T3 Countries - Transformers Rising Mobile CPI Follow-Along (33)
T3 Countries - Transformers Rising Mobile CPI Follow-Along
Traffic source: Facebook
Vertical: Gaming
Offer(s): Transformers Rising [iOS & Android]
Geos: TH, SG, ID, MY
Budget: $80/day per campaign ($2 per Advert Set/day)
Bid: $0.1 CPC - Average of Auto Optimized
Angles: I will be testing out a single angle, with multiple images and languages.
Campaign structure: I separated my campaign into advert sets based on geo, age (from 21-25), gender (all male), OS, interests (video games, online games, MMO games, MMORP games), and language (dependent on geo).
Ad Headline: Will you be an Autobot or a Decepticon? Show us what you've got! Start playing free now!
Ad Text: Transformers Rising on {{Android or iOS}}
Banners: Banner Ads Album
Landing pages: Landing Page Album
Landing pages load time: 460ms-650ms
Updates: Daily, with complete transparency. I will answer any questions posed about the campaigns and I am extremely thankful for criticism and suggestions.
Spend for the day: $2.76
Revenue: $0.80
Net profit/loss: -$1.96
Actions I took: Launched Campaign. Implemented tracking through
Current tasks: Track performance of Landing Pages.
Blacklist placements with spends of over $1.25 and no conversions. Blacklist placements with over $0.25 spend and no clicks.
Track performance of banners, remove banners with significantly poorer performance.
Is there typically a meaningful difference between only allowing phones on wifi to see an advert and otherwise in a CPI campaign?
From my initial test data today with Malaysia - Android, I'm noticing a large difference between clicks on FB and clicks in Mundo. 62 people clicked my ad, but only 18 made it to the offer page (29% CTR). I optimized my lander images, but I'm concerned that this indicates something is significantly off with my landing pages. I'm going to test tomorrow with all 4 countries on Android and iOS with the landing pages ON for 24 hours, and then switch over to Direct Linking for 24 hours and see if there's a significant difference.
The average CPC across the campaign is $0.1, and conversion rate is 1.6%. Need at least 10% conversion rate to be profitable.
fb can charge you for likes and shares that don't make it to your lander or offer
I made a small error in my
Thanks dynamicsoul,
Is there a way to prevent that action? Definitely noticed a tremendous number of likes and shares when I checked my stats this morning.
Update:
Spend: $46.51
Revenue: $1.84
ROI: -96%
FB Clicks: 550
Tracking Link Clicks: 457
CTR (to landing page): 67% (I appear to be losing a lot of visitors here, which I originally though was load time of the offer, but this is largely because of the errors I made earlier today in providing the LP url.)
LP Visits: 307
CTR (to offer page): 7% (This looks absolutely abysmal to me. Is this normal? Is this a sign that direct linking may be more effective?
Offer Visits: 22
Offer Leads: 2
Offer CVR: 9.09%
Overall CTR: 4.8% (based on track link clicks over offer visits)
I need more data for that offer conversion rate to hold real weight, but that is an inspiring number. It'd mean a revenue closer to $42 if I weren't dropping so many visitors along the way.
Additionally, I'm getting an invalid end date error for 12 of my advert sets (all iOS) when I try to upload them. "Failed to create advert set {{name}}: Invalid end date for lifetime impressions".
Firstly whats the file size of your landers and where have you hosted them? test them with : https://www.uptrends.com/tools/website-speed-test and pingdom etc to try and get an idea of how fast they may be loading in each Geo.
Ideally you want it to load in less than half or 2/3 of a second.. for a small mobile game lander. Some of these users in these geos will be on a crappy connection.. so need to be fast or they will just click back before it loads.
secondly, there's no real way around stoping people liking or sharing your link on FB.. apart from making it even more enticing to click through rather than like.
I fear it may take some dollars to get a campaign such as this profitable on FB in these geos.. You'll also have to test languages etc
Hey dynamic,
I have that information in my initial post. All the image files are under 200kb and load in about 450ms to 650ms, tested via pingdom.
