+5 votes
What is the most effective way to deploy $2500 - $5000 per month on mobile user acquisition?

Would you split UA spend across Facebook and Google only or would there be other channels that one should consider for smaller budgets?
by (400 points)
Frankly, with this budget I wouldn't open Google UAC at all. Other channels are just naaah... In case you really want try something outside Facebook (or UAC), I would go to  Appsflyer's performance index (btw they wrote me that new index for H1 2019 is coming soon) to find the right fit for my app and GEO I want to target.

To the FB v UAC

1) From my experience, Facebook outperforms UAC in most of the cases. (better targeting options, probably better algorithms, etc.)

2) UAC needs a lot of money and time o get through the learning phase.

Just to give you an example - when I am starting AEO campaign in US (Android) I start with tCPA of $30.
Recommended daily budget should be at least 10x tCPA = $300 daily.
Learning phase takes from 3-5 weeks.

It will cost you something between  $6k - $11k to do the UAC by the book, and there's a large chance that it won't be ROI+.

With a small budget like this, you can afford to do the lookalike campaign on Facebook (still not an option on UAC btw)  and cherry-pick the best impressions out there.
Thank you for the advice, much appreciated.

1 Answer

+4 votes

good question - and something I think a lot about since we're often approached by smaller apps that have limited budgets and want to get the best bang for their buck.

at $2.5k - $5k budgets I typically recommend an app’s UA goals to be:

-> test and iterate on creatives to identify themes that work best.

-> establish a baseline of ROAS(d7 and beyond).

-> understand the potential scalability of the app -> can the UA activity scale beyond $2.5k to $5k.


With that context, I will concur with what David said above -> when we work with smaller apps, we typically dont explore channels other than Facebook until we hit at least $50k in monthly spend. Sometimes we test Apple Search after proving out Facebook on $10k - $15k a month - even though Apple can be effective in terms of ROAS-es, it isn’t a high priority at this stage of an app’s UA activity - because its creative testing capabilities aren’t nearly as strong as Facebook’s. 

With that context, here are some considerations we are typically mindful of for smaller apps:

-> With $2500-$5000 budgets, we typically run a maximum of 2 or 3 ad sets(oftentimes we start with just one ad set) - because the more the ad sets you have, the fewer the optimization events you can feed to Facebook, and the worse the performance can be.

-> We use these one or two ad sets to run creative tests(and to get a baseline of CTR, CPI and ROAS). Typically we test different creative ’themes’ or directions. We usually run 10 to 12 ads in an ad set, and pause underperforms. 

-> If the account has no performance history, we typically start with install optimized campaigns so Facebook starts to build up history - and also to let the ad sets build up their learning phase. If we have some data about users/purchasers available, typically we start the install optimized campaigns on a 1% lookalike of purchasers so the install optimized ads go to a high-affinity audience.

-> Frankly there isnt a massive amount of optimization or testing that's possible at $2.5k - $5k. Especially if you monetize via IAP or subscriptions, you typically wont have a lot of purchasers from that sort of budget(things can be slightly easier if you monetize via ads). What we do do is pause underperforming creatives(often) and introduce new creative themes(less often - so as to minimize resetting of FB’s learning phase). 

-> We also typically start with manual bids - and tweak bids to control performance to the extent that we can.

Your best path forward typically is to get enough learnings about your economics to be able to scale beyond the $2.5k - $5k budgets.

If it looks like your economics dont permit scaling beyond $2.5k - $5k in spend, it can certainly be helpful to iterate creatives based on CTRs, CPIs and ROAS numbers that you see. However that can often be a symptom of a deeper product level challenge. If economics dont appear to be sustainable, we typically recommend one of two directions:

      - stop spending & go back to the drawing board to make the economics workable.

      - go back to the drawing board but keep some spend going so that you get the traffic you need to test your product changes that will hopefully make the app economically sustainable(with the understanding that you’ll lose money while you test).

Expert in Facebook by (2.4k points)
edited by
Excellent answer thank you.
-> establish a baseline of ROAS(d7 and beyond).

Do you typically see a ROI+ baseline or do you look at the economics just at the app level? A lot of younger developers struggle to understand what works best for them here.
"Do you typically see a ROI+ baseline or do you look at the economics just at the app level? "

@cfarm -> didnt quite understand what you mean by the above -> I mean to get a baseline of d7, d14, d30 ROAS etc.
Sure. Let's say the baseline is -40% ROI for a specific adset / campaign but +100% at an app level. What do they do afterwards?  The hardest decisions for companies that spend <$5k a month is where they see positive app ROI, but negative ad set ROI.
really depends on the cashflow risk they can take at that point. if they have some cash buffer and are at 100%+ at the app level, I'd typically encourage them to spend/grow so they can get installs/purchases so that they can hopefully improve the 40% at the ad set level.
How does that improve the -40% at the adset level?
you have more time/budgets/creatives to run tests at the ad set level so you can (hopefully) improve the -40% via iteration.