Home > Mobile > Follow-along Campaigns

---

Not shutting down this campaign until it's profitable. (8)

---

G'day from downunder...

I've started focusing on CPA for the last 3 months, but my focus has been somewhat scattered ( still is...), I've started numerous campaigns promoting sweeps on pop, spent quite alot of money on traffic with very little return, I don't have the exact figures, but I'm probably around -80% ROI.

I understand the secret to success in AM is choose something and stick to it, but I've been testing different methods to see which one I like best before putting all the focus there. So here's another shot at CPA with sweeps on pop.

I've been blindly running CPA offers, without much thought put into the heading/copy. So here I am wanting to do it properly and launching a campaign and hopefully with some guidance from you guys, turn it profitable.

My setup:
Tracker: Voluum
Don't know if this is a good method, let me know if it's flawed:
For every lander that I create, I always have links for 3 different offers. On the landing page, I setup the following link structure:
- Buttons or links to the main offer would link to: /click/3 then a script to automatically change all my buttons/links from /click/3 to /click/1 after 1 second. This is my way of detecting bots. So a real human user (with javascript enabled) enter the page, the buttons will link to /click/1 - while bots will see and click to /click/3

- A script to redirect all back button clicks to: /click/2

Then in Voluum, I set my lander : Main offer to /click/1, then a catchall monetizer network (YTZ) to /click/2 for all back buttons, and a bot bait offer for /click/3 - so when analyzing voluum data, you can see who's a bot, who converted on what.


Offer 1: Free iPhone 6S
Payout: $3.20
Geo: AU

I ripped a LP from adplexity and fixed all the coding and ran:
LP1: Original ripped lander, angle is the standard "Your {device} has been selected" .... claim now
LP2: I changed the angle to something like "Apple want to convert android users to iOS users, so they're giving away free etc...."
LP3: Different angle again. We're giving away iphones, to claim, you must agree to not use other brands, etc..

Results:
Traffic Source 1 (DNTX):

- Spent $27.28
- 0 conversions on main offer
- 1 conversion on YTZ for $2.60
Return: $2.60
- ROI: -90.51%


Traffic Source 2:
- Spent: $45
- 6 conversions on main offer: $19.20 ( 2 on LP2 and 1 on LP3, nothing on LP1)
- 5 conversions on back button (YTZ): $15.63
Return: $34.83
ROI: - 22.6%

Question: How do I consider this result? Reality is there's only 6 conversions resulting from my lander, so the actual ROI is -57% - the back button conversions has nothing to do with the landers, so shouldn't be factored into the ROI ?

-----------

Offer 2: supermarket voucher
Payout: $3.75
Geo: AU
Same as offer 1, I ripped a lander from Adplexity, ran the original, and created 2 angles ( converting from competitor objective angle )
Just started this, so will report on result later.

Take away so far...
Never just rip and run, chances are you're ripping from another newbie. Rip for the design/layout, then come up with your own angles or change it.

I've always been favoring DNTX whenever I launch a campaign, because there is no approval process, you can run whatever you want instantly. But this may have been a big mistake, as the result above shows, there's no conversions, compared to the 2nd source. This all comes down to being lazy and impatient.

I read in another STM'ers follow along ( cant remember who ) - but he got a campaign profitable, and then tried to scale to other sources, but every other source failed to convert or get profitable. This got me thinking, did I select a dud source from the start and would have failed no matter how good my landers / offer is?

Should we launch campaigns on multiple sources? because one of the key advice found on this forum is to stick to one source until you've mastered it, if the STM'er of the above follow along picked one of the other sources to start, he would not have gotten that profitable campaign and would have given up / lost money.

Any input is appreciated

---

overall i would recommend always to split test.
if you are split testing LP's why not split test traffic sources??

i would love if you could share some more stats like impressions and clicks as it will give a more bright view of the campaigns.
but it looks like with some small optimization you can reach a certain profit.
its all about how big is your reach - is there an option to increase the traffic or you at the Max the traffic sources can give you.

Overall the money lost here isn't that "big" (relativity) so you can consider this as a learning Tax you need to pay.

i would keep this campaign live for few more days when trying to increase the traffic a bit and see whats the results.

---

"Not shutting down this campaign until it's profitable. "

Wrong way of thinking - you need to have a clear cut off point. Given its a low payout offer that should convert easily I'd test max. 200$ and if you don't get close to profitability, move on to the next offer or traffic source

---

Conversions from the backbutton or any other script are part of the overall ROI. It's not uncommon to run campaigns that are slightly negative directly, but the backbutton takes them to profit.


I agree with this, I always try to use spytools just as inspiration myself. The remark about ripping from another newbie is a valid one too, I even wrote a thread on it if you wanna have a look : http://stmforum.com/forum/showthread...-what-you-copy!

There are sources with better and worse traffic, but since this is all an auction, the prices should also reflect this. This being said, there are not many networks where NOTHING works, it's usually a question of how well you are able to segment the traffic and find the profitable targets. But yes, it's possible to hit a network where almost nothing works.

You need, or let's say you SHOULD stick to one source and also a limited amount of verticals, traffic types etc ... at the BEGINNING. The reason is very simple, if you jump from one place to another, you will not learn any platform properly, you will not give any vertical a proper try and you will not learn enough about any traffic type - and all that because you don't know YET what you are doing and what to look for.

Once you feel that you know enough about the whole AM concept, feel free to explore more offers, more sources and more traffic types. This doesn't mean you have to be profitable already, very often it's the other way around as in the follow along you mentioned. NO affiliate can succeed in every part of AM, some are good with POPs, some with banners, some with facebook ... sometimes it requires several tries to find what suits you.

Good luck !

---

Thanks Christoph. I agree, in general this is the wrong mindset. However, my goal with this campaign is to learn to optimize. It's a solid offer where others are making money and are profitable, so if I'm not profitable, then it's something that I'm doing wrong, so I need to know what it is, and what to do to make it profitable, hence my thread title.

---

Thanks for that. DNTX was actually the first source I used, for the last 3 months, I don't feel like there's much more I can explore in their system, so I tried popcash + popads more recently, and I've experimented with various offers, but all sweepstakes, this is probably my 4th attempt with the iphone offer.

I feel Popads and popcash has been converting better than dntx.

---

Here is a screenshot of day 1:
http://imgur.com/5pzDRo8


Was ok conversions, look promising.

But, here's day 2
http://imgur.com/3CE5vvf

slightly less traffic, similar CTR, but only 1 conversion, and nothing on YTZ.....

This is where I'm baffled. This has happened on other campaigns too, at launch, it seem ok, look promising, and then the next day, it's dead.

Is this normal? or is this nor enough traffic to make jump to any conclusions?

BTW. the bid with the traffic source (popcash) on the 2nd day is $0.0065 - so adspend was around $60 ( and 1 conversion )

---

Conversion numbers can go up and down and they usually do. That's why you need more data to draw any conclusions. There could be many reasons for this, the most common is a change in position due to competition - they pushed you down and you are getting worse traffic now. Some days also perform better than others - weekend VS weekdays etc ...

The more volume you run, the more stable the results should look. But there will be ups and downs anyways, you need to learn to accept this. If you hit a bad day, just pause and run something else. Or refresh creatives, or try a different offer. Sometimes it's just a temporary problem with an offer and it will start to work good again.

---

Home > Mobile > Follow-along Campaigns