🎯 How to Do a UA Review Properly by Matej Lancaric

different findings or not even six, maybe I have 10. But you don't really need to add any any extra budget. You just need to reallocate where the the budget is more efficient.
[music] [music] [music] [music] [music] [music] [music]

Okay, real talk before we start. This episode is brought to you by Potensic. And if you're a mobile game publisher tired of leaving money on the table, then listen up. Potensic is a premium ad network built by people who have actually been on both sides of this industry, game publishers, agencies, successful exits. So they get it. Here's the thing. Potensic has direct deals with Amazon, Apple, Coca-Cola, Vodafone. It's not programmatic. It's direct. Those budgets land in your game at premium CPMs, no middleman takes. And they're handle everything with their in-house team in terms of the creatives. Plus, they've partnered with playable maker. Wink wink. So they will take brands' YouTube videos or static images and actually turn them into playable ads or even video. Proper interactive format built for gaming inventory. Brands getting a gaming native creative, publishers are getting

a higher CPMs. Everyone wins. Head to potentials.com to get started or check their creative portfolio in the link in the show notes. Thank you very much Potentials for sponsoring this episode. Hello everybody. Welcome. My name is Matej Lancaric and I'm going to be the host for today. Because now we're going to talk about how to do UA reviews properly because it was actually quite popular on my Substack. So I decided to give it a go and create a short presentation for you. If you need the whole article, it's going to be linked in the show notes. And as you might have noticed, this is a solo episode. Okay, let's go directly in and start talking about UA audits. Well, as you might have noticed, I've been in the industry for over 12 years and actually spend a lot of money on the UA side of things. And I also I've done a ton of UA

reviews. And there is a review and there is a review. So it's almost like with AI. You can create a great AI videos, but you can also create the shitty AI videos. So, and some of these some of these language might be a little bit too much, but as I mentioned, not every review is a proper one. Because some people do it slightly differently than I Well, you don't really do pull only 90 days of of data and and check the MMP or just check the creatives and say, "Oh well, the CPI is going up and you know, you're hitting a creative fatigue. You should change the creatives." Well, no Sherlock. That's not a review. I mean, that's like a commenting what's what's going on. You need to have next steps. You need to do a lot of different things and I'm going to talk about it right now. So, look.

I'm usually looking at five different things. And just to give a context. Also, the context is very interesting and very important as well. I'm usually looking at the data first and then talking to people. Why? Well, because if I talk to the people already in the first place, like they can, you know, augment my ideas or views or whatever and change my point of view on different things. I want to have raw data and raw idea raw conclusions and then I ask for context. Then I can, you know, because I might have overlooked something. Anyway. So, strategy behind the spending, what the campaigns, creatives, MMP, all the dashboards and actually people that are behind everything because that is also very, very important. You need to know who is making decisions, what kind of decisions. So, it's not only one one thing, but also only five different layers.

So, look. I was looking at one game recently and looking at almost 1.2 million spent with across six different campaigns and 19 different geos. I look. Everything was fine. Unless until it wasn't. Until it wasn't. It's also very, very important. All these UA reviews, they don't really need to suggest any increase of the spending. Just usually looking at different things and how they can be improved within the same budget. Within the same same budget. Also, when you're looking at different UA channels, you need to be sure and like 100 or 200% sure that there there's no fraud because if there is it paints a very very different picture. So, I was looking at one one game with five different channels and it looks it looks okay. But then, when you remove the the one one UA channel which was

bringing a lot of fraud, then you find out that there is one channel which is taking pretty much 70% of the spending that rest is just very small. This is not really good. So, look, if you don't remove the the the fraud, then then you see five different channels, that's pretty good. Suddenly, remove it and then the whole picture changes. So, be sure that you run all the decisions from the from the data even multiple times from multiple different perspectives. Because, for example, you you pull the data and then, obviously, it's August and you're looking at day day 60 day 60 cohorts. Not you need to go back in time properly. I just recently found out like some teams they don't even know what cohort is. And that's very bad. On the other hand, I hate UA reviews

when I'm actually looking at a really great data. I don't the hate for for myself, but I hate because I I can't add anything else that would I wouldn't really do. Anyway, so, you need to really compare data and apples to apples. And if you want to compare day 7, day 28, day 60 ROAS, you need to have fully baked cohorts for all the data you have. If you're looking at 1-year ROAS, you need to do it correctly. There's also a few a few companies and few a few teams that really are kind of working very strangely with the campaign naming conventions and sometimes you start adding different countries into the into the campaign, but there is no evidence or in the campaign name. So, when looking at the campaign names, I'm I use you need to look underneath what is actually in the

