AppLovin Max Mediation: Top Tips and Best Practices

So a good idea usually is to work with three or four other networks on top of that. And the number networks that you should start with fluctuates because basically these networks grow and shrink. But basically right now uh 28th of March 2025 uh I would recommend to start with in uh for rewarded video interials integral Google meta and unity. That's a really good start. And if you start with these networks, that's really enough until you get to about Welcome to another episode of Tangenti 101. This is Roman from Tangent. Today I am joined again by Felix from Two and a Half Gamers. Hi, Felix. Hello. Thanks for having me again. Uh today's topic is a plen max. Uh we've already done one video with Felix about

the ad mode. We'll mention it a couple of times in this video too. Uh but if you haven't watched it, definitely watch it. Um I'll give uh Felix a chance to introduce himself just in case someone doesn't know him. Yeah. Uh thanks for having me, Rowan. Uh good to be back. Uh my name is Felix. Uh I'm an ad monetization consultant. Uh basically I've been in the industry for about eight going on nine years now. Uh basically I started off working in two DSPs and one ad network always working on the supply side. Those are the people that actually buy your inventory. After that, basically working on the demand side, I jumped across and I was a head of ad monetization for Tetris, Eastside Games, a couple of other studios. And for the last four years, I guess, or three years, I I've been a consultant essentially helping people increase their ad revenue, uh, by looking into their mediation setup, uh, ad placements in their games. Um, that's what I do

full-time now. And, uh, that's why I can talk so much about mediation platforms because I live in them. Yeah, there are not so many people in the ad monetization space at least based on what I'm seeing. Maybe because it's like UA now, right? Like you know because because of automation basically like uh before when you had placements you needed a lot of us but now when it's been automized like a big company might only need two people, right? So yeah, it's definitely less of us. Uh but also Felix, you have a newsletter. I do. Yeah. So, it's a promo for a promo. Yeah, I have an ad monetization newsletter. So, if you wanted to learn some things, uh it's called the ad monetization newsletter. You can find it on LinkedIn. And maybe you can even include a link down here. And uh yeah, please sign up, subscribe, and if you want more tips regarding UA and ad monetization and game design, please check out the Two and a Half Gamers

podcast where you can find me twice a week. Nice, nice, nice, nice. Really? Twice a week. Yeah, that's a job. This is like Yeah. real job now. Yeah, it is. But the reason why I mentioned that Felix is writing some like brilliant articles and at first he wrote one for ADM Mob and like I decided that we need to do a video about it and then he done one on uh Applan and Applan is the top one source for us in terms of the ad revenue. We also published a report and uh even for 2024 it is there on the number one. I I'm sure it's still there in 2025. So well yeah the the mediation the mediation wars have heated up and it looks like AppLov Max is coming out on top. Uh yeah yeah exactly. So what we've done here, I took Felix uh article and

turned it into the slides and I've kindly asked Felix to narrate it. So I'll go through the slides, he'll narrate it and I'll ask him uh some questions. So that's the format of the video for today. Um yeah, why apply Max Felix? So right now there's basically three main mediation platforms that make up probably 90% of mediated revenue in the mobile space, right? Uh there's ATM mob which we covered on the last video which is really good at banners which we talked about. There's Level Play by Unity and then there's AppLin Max, right? Which is the largest. And what applian max as a mediation platform does really well. It gives you access to the most amount of networks which also level play kind of does right and it's good for manipulating data. But where it really shines is just in the eCPMs and

ads that you generates. Personally uh I think this is due to the size of apploven's network which essentially means that it competes quite head-on with Google and meta to push up prices. But the main benefit that publishers see from working with Max is that if you don't mediate with Max, you don't have access to the blended rows and adoras UA campaigns that are available on AppLin, but not if you don't mediate with them. And now kind of with the rise of hybrid casual gaming where basically you're getting a lot of ad views in your day one, day two, and day three perhaps basically you can't really not run these type of UA campaigns if you want to scale a game. So they've done that very cleverly. So those are the main reasons why you mediate with Max. Um do you think any of these changes with applying Selenza gaming studios? Um, yeah. I mean, I mean, I'm sure it's going to make some people sleep better

