up on copywriting skills is always good. Using PPV to identify good landers and moving them to display traffic is a good approach, but I'd like to remind you that making PPV work is actually an end in itself and not just a means to an end (i.e. a means of making display traffic work). There's tons and tons of PPV traffic, so if you can make that work for you, that's potential 4-5 figure days right there before you even port those landers over to display traffic.

It's easy to lose focus because there seems to be so many moving parts, but IMO the most crucial part is to develop a system to optimize your creatives. If trying to improve banners and landers at the same time sounds challenging, then use PPV traffic to optimize landers first. As for offers, if you'll only stick with top-performing ones on trusted networks like you said, you can't really go wrong for something like sweeps, so I would suggest to pick 2-3 of the hottest ones and use the split-test calculator to identify the best, and just use that offer to do all your subsequent lander testing.

There's no need to reinvent the wheel when it comes to lander design (at least in the beginning) - do a lot of spying, read all the threads you can find on STM on sweeps and make notes (yes - much of what worked in 2011 STILL works today), rip and mod and test, test, test! If you keep trying to beat your best lander things can only improve. They can't not! After a while, intuition and innovative ideas will start coming to you - which inevitably results from focusing on anything intently for a while.


Regarding the Voluum bill and cashflow problems - you can definitely stay within the $99/month plan while testing landers if you keep it small-scale (i.e. not try to rush the testing by bidding high and using multiple traffic sources).

Launching 5-10 offers per day on Decisive - that would be a viable approach yes, if you're talking about different mobile apps for example. If you choose to go that way though, make sure you give each one your best shot instead of half-assing each just to make the numbers. To avoid overspending on each campaign, set a low bid and low daily budget at first, weed out underperforming placements by checking stats often, and don't spend a ton of money on an offer before calling it quits! If you've done your reasonable best in choosing what you feel are your best 2-3 angles, and you're not getting conversions, cut it and try another offer!

On the other hand, if it's a big vertical you're talking about - something like adult dating or sweeps for example - it wouldn't be wise to bail vertical after $120 in ad spend. Unlike for mobile apps, these verticals will typically take a lot of banner and lander testing before you can reach profits. However, once you do get profitable, you could be reaping the rewards for a long time to come, because there will never be a lack of adult dating or sweeps offers, and you can use the same set of optimized creatives for lots of offers in the same vertical.

Rooting for your success...
Amy

---

Amy,

Thanks for the detailed response! I think I'm going to try PPV for now and see if I can make it work. Do you have any suggestions on how to best optimize the landers?

As for sweepstakes threads, I've read iAmAttila's one, but he focuses on celebrities. I've also read that using celebrities in marketing campaigns can be risky, so I'm not sure how viable his approach is. I'm also reading the books he recommended in his blog post: http://iamattila.com/cool-shiznitz/h...rofits-500.php

Hopefully, once I read the books, I'll increase my understanding on how to design the landers that sell.

---

I've already outlined the simple approach I personally use to optimize landers in post #5 above.

Celebrities isn't the only way. Also, it's one thing to say stuff like "Even this celebrity endorses this product!" when they haven't, and another thing to ask survey / poll questions about the celebrity. And I suspect that most or all of the marketers that got into trouble from using celebrities in their marketing belong in the first group. (I mean - look at all those quiz apps that ask questions related to celebrities - nobody would be able to convince me that they got permission from every one of the hundreds of celebrities they posted pictures of and asked questions about in the app!)

Reading marketing books is always helpful and necessary, but spying on landers in the wild as well as studying landers and tips shared by the pros on STM will provide knowledge that's way more relevant. Again, when first starting out, there's no need to reinvent the wheel. Pick 5-10 landers you see that have completely different designs / angles, mod them up and test them to find out which type(s) do best, then start making different versions of the best ones to test elements (sound vs. no sound, entry pop vs. without, etc. etc.) Also look into using a CDN and optimizing your server for load speed - the importance of doing this has been emphasized in numerous threads already so I won't repeat that here. :P

As for STM threads - you need to dig around! There are lots! Here are a few to get you started:

http://stmforum.com/forum/showthread...Angle-Revealed
http://stmforum.com/forum/showthread...nd-Skype-Group
http://stmforum.com/forum/showthread...paign-Revealed
http://stmforum.com/forum/showthread...ey-Money-Tips*

And here are lots more:
http://stmforum.com/forum/forumdispl...Studies-Guides


After focusing on optimizing landers for a while, you'll start to attract innovative ideas because that's just how the energy in this universe works! Start getting your hands real dirty by testing lots of stuff in a systematic way, and you shall be rewarded!