campaign and don't trust the naming convention ever ever. So, go and do a pivot table spend by campaign by geo and you'll see where everything is pretty much going in terms of the spending. I did actually this exercise on my game as well and I was very like pleasantly surprised where the spend is going outside of the of the US. US was actually dragging down the whole campaign. So, then based on this, I actually split the campaign into four different campaigns and because there are best practices they say, well, you should run only worldwide campaign on one channel. Now, well, there are best practices and best practices like we started with the reviews. So, anyway, look very much under the hood from what what you are kind of seeing in the naming convention then go on layer deeper. Because then you will kind of suddenly understand where the

where the US or not even US, but they where the spend is actually coming from and then where you should you should allocate more budget into which kind of which kind of geos. Because look, you look at the different geos and suddenly you find these small pockets of of of countries where you can actually push more more money to get better better ROAS. So, if you can't really push more budget into the let's say Canada or Norway in one of your tier one country well, campaigns, put it in tier one or, you know, who says tier one should be only UK, Canada, Germany, Australia, maybe Korea and Japan. There are other countries. Always kind of bucket the the campaigns by by geos, but your own LTVs, not anyone else's LTV. Also,

not every and I see this all my game as well. There is there are no same answers for geo versus OS. So, if you are spending against again, let's say Canada, Australia, Germany on iOS, it doesn't really mean that it's going to work the same way on Android as well. So, you need to be very much sure that you are looking at this also by OS and and this level. And when you're done with geos and and all the all the different stuff, like you should always, I mean, anyway, you're looking at this uh creative level, but you need to know like what changed and when. And like looking at different things and this is where you actually ask for the context, like what kind of change the team made to understand like why everything went to suddenly. Because when you when you go from 12.3x

multiplier before from between day seven and or between day zero and day seven, that's not great. And this should be flagged by the team already. And if it's not, then you are the one who do need to do need to flag it before. Anyway, so you you look at the different things and suddenly you have six different findings or not even six, maybe you have 10. But you don't really need to add any any extra budget. You just need to reallocate where the the budget is more efficient, which is fine. You just need to do it actually I would say on a monthly basis. You can't do it really on a weekly basis, but on a monthly basis it's very important because then you do all the hygiene stuff like here where you just cut the inefficient spend because there is always inefficient spend because you need to test different things. And it's it's okay to test different things. But then you test, you evaluate the results and then cut or scale.

Very easy. Then we had then well not we but I had a different different Then I had a different different audit where we was looking at the way more revenue or monthly spend as well. And where everything was well, let's call it the interesting. Because the people made decisions based on on blended campaigns well blended campaigns. I wish it was blended campaigns was blended blended numbers. So it was organic and paid. And that painted very different picture because then the the decisions decision was made to cut cut the budget, but well, not really great. So there are different things when you're looking at the MMP and data and and pretty much LTV also when you're looking at the the team structure and then if there's nobody that actually owns UA, well, that's

that's a big that's a big problem. Even when we're talking about the different creative testing methodologies and testing budget for for creatives, that's already a very big red flag. Anyway, so looking at the LTV curves that are actually not stopping at the day 30, but is actually kind of growing until day six, day 60, day 90, then 180 asking the actual question of, "Okay, what's your goal for this game? Is it all the goal of in terms of the payback period? Is it for like 120% on day day 90 or is it 90 like is it 120% on day 180? Very much depends. And depends on what you want to do with the game and how much money you have on the bank account and also if there is still anyone actually playing the game

after day 90 because if there is not that there's none no people like you can't really aim for anything besides like day 90. So you have quick questions that you can ask yourself because before you you start kind of looking at the UA review. So is your US campaign actually US? Do you have the the day 60 with actual day 60 raw scores at day 90 or day day 120? Doesn't matter. If you remove the the worst channel from from from the UA channel mix does the strategy still make sense or like why why are you even doing this? Like where is the the LTV like how does the LTV grow from day zero to day 180? You look at the IPM then you look at like what's the actual metric that you made a decision based on. Who owns the UA? Who does the you know allocating of

the budget and all making all the different decisions? So so who is pretty much behind the all the UA? So look it's actually quite I mean it's quite easy for me but you can you can do a pretty good UA review in in couple of days. Nowadays, with all the all the AI which helps you crunch the data to a pretty normal form, it's usually way better. But, you still need to do the the work on your end and make sense of all of it. You ask the right right question for for the UA team, for the for the CEOs, and what and who who whoever that you are talking to. Because, UA reviews are very good thing. Again, you should do it every month. You just need to do it properly.

And that's it. If you need more information, you can find them on the link in the show notes. And until next time, and until you do all the different UA reviews in your life and do it properly. Thank you very much. See you next time. Bye-bye.