Building Gaming Products | Vibhav Viswanathan CPO Mobile Premier League (MPL)
Hello and welcome to the first edition of ‘The Product Roadmap’! In true MVP fashion, this edition features Mithun (me) talking to Vibhav (who is also co-author of this newsletter) about his experience building gaming products.
A quick intro about MPL
MPL (Mobile Premier League), emerged on the gaming scene in 2018, and is one of the marquee names in the Indian product ecosystem. It focuses on real money gaming, and competes with the likes of Winzo and Zupee. Founded by Sai Srinivas Kiran Garimella and Shubham Malhotra, it has raised ~390m USD and is valued at ~2.5bn USD.
I’ve always considered gaming to be one of the sectors where product managers operate really close to the consumer, and thus are much more in control of their product’s success. Vibhav plays the part of a consumer focused PM to the T, and has had quite a journey before getting to where he is. He started his career at the Boston Consulting Group in Mumbai. He had a stint at Kalaari Capital before moving to the US for his MBA and worked at AWS in their AI/ML team. He then moved back to India to ride the Indian internet wave, and now runs products at one of the successful consumer internet companies in India!
Here’s what I talk to Vibhav about
The gaming ecosystem in India
Game development process
The metrics that govern ROI for games
Hiring and building product teams
Let’s dive in!
Vibhav, how should we visualise the gaming ecosystem in India?
So India is a large market. And the median narrative is that India is an MAU farm (it’s always been one of the top markets in terms of downloads). But that’s not really true. We know there are something like 500m gamers in the country today. It’s a very large share of the active internet audience. And out of this 25% are paying users. That’s 125m paying users for games. This is a large number by any standards. The estimated total revenue pool is over $3B USD today, and is expected to grow to ~7b USD by 2028. So there is a large an expanding opportunity in gaming.
There are few business models in games. There’s real money games (RMG). This is the space that MPL and our competitors operate in. Then there are hyper casual games and causal games (like Candy Crush). These are free to play which monetise via ads and IAP (in app purchases). Then there are mid core games like Clash Royale, PUBG mobile, and Fortnite. And then there are the super immersive AAA titles which you typically play on a console.
Out of all this, real money gaming (RMG) in India makes the most money. The key areas within RMG are Fantasy sports, card games like Rummy & Poker, and casual games like Ludo and snakes and Ladders. These are the ones that monetise the most, where users pay to be part of tournaments for prize money. The RMG platform facilitates this tournament, and then takes a cut of the total prize pool for themselves.
In India, IAP (in-app purchases) is a growing market, but is much smaller compared to developed nations like the US.
By and large, if you look at the total gaming user base in India vs that which has monetised, there is a huge difference. So there is still a large headroom for monetisation.
Can you talk a little bit about how you think about monetisation?
Sure. As I said, the largest revenue pool is within real money gaming, because IAP is not really that large in India yet.
Most games operate with specific math around a metric called the payback period at the early stages, until there is a handle of the customer LTV. Payback period is the time it takes for apps to earn what they spent on acquiring the customer. Optimal payback periods vary depending on the shelf life of the game (how quickly they go out of fashion). Typically, games that have lower shelf life (for example casual/ hyper casual free to play games) can work with a payback period of 6 months or sometimes even up to 12 months. Games that are perennial (for example real money games like Rummy), with high long term retentions, can operate with longer payback periods, even up to 24+ months.
So, money spent on user acquisition is recovered depending on how good the product itself is and how well it has engineered its payback period. The only cost to cover left to cover is the development cost which is not a large amount unless it is a AAA title.
Now in the Indian market, the reality we operate with is that of a low cost per install, so it is cheap to acquire users. But on the flip side, games here also monetise poorly via IAP and ads. Which is why the clearest monetizable area today is still RMG.
Is real money gaming so large as to make other types of games irrelevant? How do you think of IAP for example?
IAP and ads are both growing at healthy rates, but from a much smaller base. To me it looks like it will take time for these to make meaningful businesses.
So real money doesn’t make other types irrelevant.
The thing is, real money works only for specific types of games, but there is a vast ocean of opportunity within gaming. This is true everywhere - across the US, EU and Asia.
What is the game dev process like - how do you test and release games?
The game production process is fairly involved and complicated. It goes from concept to design to prototype development and then launch.
We start with the concept. The concept takes addressable market and product differentiation as inputs. Gaming studios usually conduct marketing tests to understand what works and what doesn’t.
After the concept phase, we move to game design. This is where the meat of the game is put into place. We do a lot of work on figuring out what the key mechanics of the game are, what are the retention loops that will keep people on the platform, and how the game art will look and feel.
Post this, the game moves to development. Once there is a playable build or prototype, the game is rigorously tested with employees, power users, panels etc. Critical feedback is usually worked up and potential new features are put on a forward looking roadmap. The first launch is usually rolled out to a smaller user base. Key metrics are then monitored.
Where it differs from typical consumer product building is that we spend a lot of time on the mechanics and retention loops. There are very specific mechanics in games that are known to work, for example - the ‘Match 3’ mechanic which is used in games like Candy Crush, and there is tons of literature around these specific mechanics.
