Apps: what are they for, how are they created and how best to make them? Why different apps differ not only in design, but also in stuffing. Why do similar applications from different developers have different conversion rates? Let's talk about all this today.

Let's start from the basics, from the most elementary things. Apps in arbitrage – how are they used?

Apps in arbitrage, in gambling or in any other vertical are the easiest and fastest option to launch traffic.

For example, if you want to launch on a link but you don't have certain technical knowledge, you won't do it. You have to buy a domain, make a whitepage, set up everything on the link side of the whitepage and the offerpage, set up pixels and only then launch. In apps, this is all condensed down to one simple step. You unshare the app's account and launch it already directly on Facebook.

This is the simplest possible launch option, but of course there are nuances. The link from the application is very different in terms of launching, but still the application is easier to launch if you know how to work properly with creatives. The most important thing is not to launch Facebook on behalf of the app, otherwise the app may fly off, it is better to do the launch from the fanpage.

All the risks will fly on the fanpage. Directly there will be no trigger of the app itself and you will not see a tag in Facebook on the app, for example, there an hour or a day after its release in Play Market.

With app development teams, this is a common problem right now. Facebook tagging is a very critical thing, just the app appears to the team, already in 3-4 hours or the next day that Facebook app is gone. And while this is a big problem, there are guys who have found a way out and can work with it.

There are many verticals, and there are nuances to using apps in each vertical. Why aren't apps in nutra as glaringly successful as they are in gambling, for example? What do you need to know about apps in other verticals?

Let's look at the situation from the client side. A customer downloads an app – say a fitness trainer, a weight loss programme or a health and beauty app. What happens on the app side after the download? We have a maximum of 1-2 touches – the user enters the app, looks at the downloaded landing page, leaves an application on the website – and that's it, the app is no longer needed.

In gambling, he goes in, opens the app, registers, looks at the offer, at how the product is structured and whether additional authorisation via SMS or email is required. Then the player can make targeted actions with deposits, confirmation of documents and so on. So there will be a lot of taps either way, and gambling apps play a long game, unlike nutra apps.

There is a way out of this. We won't spoiler because it's very much an inside job, but we know how gut can get wrapped up in apps. But if you assess the situation now, in general a nutra is just "leave an application and wait for a call centre to call you".

So in different verticals, it all depends on the level of engagement of the users who install that app. If there is no great involvement, there is no need to make an additional application there. If on a website there is one product in the nutra, you order it and that's it, then in gambling you can do a lot inside the app.

If you make, for example, a site with 10, 20, 30 offers in the nutra, push leads thanks to a push to the offers and wrap it all up very nicely inside, then applications will have new colours.

Let's talk about app development – how hard it is, how long it takes and what it depends on. Where do you place apps – from creation to when traffic starts to launch in? How does development even happen?

Let's take a look at what development consists of, where, what, when and how it is done The first thing is selecting the design. You analyse the market, various offers. For example, you study popular games on the offers. If you study, for example, casino top games, you will see Egypt, Zeus or fruit everywhere. So you have the first layer – for example, you know that you will make an app design on the theme of Zeus.

The second stage is developing the stub. In general, in teams, usually different people do the stub and WebView applications. In very large teams – for white pages, for white applications, there is a separate staff of people. There is also a separate staff of testers. The structure is simple: you make a white paper – not in the casino theme, for example, Cross Balls or Flappy Bird – it's just an app that Google can't charge something to. At this point, you already have a developer's account in Google, with a console bought in Google Pay, with a linked card and documents. White app is uploaded to this account. At that time you prepare the WebView itself.

What will the WebView consist of? It is a plug already in the style of the casino, you can, for example, take the white app, which you did for moderation, and then this app redesigned under the design that you liked – under Zeus, Egypt and so on. Then, after this plug, WebView with all the SDKs, fluff and additional stuffing is sewn in. The stuffing is needed for tracking traffic, so that it is not lost for the whole structure, so that the tags work, directly cloaca, which will be at the stage of active application in the market to distinguish where the bots and moderators, and where the target traffic.

When the white app has passed moderation, you flood WebView with the cloak. Then you wait for moderation to pass, then 1-2 more days and you can start to launch traffic there little by little.

How much time approximately passes from the beginning of development to the upload? After all, there are applications that fall off in 2-3 days after uploading.

The first stage – account farming lasts from 4-5 days to a week. There are situations with a long moderation with several upvotes, it can take up to a week and a half. The second stage – directly moderation of the white app itself. Officially, Google says in the documentation that it can take up to 7 days.

