GameAnalytics has recently launched Grow: the user acquisition agency for mobile games, driven by experts and data insights. We thought the best way to deliver the news and details about it is to have a chat with one of the experts behind this service, Bo Behrmann Cool, user acquisition manager.
In this interview, Bo talks about how we handle things at Grow, and imparts some awesome wisdom about user acquisition strategies and best practices.
Tell us about your career so far. What did you do before starting with GameAnalytics on Grow?
I studied consumer behaviour and then worked for Gameloft, as a B2C manager, running many e-commerce shops selling mobile games online. Back then, Gameloft was doing Java games for the mobile market. When the app stores went live, we pivoted towards that, and I started working as a monetisation manager for five of Gameloft’s Android titles. It implied working directly with the production team to improve the game’s monetisation. We were working a lot with the shops and items, but also defining what kind of data we needed… It was the early stages of monestisation as we know it today, but we made some great improvements on the games, and it was great fun to be a part of it. After a while I also moved into acquisition, handling that for all Gameloft Android titles in the EMEA area – which was my first encounter with buying the right users for a game. I then left Gameloft to work with Mingleville, an educational games developer, helping them implement analytics and use data to improve their games… So, at that time I was doing both analytics and acquisition. After that, I joined GameAnalytics, which is great, as it allows me to combine looking into a high variety of metrics with buying the right users.
What’s the process behind this mysterious endeavour? For example, how does a typical Grow campaign look like from start to finish?
We start by having a talk with the client, asking a lot of questions about their game metrics, objectives, how the game is performing, what they want to achieve with it, what their budget is, and so on. As we have a lot of insight due to the data we track in the GameAnalytics tool, we can work together and advise them on which markets would be better to penetrate depending on their game and their objective. We also know which CPI prices we should be expecting depending on the strategy we choose.
Once we settled that, we spend a lot of time on discussing creatives. I think there’s not enough emphasis in the industry on creatives. I don’t have any statistical proof of this, but I think a lot of the campaigns depend on how good your banners are. Sometimes, good creatives can cut the CPI price by 50% – we’ve seen it happen. Therefore, we focus a lot on that in this preliminary phase. We ask the client to provide a large variety of creatives, all from cover screens to in-game screens, a combination of the two and also something out of the box. If the client doesn’t have the in-house capacity to create them, we take that task upon ourselves – we have a great team able to handle that.
After this initial process comes to an end, we start split testing for the first couple of weeks for both creatives and different audience segments. Typically, we run minimum twenty segments per client. When we have the winning combination of segments and creatives, we tend to allocate the remaining budget of the campaign to the best ones. We do this on different networks, depending on where we have the option to reach the specific target group we’re focusing on, and do weekly updates of segments based on the users that come in, but we also test new creatives.
When the budget is about to be spent, we send a preliminary report to check if the client wants to continue. If we’re dealing with a long term campaign, we report on a weekly basis. At the end of the process, we send a final report.
We keep in close contact with the client throughout the entire process, keeping them up to date with the progress and results, so it’s more than just uploading your creatives into a machine and hoping all will be fine.
What’s a good strategy for UA in terms of KPIs? Does it depend on where you’re at in the launch process?
Yes. When doing a campaign in the soft-launch phase, we usually concentrate on only one country and one OS. With this the strategy is to use a low sustained budget to get users in, to tweak the game. The KPIs during this process are chosen function to what the production team wants to tweak. The most common one being retention or, if they are in a more advance stage of development, the focus can shift to ROI installs.
This phase can be tricky though, as you risk optimising towards a specific segment that cannot be scaled when going global. UA in soft launch is sensitive, you need to be careful to either also run a broader audience campaign, or you need to be very specific regarding what users you want.
Then, if you have a good game, if you have great metrics, if your ARPU is high, the objective and KPI should be revenue.
There are also clients whose objective is cheap CPIs, but we don’t recommend concentrating on that alone, as cheap CPIs tend to be synonymous with poor quality users.
No matter your KPIs, it’s very important to have a good tracking solution in order to optimise.
What kind of data and metrics do you see as being the most critical for advertisers to access and optimise towards?
It depends on the objective, but all clients at the end of the day, when asked what kind of users they want, they will answer: highly engaged users that spend money. This is twofold: of course you can only optimise towards retention, but that doesn’t necessarily translate into revenue. So, you need to do a lot of segmentation and testing. Retention is a good way to check if you’re getting highly engaged users, but even so, you need to be careful, since you can buy highly engaged teenagers, who won’t spend anything. It’s tricky to only look at retention if you also want ROI. For that, you need to optimise towards revenue.
Usually the worry is that there’s more money going out then coming in when it comes to UA. How do you cope with that? Or what should the mindset and expectations be?
That is true for most campaigns, when you have a very short timespan. The mindset with many of the advertisers is that ROI should come very fast. That’s not the case. Sure, if you have a small budget, and a very narrow segmentation, it’s possible. It’s possible to find 100 installs/day that quickly convert into paying users, but if you want to scale, and most people do, then you’re looking at a larger timespan. If you have a monthly budget, it would be realistic to go for 40% ROI when the budget is spent, or for 50-60% ROI after 60 days. This is a realistic scenario for most games. At the end of the day, it’s about getting positive ROI at some point. However, this depends on how the game performs generally and the CPI that can be achieved. They need to be realistic about their LTV and when they can expect a positive return on investment.