at nights, right? Uh, I mean, yeah, I mean, sure, but you know, they still have a MMP as well, right? So, I don't know if it makes a difference them selling their gaming unit. Yeah. Yeah. And in terms of the inventory, uh, yeah, I time time will tell, but I don't think it will affect anything at all in terms of performance on Max. No. Cool. Um, here is a screenshot from your article. Uh, why are we jumping into the creating? Uh, maybe let's start with this. Yeah, let's start here. Right. So, like if you're a newly like minted game or a gaming studio is looking to set up ads, right? Usually what best practices is that if you're doing a tech launch, usually mostly what people do then is they turn on ads but only maybe with Unity. if you're building your game in Unity and then usually what you do is you look only on engagement rates and uh impressions per DAU right then during

soft launch phase that's typically when you add a mediation platform because basically in a mediation platform this allows you to have multiple networks compete for your inventory which drastically increases the price. So that basically leads us to how many networks do you actually need and usually you need about four or five networks to get started. So on max uh you don't really have an option. So if you have max mediation you will get apploven as a network and apploven exchange as a network. Apploven is basically the direct inventory. So people coming to app 111 to launch UA campaigns or e-commerce campaigns and also their resellers which is about 20 of them. So that's app 11. Then you also get access to ALX which is I think it's the old mopub network that they used to have but exchange essentially means is a group of DSPs and basically they can buy inventory on yeah your app. You can't turn those networks off. So those are

come as standard. So a good idea usually is to work with three or four other networks on top of that. And the number of networks that you should start with fluctuates because basically these networks grow and shrink. But basically right now uh 28th of March 2025 I would recommend to start with in uh for rewarded video interstitials integral Google Meta and Unity that's a really good start and if you start with these networks that's really enough until you get to about $1,000 a day in ad revenue. Got it. And then you do a breakdown I think per each ad type. Yeah. So banners and video inventory is very different, right? Because you know one is full screen and a video. So usually the demand partners that you find here are quite different. So some networks are better than others at this. But

basically if you're starting with banners on max uh I usually recommend the two networks you don't have a choice with. So I recommend them. Mint is good with banners. Google is great with banners and digital turbine is great with banners. um that's usually enough to get you started and to scale you up. But these are great. So if you're watching this in like eight months from now when whenever we we brought this out or this gets released, a good idea is just to also before integrating networks to go on any kind of comparison site that basically says the amount of size where basically networks are doing it. There's a bunch of them out there from different MMPs. just check there to make sure that basically yeah you have a um yeah you have a good understanding that these networks that I'm recommending are still on these top lists. One question that I got um from our community is u about banners. So and question is pretty uh general in terms

of when does it make sense to monetize via banners? So I feel like because the ECPMs are so low, yeah, people Yeah. So banner eCPMs are the lowest, right? Because it's the smallest set of units which has like probably the lowest amount of conversions, but typically banners make up anywhere between 15 to 18% of games revenues. Uh so it's not going to make or break you, but it's a nice little, you know, like cherry on top. Uh banners work really well when you have long gameplay and long sessions where you basically can have the banner going uninterrupted. The best practices for banners is to have them always visible from first upload. So basically you can actually increase the amount of refreshes which increases the revenue. Uh so typically puzzle games it works super well right because you're doing something and it doesn't really interrupt the game play and it doesn't really disturb anything. So that's usually when it's good to start with banners, but Yeah, I like banners