The guys have cases when applications were released to the market in 1-3 days. But if you calculate according to the documentation, it's 7 days for the white app, plus 7 days for the WebView app and another 7 days for farming. So together it's 3 weeks. If everything goes smoothly – that's 2.5 weeks to release one app.

After an application appears, it is leased out, given to anyone who wants to upload traffic for a certain amount of money. Why do they do this, why not launch it into their own applications? How does leasing work in general?

There were many cases on the market when guys took apps directly from freelance developers who knew how those apps were made and could upload your app to the market. So you just write that you want to buy an app, transfer money, they give you some access and disappear. These developers or sellers don't explain to customers what it is and how to work with it. It's a bit incompetent, but there are really a lot of such examples. I recommend being very careful not to buy an app unless you know the person personally or your friends know them. Finally, it is much easier if you rent the app.

All the complications that await you with a purchased app are taken care of by the landlords. Why is renting better? For example, you don't know how push notifications work. In rented apps, they are customised for different GEOs, for different events, say, on signups and re-deposits go by default if the developers have set that up. Large app rental companies set this up usually in advance.

The second aspect is setting up deeplink and naming. You will be told and shown how to do all this, how to set it up and connect it. If you don't have a tracker – you can do it directly through the admin. For example, if you don't have a third server side there where you can do it and track it on your side. In general, there is a system that simplifies it all down to one company. Already within this company, you just make a rotator, which spreads your traffic inside as you need it.

To summarise.

The companies that provide apps for rent have already worked out all the wishes of webmasters. They have experience and focus on finalising some smaller points rather than on attracting elementary tracking, pooches and the like. They solve all the current issues for you. All you have to do is pay for the installs, take the naming and diplinking and start the traffic.

There are different sources of traffic. What do you have to look at when choosing apps if you're launching from Google, for example?

First, let's look at the criteria if you're launching from Google – that is, we're talking about the company's UAC. It is mandatory that the app should be at least 16+, ideally it is 18+ and it is mandatory that the app should be in the casino gambling category. Why exactly is this the case? When you launch a UAC company, you can't set up basically any targeting there. You can only add a country and run an ad campaign. If it's 12+ there, the 12 year old audience will come from YouTube, meaning kids, teens under 18 will be in the pool of audience that will be rocking the app.

Is there any optimisation from Google to ensure that the app ranks in the right vertical?

Optimisation in Google comes already at the level of setting up conversions – per install, targeted registrations and deposits. This is all customised at the stage of launching an advertising campaign. The first step of UAC launch – you accept the link. Second step – you import conversions. Third step == you add these conversions and customise them, fourth step – you directly customise your UAC like creos, headlines, texts, and launch the UAC.

What can be said about AppsFlyer?

AppsFlyer acts as a spacer between WebView and traffic. It is a service that allows you to set up and link the WebView app directly to the traffic sorts to which the ads are run, i.e. Facebook, Google, TikTok and other sorts that are integrated into it. It's like Keitaro, but on the app side – you just link the app to AppsFlyer and then you can do the integration with the traffic source.

How and from whom is it better to purchase the app? Is it better to buy from an intermediary or directly from the developer? Is there an opportunity to work one-on-one with the developer? How popular is it and does it make sense to do it at all?

Buying an app if you're just pouring into Facebook is unnecessary. If you are uploading traffic sorts like ASO, or you have UAC traffic and you want to have all the organics only on your side, or you have your own traffic sorts where you want to run an app, then only in this case it makes sense to buy an app.

From whom is it better to buy an app – from the developer, or from services that rent apps?

The second option is 100% because the service has at least some background. That is, you will know for sure that it is not a scam. If you're renting an app from reputable tenants, you know you'll be able to use it safely. If you buy an app from a developer you don't know, you might not get anything working. Or it will take 1-2 months to set it up, to understand how it all works, and at the same time you need to communicate with this developer all the time.

During all my work, I have not seen any successful cases of guys buying applications directly from developers and everything was cool. There were always some kind of problems.

Tips for buyers – what criteria should be used to select an app in general?

If it's UAC - 18+, definitely a casino category, treated by AppsFlyer.

Not all app developers and services have the problem of not treating AppsFlyer. When setting up a new AppsFlyer app, the conversions (eventals, install, signup and deposit) should be pulled to an ad campaign in Google. To do this, you need to test 50-100 installs, so that the AppsFlyer's own events are fixed and pulled into Google. In general, this is all done on the developers' side, i.e. they themselves can throw a postback in their applications and repel these targeted events.

How to choose Facebook applications? What do I need to pay special attention to?