Amy

---

Amy, thanks for all the advice! I really appreciate it. After I'm done reading the core recommended books on advertising/copywriting, I'll dive right in to test landing pages and copy/angles en-masse. So far what I've read in Tested Advertising Methods and The Adweek Copywriting Handbook has really expanded my understanding of the entire approach to marketing.

---

Amy, you are worth the $99 monthly fee for STM alone....

Your advice and contributions such as the stats calculators is absolutely golden.

When I get profitable with this, remind me to buy you a virtual coffee!

---

Hi Ocean!

Feel free to reach out to me on Skype. I'll pm you my name on here!

I'm interested to know about the campaign that showed promise and then slowed down. Was it not getting enough volume? Or you weren't sure of bid price?

---

Andrea,

Thank you for reaching out, the campaigns slowed down in terms of conversions, the traffic was still there. I have tried some different bids, but it appears the problem are the landers/angle just like Amy said.

---

Aww....thanks so much for the compliment! I'm still learning myself like everyone else. Every time I share some insight it triggers more ideas. And engaging in discussions with everyone here is also giving me lots of food for thought too. So I always get back more than I "give out".

Thanks again...


Amy

---

Np, I will be following this thread and wish Ocean success!

Amy, I wondered what spying tools you recommend for PPV? I've got a demo lined up later for WRW, but I'm torn between that and boxofads and MobileAdscout... any recommendations?

Thanks

---

I've only used 2 of these 3 tools - never used BoxOfAds so can't comment on that.

WRW AFAIK is not mainly for spying on landers (it's more for banners and intelligence data such as where traffic is run etc.) Would expert users of WRW please confirm this?

Mobile Ad Scout I use a lot. I don't even filter/sort. I just look at what people are promoting and how. And when I have a particular vertical in mind, I'd just browse through everything and look for landers I think will work well for that vertical. I'd pick and "borrow" completely different designs to test, then take the more promising ones to tweak elements on to optimize further.

Another great way to spy is by making a list of publishers (placements) on pop traffic networks, and manually browsing to them to trigger the pop/redirect (with your VPN set to the target geo of course). For example, if you go to zeropark and start setting up a "TARGET" campaign for your geo, you'll find a huge list of actual publisher urls to spy on. Best of all, this method is free.

Good luck!

Amy

---

Thanks, I'm running a three day trial of WRW now and will try Mobile Ad Scout after that. Good advice on seeing the publishers on Zeropark, will give that a try too!

Sorry for hijacking the thread, hope you have got a little further Ocean, looking forward to reading your progress

---

Have you had any further successes through your readings of copywriting books?

---

Nope, no success so far, only losing money at the moment. I've been implementing some copywriting knowledge I gained from the books I've read and I notice, through my limited testing so far, that longer, more descriptive copy seems to outperform the shorter one. My LP CTR's are still incredibly low, though, around 10-15% for POP traffic, I'm not sure why...

---

I haven't updated this follow-along in a while, but I've been doing some sweepstakes and I've been around break even for the entire time. Here are the stats since my last update:

Spent: $1870.46
Revenue: $1422.20
Loss: -448.26
ROI: -31%

As of late, I'm not even breaking even, it just can't get into consistent green no matter what I do. Tested many creatives, headlines, copy, offers (all win iphone 6 sweeps in Australia), etc. I just can't seem to get it to work and lately I've been getting discouraged. I know I shouldn't feel this way, it's all part of the process, but the feels are getting at me. I've been hitting the gym hard to try to counteract this.

The stats are mostly pop traffic, I've also tried DSP's, but they were complete failure, with exception of a couple of good placements, but now they're also in negative ROI.