as a medium for advertising. I don't think there's anything wrong with it. Nice, nice, nice, nice. Yeah, that's really helpful. Um, we have more uh screenshots for the app setup. Maybe start with uh maybe we should start with the first one, right? Yeah. So, this is the first thing you're going to do when you go into upload max, right? So, once you get approved for an account, this is what it will look like. It's completely blank. Then you need to go into ad units. So what you need to give your coders or your programmers is essentially you need to create an ad unit. And this ad unit is essentially where you're going to put all your networks together. Uh your biders and your placement networks. Placement networks don't really exist that much anymore, but sometimes they do come up. So here you need to set up for Android and iOS a separate ad unit for all of your different ad units. So banners for Android gets one. Interstitial for Android gets one. And you set this up pretty much by yeah

choosing Android or iOS. You put in your uh yeah the name of your app and you find it uh along with your app ids and then essentially you press save and that means you essentially can go to ad units and there you have your ad unit ID which you will need to give to your developers because they will need it to start calling ads whenever users see it. So this is a great place to start and this is also where you put in your bid placements from the other networks. Yeah. And I highlighted this sentence about avoiding creating too many ad units. Um Yeah. So it's an interesting topic, right? So usually I recommend having one ad units uh for everything because it just makes everything so much easier to track. Uh if you're just getting started, just have one ad unit for all interstitials and all rewarded ads and all banner ads from one place, right? So you call even if you have three or four rewarded placements in your game, they still call just one ad unit. There's a

bunch of funky stuff you can do once you get to scale, but in the start, uh just keep it simple and start with this. Gotcha. Gotcha. Gotcha. Let me take us to the next step. What is this? So network reporting is your what can you call it idiot insurance. So essentially in every stage when you have ad revenue ultimately at some point you're going to make a mistake in your setup. And what do I mean with that? That means that either you've put a wrong placement somewhere a wrong placement ID or wrong amount for placements or put a bid ID wrong. So if you connect networks under manage networks uh so what you should do if you're working let's say with mineral you go in here and you connect the API and the SDK or the API key that basically means when you go into network comparison which you can see right down there uh basically what you can do here

is check that the amount of revenue that you're tracking in max is actually correct with the corresponding revenue in the network's dashboards because if you're monetizing in acts with Mint integral. Uh the amount of revenue you see here, you actually get paid out by Mint Girl, right? And they pay out whatever they see on their dashboard. So setting up this idiot insurance make sure that you can see if there's any discrepancies. So if anything is wrong, uh so yeah, idiot insurance. So very important to set this up. Do you know if they just pull out the data via API? Yeah, they do. I'm assuming because you have to give the API. Okay. Okay, then that's it. Yeah. Yeah. Yeah, we have the same intention like we have the same reporting for the ad revenue. Um yeah, another way another idiot insurance for you. Yeah. Um uh launching AB testing and I I got like a lot of questions about it.

Perfect. You know what I think we should do first? uh if we do advanced recording because then the AB testing I think will make a bit more sense right so basically once you launch ads in max you're going to either set up bid ad units which is the most common one but a couple years ago was actually placements but the best way to track your performance and actually see how your networks are doing is inside of advanced reporting so the best way to actually set up your reports here is exactly the way I show you in the picture here, write in the first date minus 3 days and then write yesterday. And then also check yeah whatever app that you're tracking. So Android, iOS, whatever network uh network placement and um you put in the country or ad type that you're trying to track. Usually I save a report for interstitials rewarded for both Android and iOS. And essentially what this does,

it shows you a list of all your ad units, your bidding ad units and your placement ad units. And this is essentially what you can use the information from here is what you can use for your AB testing. Uh I don't know if there's any more pictures of what actually the advanced reporting looks like. Uh if that Let's go. I don't think so, man. No. Okay. So essentially, yeah, what you're looking for here is that you want networks to contribute at least 5% of overall revenue if they're bidding or placement networks. If they're on placements, what you're looking for is a magic number of 1% fill and 1% of total ad revenue per placement for the whole waterfall. I think I had this. Yeah, there we go. Yeah, I remember seeing it. Yeah, I remember seeing it. Yeah. And I think Oh, no. Latency is you go back to AB testing, right? Uh where

