Follow Along Starting out in Mobile 30 day follow along

Follow Along Starting out in Mobile 30 day follow along

Hi all,

After a few weeks of analysis paralysis I decided a couple of days ago to "get into mobile". Many thanks to Liane, TheGrumpyRussian and several others on here for their posts and help / inspiration they give to others.

I have CPVLab and Prosper202 configured on my WiredTree VPS and accounts with NeverBlue, OfferMobi, YeahMobi and a few other CPA networks. I have also funded accounts with JumpTap, BuzzCity and Admob. Budget – I am serious about this so will spend what I need in order to make it work so I am willing to spend $20 – 30k or so before getting paid by the CPA networks at the end of the month.

The goal at the end of the first month is to be able to quickly select and create a mobile campaign and be able to optimise it to profitability within a couple of days (or drop it and move on) and have a system in place so I can monitor and further optimise all of my ongoing campaigns. Monetarily I want to be making a better than 50% ROI on my traffic spend and I want to be spending $1k + per day.

Day one (Thursday) was spent researching CPA offers by talking to my AM at NeverBlue and OfferMobi and various other bits of research mostly detailed in other posts on this forum and cutting some graphics for the ads.

Day two (Friday) was spent creating campaigns for these offers. I have used a combination of the ad images and copy supplied by the CPA network and some of my own. I’ve not made any of my own LPs as yet and am direct linking everything.

Fortunately / unfortunately I had a good weekend with my kids and family and was out and about so didn’t spend any time in front of a computer – DOH.

Fast forward to today (day five – Monday) and I’ve spent $377.52 on traffic and earned $44.95 – not a great return, but
Lesson 1 : Don’t ignore running campaigns at the weekend
Lesson 2 : Check tracking is working properly before knocking off on Friday night!

OK all is not lost, but my stats are as follows:

NB – Neverblue
JT – Jumptap
AM – AdMob
OM – OfferMobi
BUZ – BuzzCity

I was expecting some click discrepancy, but this seems mad – take offer1 – Jumptap say they’ve sent 749 clicks, CPVLabs thinks it’s had 814 and Neverblue say they have actually received 286! Who do I believe?

I seem to have completely cocked up my tracking in CPV – nothing of use is shown in my campaigns so I can’t optimise at all. I’ve looked at a couple of posts on here and re-set those up in CPV – not very impressed with it (the product) so far.

Also, I was hoping to be able to drop non performing sites from Jumptap, but my advanced reporting in NeverBlue just shows "Default Traffic Source" under Site… I will call my AM about this later.

I don’t know if I have enough data to be able to say this for sure yet, but my feeling is that:
Lesson 3 : International offers convert better than US ones.

Priorities for today, sort out my tracking!

That’s it for now, sorry my first post was so long – I will keep them much shorter in the future. Any advice or criticism is welcome.


User Comment:
Great post Rob and I for sure will be following this closely!

Couple of things:

1) I would stick with 1 traffic source for now and get that to work first, before moving onto others. Park AdMob (expensive and crowded) and BuzzCity, and start with JumpTap

2) Stick with DL first then when you have something profitable you can explore whether LP improve your metrics

3) Click discrepancy is a fact of life but these are ridiculous. Bets way is to remove "steps" in the "chain" (ie less traffic networks) and check that all your redirects are working. Also – cause this is mobile – things will be slower for your user. A 2-3s delay in redirect will be much worse on mobile and people are more likely to move onto something else.

4) As it CPC, make sure your ads are clear on what the user is gonna get. Kind of prequalify them so that they wont click if they are never gonna convert.

Looking forward to the next few days!

User Comment:
Got my popcorn out.

Like tijin said concentrate on one thing.

1. US traffic DOES convert find pockets of cheap unfilled inventory
2. Jumptap is create because they pass variables to help you see whats working
3. Apps and mobile web have drastically different results test seperately

/angry out

User Comment:
Looking forward to seeing your progress!

User Comment:
Hey Rob, I’ve noticed the same thing with cpvlab, its numbers seems to be off, or I’m loosing alot of clicks between cpvlab and the networks (offermobi and neverblue). Stats are attached.

Anyone else notice this? What are you guys using to track your mobile campaigns? I’d like to find something until I finish building one.

User Comment:

Originally Posted by leber026

Hey Rob, I’ve noticed the same thing with cpvlab, its numbers seems to be off, or I’m loosing alot of clicks between cpvlab and the networks (offermobi and neverblue). Stats are attached.

Anyone else notice this? What are you guys using to track your mobile campaigns? I’d like to find something until I finish building one.

track with has-offers…..

you will have much better luck……

User Comment:
great follow along……all this mobile is making me want to dive in…….