At this point I just don't know what to do... I've been slacking off lately, also, because it's like I can't get anything to work and it's been somewhat demoralizing. The worst part is not even putting in work, it's not knowing what I should be doing to improve this. To be honest, I'm afraid what if I spend thousands of dollars and at the end have nothing to show for it, but tons of debt.

I suppose I could add more landers and more banners, however, at the present moment I'm just completely out of ideas on what I should be putting on landers. I've also used tools to see what other people are running and I've tried similar landers with bad results. The only lander that performed at break even was my own that I created and tested like over a dozen variations of.

I also know that people are driving volume doing the same offers that I'm doing and they are doing HUGE volume compared to me, but I have no idea where they are getting so much converting traffic.

I've been thinking maybe I should switch vertical to something easier, like app installs. It appears that sweeps are pretty hard to get to work, especially with limited experience and in a geo like Australia, where CPM's are sky high.

Any suggestions on what I should do next?

---

i would say sweeps is pretty saturated at the moment. and you're competing with people who have really dialed in campaigns.

i don't know if app installs are easier.....

but i would recommend taking a step back and spending some quality time spying to see what else is being pushed besides sweeps. there may be some new trends forming.

---

Yes, definitely switch it up a bit. Your stats are actually pretty promising - having only invested $400-ish for the amount of learning you've gained is really good. But I appreciate it might not feel that way right now!

If you've just been testing in one geo, I'd probably recommend switching geos before switching verticals. AUS is usually a pretty competitive vertical - the fact that the payouts are high and it's English-speaking attracts a LOT of people.

If you haven't tested a second or third traffic source, that's worth a try too.

Looking forward to your next update!

---

@ocean25. Negotiate a higher pay out.

Looking at your stats and knowing you have optimized everything there is you can optimize, the only thing you can do is get an increase in pay out.
If the advertiser/networks bumps your pay out with 20% - which they might do given you are doing decent volumes - then you're in the green.

If you cant get an increase in pay out, i'd drop the campaign and focus on something that makes money.
spending so much time on optimization and still losing 400$ doesnt make any sense to me.

---

I second what Caurmen said - having learned as much as you have and being out by only $400 is a feat!

If you're doing 10%+ CTR on your landers - that's already better than what I'm doing! Ultimately it's CR and not CTR that matters, but still! As has been pointed out, AU is pretty competitive. However, it's a great geo to use for testing landers because 1)it is sure to have offers that convert reasonably well in any point in time, 2)no translations are needed, and 3)there's lots of traffic available for testing.

Now that you have some good landers - talk to your AMs and put together a list of good sweeps offers in different geos. Get your best lander translated into different languages. Set up camps at min bids to test these offers to see which ones are the most promising. It doesn't take much money to test each offer - e.g. for a $2 offer if you don't get at least 1-2 conversions on a $10 spend I would just pause it. Arguably that's not enough testing to gauge an offer, but if you're testing many offers on a budget, the advantages of this approach will outweight the drawbacks. You may cut some promising offers prematurely, but as long as you uncover some winners I'd say who cares. I recently did just that - tested 40+ offers, and was able to find around 4 that were profitable. Next I'm scaling these while continuing to test all my best landers on them (plus new ones) to optimize further.

Someone correct me if I'm wrong, but there seems to be a serious lack of mobile CPI offers around lately. I moved into pop in the first place because of that actually. Several months ago I picked 3 apps, brainstormed angles and headlines and had them translated into multiple languages and made hundreds of banners - all 3 offers were pulled in the same week. I took that as a sign. Things may be better now though - definitely talk to your AMs to find out.

It's really good what you're doing now - taking a break from it all. When you're ready to take another shot you'll know because your heart will tell you, and I would not recommend to force yourself to get back into things until it does. AM isn't the easiest thing to do. I'd say you've done extremely well considering you only joined STM this year! I talk to a lot of people, and trust me when I say that you've progressed a lot further than many of the people I've spoken to. Just thought I'd let you know.

And when you're ready to take another stab at this, let us know...


Amy

---

Thanks for the feedback, I'll try to implement the advice given and will post back once I get some results.

---

Home > Mobile > Follow-along Campaigns