we started. Mhm. So launching AB testing. So once you've actually pulled data from these advanced reporting, this is when you can actually set up AB test. So when should you do an AB test? You should do an AB test whenever you're launching something new or changing anything in your ad setup. So let's say you start with the the four networks that I recommended and you want to add a fifth. So let's say everything is going super well and oh I want to add iron source on iOS. You basically launch it by basically setting up an AB test uh that you run. So in the exposed group you launch with iron source. So how you do that is you go to the three dots in the ad unit and then you say yeah create an AB test. you name the AB test and then in the B category or exposed group that's where you would set up the iron source bid ID and then what happens is over the next couple of days you can actually see what the performance is like and you can see that is adding that

particular ad network a good or bad idea from a financial standpoint. I see. Um, yeah. Hard to describe it without the visuals, but is it like straightforward? Are there like any some like it's it's quite easy. Yeah. Yeah. Like it sounds more complicated and then you actually do it and you're like, "Okay, this is pretty easy." Okay. Okay. Okay. Got it. Yeah. Uh we have another 1% thing. Uh it's also like in continuation of AB tests. Yeah. So the the how long you should keep AB test depends entirely on how big your app is. So I've managed apps, you know, that have millions of users. If that's the case, you can probably get a statistically significant result in a couple of hours. Or I managed games that are like 15k DAO and you know it takes three weeks to get the same results, right? So it's just amount the amount of

data points simply that you need to track. Gotcha. uh adding new networks. Yeah. Uh exactly what we talked about before, but the only thing to add here is that you can actually also check for crashes. That's also one of the main things that you need to check for when you're rolling out new networks. Typically, you roll out new networks if you have a big app, 25% and then 50% rollouts. I usually just do 50% rollouts starting away. Uh sometimes you can get in trouble for it, but the amount of times it works compared to the times it doesn't is quite uh big. So nothing to majorly worry about. And do Yeah, of course you do one network at a time. Do you gradually go uh placement by placement or all of them together? So depends on if you're running a lot of AB tests. So say you have uh Android banners, Android interstitials and

Android rewarded uh all live, right? And you want to launch one new network. You can set it up all on the different ad units or you can wait. Uh there's no real best practices here. It's just like rolling out whatever makes sense financially. Mhm. Got it. Um is this Yeah, this is a we talked about this. Yeah, the 5% right. So yeah, so what drives high ad art? So ad revenue per daily active user is essentially the only thing you can do as a publisher is increasing the amount of competition. Uh that's basically bidding for your inventory. So if a network is actually not bidding at least 5% of revenue, they're not really contributing to the overall competitiveness of your ad stack, right? So, if they're not contributing 5%, you should probably move on and try to find another network because it's not it's kind of the network's buying, but it's also them losing and winning that basically pushes up the price. So, it's a bit like an

auction. It's exactly like an auction, right? If you only have people in an auction that are not bidding or not like winning, then it's not going to drive up the price, right? So, same thing here. So, that's why this rule is kind of Yeah. out there. Yeah. And we discuss it also in our ADM mob uh video. So if you watch doesn't change is the same. Exactly. Yeah. I'll stop Felix before he ruins my tease for for the adop video. Uh here are some tips that you shared for the bid floorers. Yeah. So two years ago we were probably like in terms of revenue uh in max that came from biders was like 35 or 40% and the rest was coming from placements. The good thing about placements for publishers is that it gives you higher prices and more control where you're

selling your inventory. The bad thing about placements is that they are bloody annoying to manage because it takes so much manual labor. Right now we are in a environment where we're probably 90% coming from header bidding. And this is great for networks because essentially the average CPN have kind of gone down, right? So bid floors is something you can set inside your maps console that you won't accept any bids below a certain price by geo. So usually I find it that the biggest gains you can drive from setting bid floors is usually in Q4 because there's more bid density and higher demand, right? But the thing is you have to be very careful with setting when setting bid floors because how demand fluctuates with supply, it can really burn you, right? So you should always AB test bid floors before setting them and you should always AB test to take them off, right? So let's say like things are looking really good in

