My grandmother told me that great men have their faces on bills. Observing the picture above I believe it’s fair to say she raised me well! Or that I asked Pedro to put my face in it. You’ll be the judge of that.
On to more relevant matters…
Monetisation is the end result of our analytical journey, The Player Lifecycle. I’m a firm defender that retention is king and engagement the successor to the crown. After all without players that stay and love our games there won’t be anyone to monetise. But not admitting that the end goal in freemium games as a product or service is monetisation is to be naive.
What is monetisation?
Depends who you ask. It can be a KPI, like monthly revenue or a product metric like ARPDAU. It can be in game advertising, in app purchases of game products, additional content and DLCs. By the way, your friendly glossary is at the distance of a click if you need!
In the context of this blog and for the sake of simplicity, monetisation is the game’s ability to produce revenue excluding the purchase of the game, which in freemium doesn’t exist anyway. My focus is in app purchases and advertising revenue.
The many faces of monetisation
In terms of diversity, no other area rivals monetisation.
Monetisation is found as an absolute value in the form of revenue or as relative values in the Holy Trinity of Monetisation, which, I’m proud to announce, I just made up! I’ll address the Holy Trinity of Monetisation on a later post but to avoid leaving it hanging here, it is made of ARPU (average revenue per user), ARPPU (average revenue per paying user) and conversion rate.
Monetisation is used in the following contexts:
- Reporting. Although I don’t expect to cover it deeply in the blog, total or segmented revenue is a typical monetisation KPI. It is often expressed in units of time, e.g. monthly revenue. The Holy Trinity is also present in reporting, often on daily KPIs.
- Cohort Analysis. Retention is a cohort analysis metric. Similar to retention, monetisation cohort analysis calculates variables of groups of the population segmented by their acquisition date. Instead of return rates, we are interested in the Holy Trinity variables, in the form of cumulative variables. These cumulative values are always calculated with the cohort size as the denominator. This means the cumulative ARPU of a cohort is equal to the sum of all revenue, divided by the DAU on acquisition day. It is common to observe these cumulative variables to measure UA campaign performance in the form of ROI. As an example let’s say that a campaign is acquiring users for cost per user of $1. If we calculate the cumulative ARPU of the cohort to be over $1, the ROI is positive and the campaign was successful. If possible, predicting the cumulative ARPU and the ROI early on allows to have prescriptive recommendations like increase spending or stop the campaign. The cumulative ARPU is also the most basic form of LTV but since LTV is a whole different ball game, I’ll address it in the future… and not a near one, I’m afraid.
- Ad-hoc analysis and reporting. It shouldn’t come as a surprise that many questions involve monetisation. Monetisation can also be used with retention, engagement and virality variables. Monetisation variables changes are often the hypothesis of randomised controlled trials and AB tests.
A little big trick to monetise better
So monetisation is not that hard to understand. Basically it is how much money we made divided by players or segments of players. But there’s a tad more to it. Often people obsess with monetisation.
Let me give you an inside view about my work with Miniclip’s producers and stakeholders. By default I want the AB tests to be ran for as long as possible. Producers and stakeholders often want AB tests to be as short as possible. There’s a fine tension in this, one that we live well with to be honest. On one hand I want to have as much data as possible to guarantee the quality of my team’s reports. On the other hand our games are fast moving services and producers want to ship features, tests, content, etc.
There is one time when this balance flips. When I’m more than happy to report and finish a test and producers and stakeholders want more… When is it?
If you asked me this 3 years ago I would likely say “when they are not sure there’s a revenue increase” and I would be wrong. The thing that makes monetisation look less relevant is retention. If D1 retention rate drops 1% on an AB test it is likely that we will analyse the drop and also likely that the feature or change is not implemented in case we find a statistical significant drop caused by the feature or change. That’s not a drop from 40% to 39% by the way, that’s a drop from 40% to 39.6%!
Yes we do take retention very seriously. So seriously that even if I find that the drop is not significant, I might test if major engagement metrics were affected.
That is the little big trick on how to monetise better: don’t increase your monetisation KPIs at the expense of retention or engagement. You won’t be making that extra 10% revenue tomorrow but you’ll make more than that in the long run.
The hard truth is that if you are willing to loose players for short term profit then you either don’t love your games or your players (this is my visceral love for games speaking) or you don’t understand the freemium model (and now my cerebral analytical alter ego). Whatever it is, it won’t get your face on a bill. In the long run you might not even get the bill!
And this is it. I owe you a travel through the fantastic world of The Holy Trinity of Monetisation.