Willing to lose a small fortune in mobile to make a large one! (38)
Willing to lose a small fortune in mobile to make a large one!
Background
I started out on POF a few months ago and created a follow along for that. I continued running campaigns after I stopped updating the follow along and managed to improve my results but reached the point where I had learned a lot but felt it might be worth trying an area that has greater growth potential and a lot more up to date support on the forum. I have paused my POF campaigns and am going to focus 100% on mobile and will stick with it until i am stacking that money!
Setup
VPS with Wiredtree
STM Mobile Tracker
Decisive Traffic source
Antivirus offer targeting Brazil
Linking to Lander
Status
After having two rounds of rejections from Decisive, I finally had my ads approved on Thursday and ran two full days of traffic. I was blown away by how fast the volume and clicks came in. However, I have paused my campaigns for two reasons:
The tracking doesn’t appear to have been set-up fully as conversions are not being tracked in decisive / STM tracker but they are showing in F5.
There is a huge click discrepancy between decisive / STM tracker and F5. Decisive is showing 1760 clicks, STM is showing 1789 clicks and F5 is showing 203 clicks!
I really need to get these areas solved before I can restart my campaigns and actually work on optimising the banners etc. I have read through a few threads and tried to solve the solution from that but am a little stumped. I would really appreciate it if anyone can’t lend me their experience in this area so that I can get things back on track again! 
Thanks in advance!
Hm, that click discrepancy issue seems to be happening a lot. Talk to your AM at F5 and see what they say about it.
Tracking: can you post your offer and postback links here (change the domain and affiliate id, of course)? We can then debug them.
I have decided to set-up
I sent a message to my F5 AM and am waiting on a response. Decisive mentioned using their click explorer feature to see a breakdown of where the clicks came from but didn't actually say anything about bots etc!
I made a major newbie mistake, the clicks in decisive and prosper were related to clicks from the ad to the lander. The clicks in F5 are the clicks from lander to offer. Doh!
Thanks to F5 for pointing this out!
Whoops, should have spotted that you were running a lander. Glad you got it sorted out.
Update
I have had a couple of days traffic now and it really comes in fast! I had four campaigns running as per the recommendations, wifi / carrier, app / site and have had to pause three of them.
Here are the stats:
106k imps 1056 clicks 1% ctr 0 conversions 25c cpm 26.50 spend
59k imps 466 clicks 0.78% ctr 2 conversions 31c cpm 18.43 spend
76k imps 1114 clicks 1.5% ctr 2 conversions 32c cpm 24.62 spend
I still have one live currently at:
26k imps 300 clicks 0 conversions 38c cpm 9.94 spend
I am split testing lander against direct linking and the direct linking got 3 of the 4 conversions. Overall pretty bad ROI on this so far.
So I guess I now need to go and build a load more banners on another angle for the same offer and see how that works?! Any input or suggestions at this stage would be great. e.g. is it worth testing this campaign on another traffic source as well?
I've seen some click discrepancies as well.
I was wondering if it was a time zone thing. eg some of my conversions show as 'today' on decisive and 'yesterday' on f5
it wasn't a time zone thing for me, I just got my sums wrong!
1000s of clicks and only a few conversions isn't great - is there much fat you can trim yet e.g. poor performing placements?
Also, what offer is it and how many people from Brazil have already installed it?
The Brazil AV market is getting smashed right now and you may have trouble getting traction without a good angle and when new to the game.
There isn't much fat that can be trimmed in that area. It is P Safe. Would you advise ditching it and trying a different offer / GEO? What level of conversions in 1000 clicks should I be looking for at the start of a campaign?
Psafe has gotten quite saturated. I ran it when it first came out and then it was on fire. Split test some offers using the top 5 angles you can come up with for each and let the data do the talking 
Thanks for that input, I will mix in some new offers. Would you recommend trying another GEO?
been recommended armor for android in nz / ng. Is that one still worth running?
i was also considering cycling in another traffic source for comparison, would you say that is too soon?
Yes to all of the above except Armor For Android, which I haven't tested. However, definitely try other geos, try other offers, and try another (mobile display - keep in the same area) traffic source or two if you have the budget for it.
I've loaded another campaign up in decisive for the armor campaign in nigeria so will see how that does. Will then look to do another network next week perhaps. i ended up spending 75 bucks on the last campaign which was obviously a large waste of dough for such a small payout offer. When should I have cut that one?
Originally Posted by cooperoil
So if it was a $0.50 payout offer I would probably spend ~$10-20 per angle with multiple banners running in parallel.
OK, thanks. And you recommending running two angles on an offer before trying a different one? I could run out of offers pretty quick! It could be a good offer but a shitty angle as well!
Try multiple angles - as many good ones as you can think of!
But set a limit to your spend - i.e. if you spend $80 on 4 angles total and none give anything above say -75% ROI, it's probably a dud combo of traffic source/offer.
Angles are critical but a good offer will usually convert poorly pre-sold traffic reasonably well - i.e. not profitably, but enough to demonstrate potential.
Thanks Zeno, very helpful.
My new Nigeria campaign is giving me very low CPMs 5 - 16c, volume is relatively low but conversions are coming in slowly on a couple of campaigns which are around -30 - -50 ROI. I am using smart CPM at 1.5x payout, should I raise this a bit to get a higher CPM and more volume or leave it for a bit?