December and you put a bid floor of $3 on interstitials. So basically everything below you won't accept showing an ad for then essentially that means you know that could push things up because it's December. But then when demand drops on networks in January if you still have that $3, you might lose a lot of money if you still have that in January because the whole equilibrium point of like ads moves down. So bid floors can be very powerful, but they can also hurt you quite a lot. So always AB test them. And if you have a bid floor, you should always turn test turning them off in an AB test maybe every two weeks or so because it can really burn you and that's how often things fluctuate like seasonal with seasonality. And do we recommend like indie devs to touch this at all? Yeah. I mean, like once you're over 3K a day, maybe it's still worth the testing it around a bit. Yeah.

Ad review. Yeah. So, ad review is a separate SDK connection when you do make the max link that you also need to install. Uh in my experience uh this SDK ad review can cause crashes which is super annoying because essentially it's a very integral tool to managing the ads that are shown in your game or in your app. So ad review in Max is where you can go if you have it installed to review all the ads that have been shown in your game in the last 30 days. And this is more of a pro tip. So basically you can look here like what's actually being shown. So if you're getting complaints you can then ban the ads. Um you can ban the ads or reach out to the networks to ban them. Uh but here is just kind of the command center for what's actually being shown. And if you're actually getting a lot of scale, this is where you can track like if you're getting particularly large buyers, if you want to make direct deals, which is quite hard to do in mobile, but yeah, it's just a good idea

to kind of know what's being shown in your in your in your game or app. Got it. Um, banner refresh rate. Yeah, we touched that pretty quickly. Yeah. So, when you set up in max, you're set on in your ad units, you're set up on something called uh max standard, I think, or optimized by max. And that is basically code word for refreshing every 10 seconds. Uh, so basically, you can set up anything, I think, up to 60 seconds, 30 seconds, 20, 25, whatever you want. going down to 10. Uh, generally it depends a bit on what type of game you have. I usually recommend to start on 15 and then seeing basically what it does and then you can AB test by going down to 10. Uh, but basically depending on the type of game that you have, you'll find the optimal refresh rates. So, what you're trading off there is volume over eCPM. And sometimes, you

know, going down to 10 second yields you more money uh because even though eCPM drops, you're showing so many more banners. Uh uh a good thing to do is also track retention while you're testing this because you know if you're on a 10-second refresh rate when it's flickering, that might actually harm retention in the long run. I see. Um I think two things here for the retention. Uh I assume then you have some common control where you see the data for retention and not index. You need to track that yourself. I see. Yeah. So and do usually add monetization people uh just go through different dashboards. Yeah. I mean depends on the company, right? Uh I've done it and I've been in companies where you're not supposed to do it. Uh I usually think it's a good idea that you have a idea of what's happening uh when you're changing stuff

on the ad sites. It's good to give an ad one person access to it. But you know like if you're just getting started, if you're an indie dev, you know, you're doing all these jobs anyway uh as one person, right? So yeah. Yeah, that's helpful. And for the AB test, is it the AB test that we just discussed or that's also your own? So you can actually set an AB test by basically in the exposed group you have 15 second refresh rate or 10 second refresh rate. Yeah, got it. Uh we received uh around five or six questions from our Discord community. Um so I'll ask uh Felix all of them now. Does it ever make sense to have more than one ad mediation? But like there's um essentially I've seen it being tried but it usually just results in a lot of crashes and ANRS. So nice idea in practice. Yeah, nice

idea in theory but in practice it just always crashes and burns. So don't waste your time on it. Okay. Is it a good idea to set billion floors already? very good. It's a very good idea to set bid floors, but you just need to AB test it. And basically AB test, if you turn on a bid floor on a Friday, the following Friday, it might not be a good idea anymore, right? So these all things need to be tested, right? Right. Right. Right. Yeah. And you're testing this is not not related to this, but you check your uh monetization data like every day or I mean it depends on it depends on scale, right? So like it's a question, right? Like so I can spend two hours setting different bid floor tests, right? So when is that a good idea? So you need to think about tradeoffs and opportunity cost. So basically, if I'm earning $1,000 a day, me playing around with bid floors might