Currently I'm testing Malay, Thai, Indonesian, and English on the headline and ad text. I've tried to make the creatives relatively language-agnostic to reduce the amount of translation work that needs to be done.
Ah ok, I didn't notice that bit
Well it's something else then.
Update:
Spend: 56.16
Revenue: 3.28
ROI: -95%
Action taken:
I'm switching all campaigns over to direct linking for the next 30 hours to see if that impacts the performance of the campaign positively. Once they get to the offer page, the conversion rate is decent (6%), but I'm losing 90% of my traffic on the landing pages.
Update:
Scrapped and restructured the campaigns based on performance over the past two days. CTR in ID and SG was absolutely abysmal and I noticed a lot of traffic was dropping off (and understood why a little better after reading Zeno's excellent coverage in the tutorial section).
I took action by creating 4 new campaigns, Two for TH and Two for MY, representing Android and iOS versions.
I doubled the budget so we are now at $100 a day spend, but I have great news.
I added Transformers to the interests of each demographic (Duh, I know, should've done this earlier), and after consulting the ~800 visitors I'd received and seeing that almost all that converted were on wifi, and speaking with a few other guys on here, decided to restrict traffic to ONLY Wifi (the transformers app is massive at 100mb, and can not be downloaded on iOS devices without being connected to wifi, I didn't know this so a large amount of my iOS spend was a wash).
The results of this were apparent within an hour of uploading (uploaded a little before midnight EST).
Spend: $5.08
Revenue: $4.00
ROI: -22%
Still operating at a loss, but so far revenue has beaten out what since the inception of the campaign. I'm going to take this as a win and see what happens in the next 8 hours. I'll cull off any ads that aren't receiving clicks after $0.25 spend and aren't converting after $1.25 spend.
If anybody has input, please reach out to me at dmchale92 on skype. I'm pretty new to this and it is blowing me away how much I've learned in the last week, especially since I started running campaigns in earnest.
This campaign would actually be profitable right now, were it not for the iOS traffic. It isn't converting at all (direct-linked to the offer) and I can't figure out why. I'll see if that changes in the next 8 hours, but it could mean cutting those campaigns if it does not. Half the current spend is from iOS traffic.
Do you have 2 offers with rules in place on
Hey Jeremy,
That's an excellent observation, and after reviewing the data and making that post, I actually cut iOS from my offers and am now focusing exclusively on Android in TH.
Spend: $41.55
Revenue: $24.36
ROI: -41%
I'm not sure what action to take from here. TomCPALead suggested making a dark post of the most successful ads and pursuing CPM bids (starting at the lower range of possible bid and incrementing as time goes on until I get decent traffic) in order to reduce the CPC and make the campaign more profitable. I did so early this morning, but I'm not getting much reach with it. Another step I've taken is to convert the page I was using to an Entertainment Website category from a Health Brand category after suggestions that this may increase the reach and relevance of the audience.
I saw that you ran a lander at the beginning and switched to direct linking, but I didnt read anywhere, that you split-tested them. Why dont you try split-testing direct vs landers? Also, setup your landers on a cdn with servers in Asia
I had a 24 hour period where I left it to run with traffic split between three landing pages per OS, then a 24 hour period where I changed to direct linking without altering ad spend and saw a massive uptick in clicks and in conversion rates.
I tested making dark posts and optimizing for post engagement which wound up being about 6x as costly as CPC (I don't see an option for oCPM in the power editor, I think I must be missing something).
I was using GetResponse to host my landers, I've heard Unbounce is pretty good and they use AWS which has servers in Asia. Would that be a faster solution?
Edit: Just did a speed test (originally I used pingdom for my speed estimates) with https://www.dotcom-tools.com/website-speed-test.aspx from 22 different countries.
The tests for China and India were between 4,000 and 45,000ms. Absolutely insane as GetResponse does not have CDNs in Asia. Lazy solution, definitely not a good one for anyone following along (unless you only want to use it for North America and Western Europe).
Here's a picture of today's ad spend to highlight the differences between a set of dark posts with my most effective ads using post engagement and the same with CPC optimization.