A bit of an update.
I have been running the Nigeria Antivirus campaign since Friday and have some potentially promising results I think. My angle is not original at all, just focussing on battery drain but I am going to work up some new angles later. I have four campaigns as suggested and below are the results. Only one campaign has a decent CTR and the others are 1/3 - 1/2 of the leader. I think based on what Zeno said above, the offer I am working with is the strong element as it is converting where other elements such as CTR, angle etc are not as good as they could be. Also, I am direct linking.
Total Stats since Friday
Wifi Site
Imps 64k clicks 389 ctr 0.61% conversions 2 CVR 0.51% CPM 0.08c cpc 0.01c cpa 2.69 spend 5.38 revenue 76c
Wifi App
Imps 25k clicks 269 ctr 1.1% conversions 5 CVR 1.9% CPM 0.19c cpc 0.02c cpa 95c spend 4.74 revenue 1.90c
3G App
Imps 25k clicks 112 ctr 0.45% conversions 2 CVR 0.45% CPM 0.08c cpc 0.02c cpa 1.01c spend 2.01 revenue 76c
3G Site
Imps 31.5k clicks 105 ctr 0.33% conversions 0 CVR 0.0% CPM 0.07c cpc 0.02c cpa - spend 2.06 revenue 0c
These are total stats since starting, although I got no conversions on Friday and yesterday a couple of campaigns got close to break even:
Wifi App
Imps 11.4k clicks 141 ctr 1.2% conversions 4 CVR 2.8% CPM 0.17c cpc 0.01c cpa 49c spend 1.97 revenue 1.52c
3G App
Imps 13.4k clicks 68 ctr 0.51% conversions 2 CVR 2.9% CPM 0.07c cpc 0.01c cpa 48c spend 95c revenue 76c
Main issue is really the volume that I am receiving. as previously mentioned, I am using smart CPM and it is giving me very low CPMs, I am wondering if I should up this a bit to see if it results in more volume. What do you think? Also, these campaigns are approaching a collecting $15 spend, where would you go next?!
So close! Test more banners. Throw a lander in the mix.
A boost in CTR and CVR could bring you into profit land, and then you have placement/carrier/device optimisation to help as well!
I'd let each campaign get to at least $7 or so spend before making any decisions.
The usual rule I quote is $15 per campaign, but in light of the low payout you can probably reduce that a bit.
As Zeno says, test moar! It's looking promising.
Thanks for the feedback guys! I am going to build a load of new banners as CTR improvements will really help take me profitable. When should I cut the banners in the campaign? X spend and no conversions? x spend and less than x ROI?
Those are the kinds of optimisation decisions that you need to decide on and write down!
Example:
If ad spend = 2x payout and no conversions >> kill
If ad has > 5 clicks and CTR < 0.5% >> kill
If ad has > 3 conversions and profit <$0 >> kill
How you do this is totally up to you and will vary with other factors - e.g. offer payout (might wait longer or less for data), historic performance and how well you know the demo, etc.
Have a read through the Mobile Appetiser or the Mobile Main Course to get some suggestions on cutting rules.
Thanks guys. Good stuff there. Well, I cut poor performing banners and posted some new ones with a new angle that went live last night and so far today my CTR has improved (doubled) and conversions have completely stopped!!! I will keep this running for the rest of the day, as I am getting data. It is an antivirus offer and the last angle was around battery life which is a bit tired, so with this one I focussed on government spying on your messages etc. I imagine that it maybe doesn't fit with the lander / offer which is why it isn't converting so well. Any suggestions?
Are you running a lander or direct linking? And what's your spend so far today?
I wouldn't make too many decisions based on one day's data - could just be a bad day for some reason.
I'm direct linking. Spend today is not that high between 50c - 1.4c on four campaigns. the highest spend does have a conversion now. I won't make cuts too soon but it is a low payout offer so don't want to burn too much. Also, volume is coming in low, should I raise my CPM a bit to get some data faster?
Originally Posted by cooperoil
Your spying angle may benefit from a lander... I would whip one up and split some traffic to the lander, some DL to the offer.
Yep, definitely - that's a bit too low to make any decisions. Raise in small increments until you're getting enough volume to make decisions on.
ok, going to up bids now and then try and get a lander going in next couple of days
Your so close! Keep at it
I upped my smart cpm CPA goal to twice the initial CPA and it has had no effect on volume. Is there another thing I can try or should I maybe ditch this geo as the volume just isn't there on decisive. Or should I try running it on another network? I have tried two sets of banners here and have attached total stats below. Should I try another angle, traffic source, geo, banners? Or another offer on another geo?
Thanks!
If upping daily budget + bid gives you little more, it's likely there is not much traffic and you are being competed with well.
If the campaign is showing potential, I would consider testing a traffic source that has a hefty amount of Nigerian mobile traffic.
ok, next step, anyone know any good sources of Nigerian mobile traffic?!
That would be a good question to ask as a separate post - you'll probably get more responses that way, not least from traffic sources with Nigerian traffic! Stick it on the general Affiliate Marketing forum.
Will try that, thanks
Home > Mobile > Follow-along Campaigns