yield me anywhere between 5 to 10% uplift per day. So if you're earning $1,000 a day, that means you can talk about $50 to $100 a day improvement, right? not probably worth my time what I should spend the whole day doing right but suddenly if you're a publisher earning 100k a day that's very worth my time right so it's really related to that right and when you're tinking around in the mediation platform like tinkering with these type of things it's like you can get eight to 12% gains right so if you're $500 a day like don't waste your time on this do other stuff that actually drives the business forward but if you're 100k a day Yeah, tinker. It's a great idea. Nice, nice, nice. Yeah, super helpful. Um, how important is it to keep app ads txt up to date? Uh, very important. I I I think it's like it's exactly what I said before, right? Like it depends on your scale, right? So, if

you're big, this is very very important because essentially what an adapt txt file is is a what can you say? a contract or you can say like a seal of approval what resellers are actually allowed to buy and bid on your inventory. So these lists like they update with networks every month or so and because they change resellers. If you don't have these lists up to date that means you're not getting 100% of demand from your demand partners. How much does this impact you? Hard to say but if you're big the impact can be big. So, if you're about $1,000 a day, maybe update these once a quarter. If you're $10,000 a day, you should be updating these every month. Nice. Got it. I know the big one, big companies are doing this every two weeks, for example. So, yeah. Yeah. Uh I only started running some tests and my app got blocked. What did I do wrong? Is there any chance I can get my app

back? Is this blocked on the max side up or no? Appl uh they basically usually block people they don't know. So you need to speak to someone there. Uh so basically if you've been blocked you just need to speak to someone at leven to unblock you and then you can get your app stuck. Got it. But easier said than done or what's your I mean uh if you know them it's super easy. If you don't know them, it's super hard. It's unfair, but that's the way they like to do it. Okay. Real business. Yeah. Uh, what is the current best practice setup on Max when a lot of network move to be only real life example hinting at segmentation? So best practice depends entirely on your revenue model. So let's say you're 70 30% AP ads. So relying more on APS. Uh best idea here is to start with just

uh four or five networks outside of AppLin and AppLin exchange and just keep them on bidding. Don't make it super funky, right? Uh if you're 100% ad revenue uh basically or if you're 50/50 uh IP and ad revenue, it's a good idea to start thinking about segmentation. The most effective segmentation I've seen done is basically you segment uh using your MMP on the type of UA acquisition campaign that brought your user and then essentially you have different ad frequencies for the different type of UA campaigns that bring the user. So ad row and organics are treated the same and basically you have a high ad frequency and really quick roll out on ads. IP campaigns very low ad frequency and roll out quite late in the game or user life cycle. Uh if you're 100% ads uh you know um you try and just for the best of your life showing as many ads as possible and

increasing your ad stack bid stack. So there is just a good idea to try to optimize everything. Super. Um and what did we miss? Is there anything that we didn't touch on today? I would say just get started with ads, right? Uh, usually also what helps is knowing that there's different types of end card formats and ad aggressiveness when you go on max. And basically that has to be set by your account manager. So depending on your revenue model, again, it might be a good idea to opt for less conser or more conservative or less conservative addend cards that basically finish the ads quicker. That means you also do have a trade-off in eCPM. But yeah, depends on your revenue model. That's the only thing else I can add. Super. All right, Felix, that was a very very good 30 minutes. Uh, I learned a lot. Uh, hope our audience also learned

something about applying max and their ad revenue will only grow from this moment on. Uh thanks a ton for joining us. Um we'll link uh the video for ADM Mob uh also here at the end of this video so you can enjoy more time with Felix. Um subscribe and uh until next time. Bye-bye.