Facebook gave me some interesting data and showed me this is a campaign with huge potential. Looking for suggestions of other traffic sources. Going to get a LP hosted closer to Asia and then localize my best performing creatives with some Thai Copy and investigate LeadBolt, TapIt, Airpush, and MMedia. If anybody has suggestions for hosted LP providers that'll give me good performance in Asia, that'd be kickass.
Stripped out the English Ads from the Dark Posts and manually bid on impressions ($.42 CPM) and it improved their performance 6x, they're averaging $0.08CPC versus an average of $0.06CPC with my original ads.
I transferred all of my landing pages over to Unbounce (which tests much faster than GetResponse, and gives me a bunch of high-converting templates tested by the community, alongside a really interesting front-end design interface.) with a coupon for the first month free and 50% off my first three months (tmonkey503 for any interested). I also updated them
I'll launch a proper split test between my direct link and LP traffic later today, along with developing mobile banners from my creatives and testing the waters of Go2Mobi (recommended by jeremykng), AirPush, MMedia, and TapIt!
Paint.net is going to be my weapon of choice here, after getting the copy translated by OHT. If anybody knows of a better way to do this, I'm all ears.
FB(today):
Spend: $31.5 (this would be much better if I dropped my dark posts ($22), but I really want to get more data and see if I can get the cost of that traffic down. If anybody can explain oCPM to me over skype, I'd appreciate it!)
Revenue: $8
FB(total):
Spend: $96
Revenue: $38.2
ROI: -61%
look promising
It's a little early to tell, but switching to a new CDN and putting up better landing pages looks like it may swing this in the right direction. Once I get my CPC under $0.05, this will be a very profitable campaign.
which campaign objective / bid mode are you using?
In FB I'm using Click To Website with CPC bidding and it's giving me about $0.06 average, for dark posts using impression bids and bidding $.7 which is giving me $0.10CPC on average.
In Decisive, I'm using Smart Bidding and so far the CTR is decent, but no conversions. Following appetiser guidelines for that campaign. Here are the banners I'm using.
Using dark posts with impression bids on FB I've managed to get my CPC under $0.01 by split-testing 7 different images with different Thai headlines. I'm going to keep optimizing this and see where it takes me, I'm really getting a feel for what works on FB ad-wise, exciting stuff. I think we'll see a positive ROI by today or tomorrow, very exciting stuff. Big thanks to everyone that's given advice and criticism!
Hi there! I was reading your thread and now i am wondering what 'dark posts' are? Never heard of the term before :-)
Good Luck with your campaigns!
It's where you make an unpublished post through the manage pages tool in power editor and then use it as an advertisement.
some of my FB ads got under 0.01 CPC for CTW (for Indonesian user)
how? "WOW" Post Headline + "eye-catching" image + call to action post text. and also match the target interest for sure..
Going to run this for one more day, but it looks like I may have to cut the offer if performance doesn't pick up.
From FB, I've got $175 spend and $60 in revenue between the dark posts and normal CPC campaign.
From Decisive, I've got $25 spend and $0.7 in revenue, from a single conversion between four camps (site, with wifi and without, and apps, with wifi and without).
u're making sure you're looking at "website clicks" CPC, not the generic CPC, right?
you should ideally run DL/LP test at the same time.. not in sequence
Aaaand Offer got paused. I'm going to call this one dead for now guys. Stay tuned for Mobile Pin Submits or "How I learned to stop worrying and love antivirus offers".
Could you run some tests on this for a younger demographic? If possible (given it's a game install it may be), that's definitely worth doing - the demo for "Transformers" skews very young.
See http://www.hollywoodreporter.com/new...younger-208116 .
Hey Caurmen, that's a good call. The Campaign is stopped now, but I was getting decent conversions and much better CTR and lower CPM for 13-19 (did demos with 10-19, 20-29, 30-39, 40-49).
You continuously mentioned Dark Post, What does it mean?
aimisaudi Dark Posts is synonymous for Facebook Unpublished/Sponsored posts.
Originally Posted by Mr App
Originally Posted by aimisaudi
Home > Facebook & Instagram > Follow-along Campaigns