User Comment:
Thanks all for your replies.

Good advice there tijn and Russian, I will focus on getting Jumptap working and bin the rest for now.

Russian, where you say "find pockets of cheap unfilled inventory" do you mean use all the tracking variables from JT to weed out the non performing sites? Then presumably also weed out the non performing handsets / carriers?

I think I will just focus on Neverblue for now as well since they have the advanced tracking stats.

Polarbacon, I’ve been seeing hasoffers crop up more and more on here so have signed up for a one month trial with them.

No new stats to report as yet since I amended my target URLs to hopefully pass the right JT_ variables through to CPVLabs, my campaigns have been pending approval all day – BAH.

I will create some new campains in NB now and point these at my hasoffers tracking – could be a useful fact finding mission for us all there!



User Comment:
Some quick thoughts:

1) If CPV Lab is showing more clicks than Jumptap…it may be that

a) Jumptap caught some dupes / fraudulent clicks and took them off your reporting

b) Jumptap hasn’t updated all your clicks yet (check again in a few days – do the #s change?)

c) Someone refreshed post-click before the redirect happened.

2) Also, wanted to make sure you saw THIS THREAD where I posted how to set up the JT variables in CPV Lab.

3) Regarding the really low click count on NB…I’m not sure how they track, but check with Marty and see if they’re seeing your clicks:

a) from wrong geographic locations? That happens sometimes with mobile, really poor targeting from the ad network OR fraud they haven’t caught yet.

b) duplicates? Back when Offermobi ran Linktrust, user reporting used to not show dupe IPs so you wouldn’t see many clicks in your reporting…even though they were there. It made the conversion rates look higher sometimes than they really were, and made it difficult to optimize and get rid of crappy targets.

c) from banned referrers? Make sure you’re targeting properly (carrier / handset restrictions), and also find out if there are any blacklisted pubs you weren’t told about.

Hope that helps.

User Comment:
Thanks Liane,

I will check JT figures in a few days and report my findings here.

Also, thanks for your detailed shot of setting up JT variables, it was that post that made me re-set up my CPV tracking however since my campaigns are still waiting approval I don’t know whether it’s right this time or not…

I’ll give Marty a shout tomorrow.


User Comment:
Not sure if anyone else has noticed this with CPV Lab, but when I’ve adjusted the variables without creating a new campaign in the past, stuff still didn’t go through. Might be worth re-creating that campaign with the new variable setup and submitting the new tracking link to JT.

User Comment:

Originally Posted by polarbacon

track with has-offers…..

you will have much better luck……

Good advice, setting up has offers now.

How are you guys setting up the cost associated with the click?

Also does has offers show ROI or just the RPC and profit?


User Comment:
OK, got my tracking sorted out and that feels better, still a discrepancy, but that is understandable on mobile.

I’m only focusing on Jumptap & the top two mobile offers on Neverblue now (Jamster & Thumbplay in the US).

I won’t go into detailed reporting for obvious reasons, spent another $311.30 on traffic yesterday from Jumptap and got one conversion at $8.00.

Pretty dismal, but the most worrying thing from my point of view is I have now spent over $700 on this test and made just over $50. More importantly I don’t have anywhere near enough data in order to start dropping non-performing websites, handsets, carriers etc. because NOTHING is performing – no pattern at all.

No lesson learned in the last day. On the plus side, if it is this hard (expensive) to get started, then hopefully not too many people will swamp the mobile markets. On the negative side, I will burn through my money very quickly and not get the chance to optimise at all at this rate.

Thoughts: I probably need to improve my ad creatives – I am just using those supplied by NB at the moment. Also, looking at the small print of those two offers there is no way in fucking hell I would sign up for either of them, but I suppose marketing isn’t about what I want…

Reality: I’m off to Ad:Tech London this morning so won’t do anything for the next couple of days.

Looking forward to beers with some of the STM guys tonight.



User Comment:
How’s the general spread of traffic? I mean, is one carrier eating up more clicks?
Consider splitting up your campaigns by carrier. See if that makes a difference in terms of data.

User Comment:
Been out at Ad:Tech for a couple of days during which I dropped the daily amount of these campaigns right down.

Have just noticed ALL my conversions are from one carrier and in the evening so have dropped all te other carriers and date parted these ads.

Will see what it looks like in the morning and setup a few more camps tomorrow.

So long as I remember Lessons 1 & 2 from my opening post to this thread I should be good…

User Comment:
Sounds like you’re doing it right. The thing with mobile unlike other traffic sources the traffic that doesn’t convert DOESN’T convert at all, so as soon as you see tons of clicks and no conversions cut it ruthlessly. Its like shaping marble you just keep chipping away until all thats left is the profit. But don’t chip off to much because all you’ll be left with is a pile or rocks.

