Skip to main content

[CASE STUDY] Horizontal Scaling Mobile to 10 Different Traffic Sources

I spent $500 dollars trying out 10 different traffic sources to scale horizontal running an offer for mobile CPI over at AppFlood [sign up here]     Many gurus out there say, if you have a winning campaign you can just cut n paste it to other traffic sources and you will be stacking that money.

I wanted to put that “simple and easy cut n paste method” to the test , so I took one of my killer, strong campaigns where i spent high xxxx on data, and had optimized on my favorite traffic source and decided to load up $50 bucks into 10 different traffic sources and bid everywhere the minimum CPM or CPC they require.    Now on some traffic sources this was $2.50 CPM,  on others 0.05 CPC, and so on so forth.

Here’s the screenshot of the week long results after all $500 was spent everywhere.

It is CONFIRMED:  You cannot just cut n paste a kick ass angle & winning banners onto other traffic sources and StackThatMoney – more goes into that than simple COPY+PASTE.      I continue below this screenshot…

So,  as expected its not so easy to take a campaign thats making a couple hundred a day,  copy paste it to new traffic sources and stack the money so that campaign will make you xxxx per day scaled.   Nah..ah…     I wasn’t expecting that to work like that,  so even though I lost $350 out of my $500 investment I did learn a couple of great things!

1)  The box in red, this traffic source delivered a shit ton of visits on their minimum CPM bid using my banners (the cost is actually less than 1 cent per click for those of us mathematically inclined haha)  anyway,   it was negative in a big way but since I got 6594 clicks for $50 bucks and generated back $37.80 its not a bad potential ‘winning traffic source’

How would I turn the big volume for cheap into an advantage?   I’d look at the site placement, pick the winners, separate em into their own campaigns and run some more tests.  Because there were some real gems in there,   ones that were doing 5% CR++   .

2) The box in BLUE – Volume here is limited that’s probably because my bid is low, but the CR is nuts. Look at it in CV column – 6.05%!!!  I can work with that, again I’d look at my placement IDs and see winners, then separate em into different camps and go nuts trying to increase my CPM and testing other banners to bring my CTR up.

3) The box in GREEN – this traffic source didn’t spend hardly any money,  I doubled my bids and still not much volume all week –  they didn’t have a campaign planner so the only way to find out of they had volume or not is to make a campaign.    So this is what it looks like when traff source has no volume, even though I was bidding sky high like xxCPM per day!

In Conclusion

Were the ‘experts’ right when they said its just a simple COPY+PASTE?   HELL NO!
Were the ‘experts’ right when they claimed you can stackthatmoney if you scale – YES, but it takes optimization.

Morale of the story – if you scale to other traffic sources, expect to do the exact same thing as on any other traffic source – COLLECT DATA, LOOK AT IT, OPTIMIZE IT, PROFIT :)

Hope this helps,   all the best

-iAmAttila

PS: I am aware there are more than 10 campaigns there, because I actually tested more than 10 traffic sources but at minimum bids I got hardly any volume, or some simply denied the creatives saying it violates their retarded ad policy :P

PPS: It is always the best to start and stick to 1 traffic source, and only attempt to scale if you have enough data that proves over time the angle/banner combo you are running is truly the best/strongest ;)

FacebookTwitterGoogle+LinkedInReddit
  • cnpabi

    bid the minimum bids causes deny the creatives? HOLY SHIT Traffic source. Fuck that.

    • no, you misunderstood – i always bid minimum to see if i get volume, in some places i got not much clicks because they denied the creatives so a few clicks is only them looking at the conversion flow/campaign.

  • Eric Ng

    thanks for the great post ! Can you share more how you would do the optimisation for point 2 – look at my placement IDs and see winners, then separate them into different camps or is there a link to read more about the stragegy. Thanks !

    • If you are on my mailing list you already should have received more details on how to optimize.

      If you havent yet signed up i suggest you do on the right hand side at the top so you will get real golden nuggets i wont post publicly.