The aggregation of marginal gains (20)

There is power in small wins and slow gains - this is what I learnt after spending 4 days in Bangkok with Xavier and Jeremy from @acepowermarketing. Xavier also shared this article with me to further explain what he was trying to say.

After a long time away from AM with just one test ($150, zero conversions) under my belt, I have found some time and budget to get back into it, so here goes. I'm still very much interested in the mobile apps space (mainstream, non-adult), so this is where I will stay for now. Here are some basic rules I have set for myself to ensure that I do not crash and burn before I "get it".

• Don't be too creative for a start, just copy what works for now
• The top 10 things to remember are: Geo, Offer, Geo, Offer, Geo, Offer, Geo, Offer, Geo, Offer (Thanks Jeremy!)
• Keep it simple for now, test banners + DL
• Day job comes first (it pays the bills, and funds this venture)
• Seek 1 percent margin for improvement in everything I do
• I have very limited resources (time and money), so every dollar counts

• Hola Screen Lock
• UC Browser

• Recommendation by affiliate manager
• Recommendations matched manual spying (more about this below)

• Indonesia (High volume, cheap traffic)

• Go2Mobi (MoPub)
• Go2Mobi (Nexage)
• Instagram

• Connect to a VPN in my selected geo (Indonesia)
• Go through a list of mobile websites I curated
• Grab all the banner images and URL to a landing page (if any)
• Count and time stamp everything and put it into a CSV file
• Repeat every 5 seconds

• Got back to AM, yay!
• Made a working tool
• Setup 2 campaigns
• Started this follow along

you might still want to do some manual spying to see something that your program may not, for example, which banner or lander appears more often, and how they are linked together, which domain and placement you usually see them.

i think a proper tracker is needed regardless of your budget. its an essential tool.

for bid prices as from the traffic source manager or look at the planner and test it

Definitely a genius way of inventing your own spy tool man - wish I was this technically savvy haha. Good luck to you

Very interesting approach @getoutsg, Xavier always has some interesting insights about data processing.

May I ask what programming language or technology you used to develop your tool?

I didn't think that would really matter, but I created a macro in Firefox with Javascript. Not the cleanest of solutions but definitely works for me right now!

Update, so I'm trying to figure out my tracking issue (see post here). The problem is I don't really have the budget this month for a Voluum subscription, and was hoping to get some help to use the tracker provided by Go2Mobi. Someone else suggested I try TrackingDesk but I'm also having some issues there.

Waiting for the tracking issue to be resolved before activating my campaign.

Awesome start, looking forward to seeing how you go on!

You will be killing it man!

Can try tracker https://www.Adsbridge.com/index/pricing. It's free as a basic user.

Originally Posted by digitoma

Can try tracker https://www.Adsbridge.com/index/pricing. It's free as a basic user.

Ok so I'm still waiting for Go2Mobi to approve my campaigns...wonder why it's taking so long

...meanwhile in other news... since I started this follow-along thread and shared the tool I wrote (because I'm too poor to afford WRW or a VA), I've had a few friends in my mastermind group who contacted me who wanted me to share this tool with them. It's too raw at the moment to share to be honest, but they suggested that I offered to help others who are also cash-strapped but still need to do some spying. There's some small cost to this (additional vpn, additional VPS instance etc.) but I'll be happy to offer this for a very, very affordable fee. It won't work for all sites right now, but I'll be happy to test your target spy site, and if it works according to what you want, then take the service.

I basically need the money to fund my follow-along campaign and also hopefully, for a Voluum subscription!

If you're interested, drop me a message on Skype at felixs.im

Otherwise, just ignore I said this while we all sit here twiddling our thumbs while Go2Mobi approves my campaigns

So I got approved for Hola, and here's the results of the first $1 I spent:


Got rejected for UC Browser (?!??!) asking me for the following information:

Here's an update, got approved for UC Browser:



I'm not sure why my CTR is so low, but I'm gonna stick with it for a few more days before making any optimisation.

ROI = (earnings / spendings) / spendings * 100

Earnings = Impressions * CTR * CR * Payout

Thanks! I've stopped Hola and will review it again later today and post!

Hola launcher is a doable app, we just launched some tests on it and it's profitable off the bat for some small geos. However you could have been on the wrong affiliate network.

The easiest way to think about banners is what makes the user reach the end point? Users could like it in your face, or not in your face. That depends entirely on the perception, not ours. A good guide will be to study existing banners and see why they work(not why they don't make sense).

Ctr is a intermediary metric. Try to use cv instead if possible, app in the first place offers the advantage of high conversion data. Using ctr here is not a good judge.

Don't assume that using landers or direct linking is the best. Sometimes it's the best, sometimes it isn't. I remember some case studies where others did xxxx/day direct linking, and also others who did a landing page. It's just a "method". Sometimes it works sometimes it does not.

Manual spying is fine but I think a tracker is a necessary expense, cos hitting profitability without any optimization just isn't that likely, I've only managed to do it a couple of times.

Good luck, and look forward to updates!

Great follow along so far! I have a few notes based on what I've read above...

I would definitely recommend testing Smaato, even before testing MoPub & Nexage. Utility apps (like the ones you're running) have historically worked the best on that traffic source. MoPub is pretty expensive and compliance is very strict- net result is typically wasted spend for very little return. Nexage has more potential, but is better fit for lead gen and sweepstakes offers. Obviously this doesn't mean these sources won't work for you - I would recommend testing everything - but if you're on a limited budget, I'd recommend Smaato first and foremost.

As far as approval time goes, based on the post dates I assume you submitted over the weekend. We don't currently approve campaigns on the weekend, so anything submitted after business hours Friday (PST) will unfortunately have to wait until Monday. We know this isn't ideal, and are definitely looking to have approvals performed 24/7, but for now that's the process.

Also, about the "Won't Render" rejection...you received that email because we were not able to render your direct linked page. We don't have issues rendering Google Play Pages basically ever, so there's a strong likelihood there was an issue with the tracking issue. Always best to make sure that everything is indeed working on your end when you get an email like that. Assuming that you had everything right on your end (I'm sure you did), if we can't render the page, we still need additional information to approve the campaign. So just follow the steps provided in the email and we can get you up and running- simply letting us know that you're direct linking doesn't really help, as we don't know what you are direct linking to. We do our best to approve every single campaign that comes across our platform, but if a campaign isn't rendering...it isn't rendering, and we can't approve it for compliance reasons. I know it might seem unnecessary, but we just want to make sure a) you don't burn hundreds or thousands of dollars on traffic that leads users to a dead link, and b) that everything you are running is compliant with the traffic source policies.

Love your thread, So funny, before I read your thread & just finished setting up a campaign for UC browser in Indonesia for go2mobi also (also starting so no Voluum for me either). Did one for UC browser in india & CTR was only 1.16 so I quit that. Think UC browser is heavily saturated over 6 million downloads. Just gonna throw a couple of buck at indonesia to see if there's anymore potential then probably move on.

Found this thread by thedudeabide that seem like a really promising strategy. Worth checking out.