The first thing to look at is the design of the app, what the app is and what the Android version looks like, as well as when the app was released. The date is purely for your internal analytics. You will realise that if the app has been in the marketplace for a week, you can try to run on it.

A couple of words on gambling designs – those that from Facebook customers associate with games. That is, a person clicks "install application", sees a creatives, for example, Egypt or Zeus, and recognises a game, say Gates of Olympus.

When everything is set up harmoniously, the first touch will be the creative, the second touch will be the app, the third will be the offer, and the fourth will be the targeted actions that can convert cool numbers. At each step, it's worth tracking these dependencies and expectations of the person who downloads the app and reaches the end goal, that is, so that the user doesn't have a resonance from the first touch to the last.

This is already a matter of taking a serious approach to launching traffic. A person who meticulously checks the app, design, Android version, release date, updates, reviews, rating, usually achieves his goals. There are times when an app with Android version 10, but is launched on Facebook on Android version 9, and some traffic is lost. People who see the adverts just can't download the app.

So, sometimes traffic is drained into milk if certain criteria and parameters of apps are not followed. There is a mass banning of apps in the market. What are these mass bans related to?

This is a constant story. It is rare that 2-3 months go by without an app being banned. It's Google's usual storm, it starts to remove applications with the same designs and categories from the Play Market en masse.

Now we are testing a theory about bans. Imagine that four teams release 3-4 similar apps in a short period of time. Immediately only one of the development teams' apps are banned, a week later another, and another week later a third. We know that Google starts looking at these applications and sees that on the first day they have no installs, and on the third day they have 5-6 thousand installs, even though they have no ads. There's an instant boom of installs. Developers work faster than Google, release applications faster, and he does not have time to track them and take them out for infringement at once. But the day comes when it tracks all the apps from the team and starts removing them from the Play Market. But that's just a theory we're testing.

There are different versions of apps on the market, there was a period when PWA apps appeared quickly and disappeared just as quickly. What is the difference between PWA and WebView?

PWA lived for 2-3 months on the hype and disappeared on the same hype. We tried and made PWAs, launched them simultaneously with applications, and made two ad campaigns. PWA (progressive web app) works like this: you make a website – a copy of Play Market, you make an app, wrap it all up so that the APK is downloaded to a person's phone, the person opens the APK on the phone, and WebView opens. PWA runs on the link, normal WebView apps run as apps.

It's hard to say that PWA has any envelope issues or anything else on the market. There are teams that only work with PWA and they're doing super, there are some that haven't done well with PWA at all, they only work with apps. You have to have an approach to everything. Many people think that they just need to put a direct link to the land from which they are uploading, forgetting about the many nuances. Besides, when the user sees that it's not a real application and something doesn't work in it – he can just leave.

Now there are PWA builders where you can set how much the application will weigh – for example, 10, 20, 30 MB, make ratings and reviews, translate reviews and applications if you upload to multi-GEO. In other words, the PWA app creation system is so well thought out that a person without technical knowledge can create an app in an hour and run an advert on it.But the only question is whether you know how to work with PWA. If you have always launched on a link, and abruptly switch to PWA – there are a lot of nuances to deal with.

What are the differences between Android and IOS apps? For one, there are fewer IOS apps on the market – what's the problem?

Making IOS apps is not difficult – you just pick a programming language and make a simple plug in which WebView will be wrapped.

The problem is how to work with them further. The problem is an account that can hang on farming for 30 days. After farming the account, you have to fill the white app, and the usual crosses and zeros do not pass – it must be "Crosses" with animation and bonuses. It has to be a game that will live on the account for a long time, and eventually it will live on the account for a long time with the app that the adverts run on.

Third, when you make a WebView app after white moderation, the design of that app should not be gambling. It should be dragons, candy, pirates – without anything that could be an additional trigger for the app to be in the App Store.

Is it true that iOS development is more expensive?

Yes, and that makes sense. Both by development and account, and if you count all the small elements – even design, it's more expensive. But at the end of the day IOS is not much on the market because there is no understanding of how to work with it in a way that makes it fun. But I know 3-4 teams that work with IOS apps.

The main problem – when you want to run Facebook ads on an iOS app, accounts will have a limit of 9 slots per account. You can't expand them – you have 9 slots under the entire marketplace per app. You can poke one account around because the other 8 slots the developers will give to other teams that will run traffic. In Android, for example, you can do unlimited slots on Facebook. And for iOS, you have to find 9 top teams that will do the maximum in the short life of the app. IOS is a big market, Ukraine, but also in Europe. There are Tier 3 countries where Android is still predominant, and you should not launch IOS there.