What should anyone thinking of doing UA for their game consider in terms of budget?
A typical mistake is that clients come in with a massive budget and they want to spend it all at one time, in many countries. Our advice is to split it up, to find the right country to start in, testing parts of their budget in 2-3 countries before we spent the remaining 80%. Also, to test the platform and device, which country has the highest conversion rate, the right prices to get users for each country. It’s about taking things slowly and split testing. Be careful not to run all your budget in one go, try a few things out before you go full speed.
How can one compete with large publishers that can afford rolling out massive campaigns and high bids, when one has a medium size budget?
Well, it’s true that they’re buying a lot, and due to that the CPI prices go up. You can see that on Facebook: the CPIs are rising in many markets right now, and this is both because everyone is trying to get in, but also because these big companies are buying a big chunk of the traffic. There’s a lot of games with a good performance that are having a hard time because of this.
But if you have a high LTV user, and your CPI price is good, then even with a small budget you can achieve positive ROI. My advice is that if you have such a business case, you should go out and find investors that can help you scale up. You also need to be clever and try to be successful on more channels other than Facebook.
But the thing with UA is that it depends a lot on the quality of the game. You need a production team that continuously improves ARPDAU in order to compete, and if you can buy installs that are profitable, then you have a good starting point. That’s the most important thing: I don’t believe that being good at user acquisition alone can bring you a fortune, nowadays. It’s a matter of having a good game and a team that can quickly improve the metrics of the game, in combination with buying the right users, on the right channels. UA is not a stand-alone organ in your company that can bring you a fortune.
What is the most important thing anyone should have in mind when they start doing a UA campaign?
First, there are some things you should consider: what development phase are you in? do you feel that your product is ready to go global? what do your metrics look like? is it good enough to spend an X amount of $ on acquisition? how is the game growing organically? do we need to work more on the game? is it better to put that towards production? If you decide to move forward with UA, then you need to be very clear with what you want to achieve, your objective.
That’s step one. Then you need to be able to track your acquired users, and a great analytics tool. A tracking stack that can be integrated with your analytics solution is best. With it you’ll be able to follow the installs on all the metrics in your analytics tool, and all the events you’ve set up. Many of the tracking tools have custom events features, so if you’re a small developer this can be an OK way for you to get started. But if you want to get more advanced, you should connect you’re tracking stack with your analytics tool. This way you can benchmark and have a quicker overview.
What do you think is the most overhyped or misunderstood thing about UA strategies?
A lot of developers and advertisers are extremely focused on cheap CPIs for some reason. I don’t know if it’s because of their investors’ objectives, or not, but this tends to translate into buying poor quality users. We’ve seen many burst campaigns being very unsuccessful, having a bad effect on retention rate, many a times resulting in the game dropping below its previous position in the charts. Most times it’s better to grow your chart position with organic users, or quality installs.
Combining burst with quality is also a way, but we see too often that burst campaigns are not so successful on the long run. Of course, we’ve also seen successful burst campaign, but in order to achieve that, your game should be broadly appealing and have high install rates… I can only recommend this strategy to a handful of games out there.
What’s the biggest challenge games are facing today when it comes to UA? Is there any magical piece of advice on how to overcome it?
It comes down to improving your game metrics in order to be successful at user acquisition. It’s the most important thing. If you have good metrics, then you can start doing UA. It’s a combination of bringing the right users in, but it’s also a matter of you having a high enough LTV. In the end, it’s all about getting your LTV/user to be higher than your CPI. Make sure you can calculate your LTV and that you can estimate the LTV of your acquired installs: then you’ll be on solid ground.
There has been a lot of talk about re-engagement, lately. What’s your take on it?
If you’re collecting data from your users, it’s definitely something you should try out. We’re experimenting with it right now, and reading a lot regarding what’s going on. You need to have a lot of users in you game, about 1M… And you want to target only those who used to spend, but are not active anymore. It’s not about retargeting all your users. You need to retarget the right ones, the ones that are really valuable to you, and that you know will be re-engaged.
What are the most common mistakes advertisers usually make?
I think that they conclude too fast if something is successful or not. Again, it’s about trying out a lot of creatives, segments and channels… Many are focused only on Facebook, but there are a lot of things going on, a lot of other interesting channels. So, I think a lot of advertisers should have a wider perspective on UA. It’s true that Facebook has been the leader, and probably will continue to be, but there are other channels on which you can be successful, at a cheaper price.
What kind of mobile app ad campaigns are the most successful?
Tricky one… That’s the setting where the dev is providing the best creatives, and also has a top landing page and awesome game to go hand in hand. There are four parts: creatives, landing page and a game that has good metrics, plus a good monetisation model on top. That’s every advertiser’s dream! Then all you need is a good tracking solution, a clear objective on what your LTV is, and a budget that will allow you to expand into multiple channels.
If you want to learn more about what we offer and how we can help you in your user acquisition endeavours, check out the GameAnalytics Grow website.
Great stories about great game developers, and how they thrive in the era of data.
* You get our industry report the minute you sign up.
Editor’s Note: this post was originally published by Kalle Heikkinen, Chief Game Analyst at GameRefinery. With almost 400 game analysed under his belt, Kalle has a solid understanding of what works and…
Your are reading
Interview: Bo Behrmann Cool on User Acquisition and Grow