User Comment:
Mr. Russian,

What kind of traffic source will give us so much flexibility in terms of cutting the bad performers?

User Comment:
@ NumeroUno > JumpTap hands down

User Comment:
Goddamn, I must be unlucky or not Angry enough with them, my rep is giving me a hard time to simply cut the fat for my campaigns, its like I must spend 1k before I can start killing the bad performers, I’ll talk to her again tomorrow.

Thanks a lot.

User Comment:
OK overnight spend was $177.30 and revenue was $40 so still a lot of optimisation needed…

I like the analogy Russian.

Numerouno – I’m getting the same bull from JT as well. Anyway I’m gonna stick with it, have a stack of sites I want to drop and will see what happens from there…

User Comment:
Guys guys.. working with traffic reps is similar to Mr Greens post about picking up women. You have to "tickle their balls" metaphorically so applies if they are women too to get anything done.

Here is how I cock tease them… before I even run campaigns I call them up very professionally explain I’m an agency (true) with a multi million dollar budget (true if it converts) and need the best AM. I will then interview whoever they send me ask how long they worked there and get a gist for if they know their shit… IF they don’t you just flat out tell them I want to work with someone else!

Once you get your Ad Manager this will be your best friend, built rapport and keep telling them you will spend more money. If they tell you drop 1k compromise and drop $500 after negotiating then cut the fat. Their job is to make you money but they HATE tire kickers so let them know you’re serious and they will take care of you.

Example instead of depositing $10 bucks at a time drop $1k and see how much quicker they are to help you. Again, this is if your serious about running with them if you’re still "kicking tires" then don’t expect them to bend over backwards to help you.

Just my .02

User Comment:
Any updates?

User Comment:
Same story here with JT: asking to drop about $1k before cutting the fat, which kind of makes sense.. AngryRussian, what ROI do you usually aim for on mobile? 50-100%?

User Comment:
@netfig read this "Stop Focusing on ROI and Concentrate on VOLUME"

User Comment:
It got a bit depressing starting off writing a daily post with bad news so I would end up not posting and promising myself I would do so when I had a good success story to tell.

It’s been a month now, so here is an update of sorts.

I’ve tried a dozen offers now in different niches and the best I’ve managed is break even.

I decided to stick with Jumptap as a traffic source. Jumptap have been very helpful – even extending me $10k credit on net 30 terms which was handy since as I’m in the UK they don’t like any of my credit / debit cards. I’ve created separate accounts for control and dating and I’ve not really experienced any problems with Jumptap.

I’ve tried tracking using CPV Lab, Prosper202 and Hasoffers – all of them are pretty useless due to such MASSIVE discrepancy. The best tracking I have found is not to use any! Well sort of – after watching a Webinar on Offermobi a couple of weeks ago, they showed how to get the JT_ variables sent to their Subid1 – 5 fields. This cuts out the extra hop and time of hitting the tracking software, but means you can still optimise based on site, carrier, ad, mobile, keywords.

So what I have been doing is start a new campaign, use some creatives supplied by the CPA network and create some of my own. Analyse the stats usually -90% ROI after day 1 and drop any handsets / sites that are soaking up a disproportionate amount of clicks with no conversions. Rinse and repeat for a couple of days. Look at what time the conversions happen which is usually in the evening so incorporate date parting. After 3 days ROI is typically -30%. Keep doing that for a week and usually get to about break even. Further optimising at this stage just kills the volume so I change the creatives and that doesn’t make much difference either.

I’ve learnt a lot over the last month and spent a lot without any winning campaigns. I’ve followed the best practice I’ve read on here and in Liane’s training vids and webinars. I wanted to give mobile a good test and feel that I have, but with no success.

Fortunately I have other income streams as well, but to say I’m disappointed is an understatement…

Sorry for the doom and gloom, but that is my experience of 1 month in mobile.

User Comment:
Keep your head up. Follow my 14 day plan and you will no doubt find profit. Mobile isn’t "easy" but look at it this way, it makes the barrier to entry higher and competition modest.

User Comment:
Cheers Russian,

I’m sure your 14 day plan will be what I need to get me past this sticking point.

Respect to you for sharing it here.

User Comment:
@thistimenextyear – did you see this thread:…p-Optimization

Many similar experiences. I’ve seen this a lot when breaking out successful carrier/handset combination into separate campaigns

The Article Published IN 09-19-2011 02:37 PM

Share To More ()