For example, if I figure out that 35 year old males in coastal America are underserved, then I could build a game for them which would use the same mechanics as Candy Crush (match 3), but the game art would be very different. It would be very testosterone heavy for example - filled with cars, bikes, angry colors.
You mentioned that the concept phase takes addressable market and product differentiation into account? What do these two terms mean?
So what we do at the concept phase is to pick a very specific audience which is underserved and build the game for them. The critical understanding needed is to identify personas (Ex: Gender, age, State/language etc), and then build fresh content for them. When I say fresh content I mean that the game should connect with the target audience, but the game mechanics can still remain the same.
For example, if I figure out that 35 year old males in coastal America are underserved, then I could build a game for them which would use the same mechanics as Candy Crush (match 3), but the game art would be very different. It would be very testosterone heavy for example - filled with cars, bikes, angry colors, you get the drift.
I’ll go a little deeper. For a standalone gaming app, it is all about the quality of cohorts that are acquired. You have to ensure that you have predictability of the LTV/CACs of a cohort within the first 2-4 (D14 or D30 at the latest) weeks of engagement. With this knowledge, you will have a directional sense of the payback – is it 9 months/ 12 months/ 15 months etc. This gives a clear idea of how the marketing $ rotate in the system. Once you have this information, it is about increasing LTV (through product features) and decreasing CACs (more efficient marketing) over time. Obviously this is easier once you already have a successful game, but harder during launch with almost no data. So, it is important to keep this framework in mind, and start building data during alpha/beta launches.
One company which has done this well is Triple Dot Studios. They have this game called Woodoku, which targets middle aged women (35 Yrs) in the US. It launched in 2020, took about five to six months to develop and has since been downloaded 100 million times. They delivered fresh content for this target group but the core game mechanic was still a combination of Block Puzzle and Sudoku.
That makes a lot of sense. So now you pick large underserved audiences, and build games for them, assuming you can find an ROI effective channel to reach them.
Yes, that’s it in a nutshell.
What are the metrics you track once you launch the game?
A good leading metric is obviously retention. We want to keep people on the platform. So now if you look at the whole loop, you’ll have CAC, you’ll have retention, and you’ll have some sense of monetisation. So you will be able to build scenarios - at what CAC is it ok to acquire a user who retains for 30 days, and monetises to the tune of X dollars. And that is now the equation you try to balance for every game. In general we are able to predict within a few days/weeks after launch how well a game will do.
It’s also super hard to build a game that retains well. Specifically in RMG tournaments, we have to make sure our matching algorithm is best in class. You want to play with someone who is broadly in your league, not too good and not too bad. So we spent a lot of time getting our matchmaking engine to work really well.
People are looking for hits of adrenaline when they sign up to play a game, so the threshold for ‘being a bad product’ is extremely low. People have lots of choices, they’re not going to stick to your game if you provide even a slightly sub par experience. The advantage we have in RMG is that there is prize money involved, so that gives us some leeway, but we still index extremely hard on building a great user experience. It has to be super tight.
So the way to look at this is a trifecta - No bots + No cheating + Great matchmaking. All supported by a bug free playing experience.
Can you give us an example of what you mean by building a great user experience?
Yea so one thing I did when I came in was to make sure we have very few technical issues. That’s absolutely non-negotiable. We aim to have less than 1 in a million games to have any sort of user experience issue, measured by customer service tickets. When gamers have prize money at stake, they have very low tolerance for bullshit. So making sure the game doesn’t freeze, it works well in low bandwidth - cases like these are now table stakes. So you can imagine we have invested a lot in engineering for this.
It’s also really important to customers that there is no cheating in the game. In fact, in one of our brand campaigns we spoke explicitly about this - we do a great job in fraud models to make sure that there is zero player fraud in the system.
Users also have this fear that they are playing against bots, and that the platform is out to get them. And users can figure this out. So we actually enable users to download excel sheets which show the card distribution in their historical games, so they can do their own calculations, and see that there are no bots on the platform. We actually know that for people who experience cheating - their churn probability is 20x-30x.
So the way to look at this is a trifecta - No bots + No cheating + Great matchmaking. All supported by a bug free playing experience.
Ok, that’s fantastic. Let’s switch gears a bit and talk about product team structure.
How are your product teams (pods) organized? And maybe leading on from that, how do you split execution and strategy between marketing and product?
Pods are structured - formed and killed - based on the top level themes that the management wants to solve. Pods usually have owners for all functions i.e product, product design, game design development engineering, quality engineering, revenue, live ops, data engineering, data analysts. So that’s product pods.
Marketing consists of brand and performance. Product teams work very closely with performance marketing to ensure the product is marketed to the right set of users. This is what I talked about earlier in terms of finding a sharp underserved market and reaching them cost effectively. There is a lot of technology that goes into this.
‘Formed and killed’ is interesting - I have always thought of pods as long running structures. Do you follow a different approach?
I’d say about 80% of the structures are the same. We change up 20% depending on the new initiatives / changes in the business. However, we are willing to change more than 20% if needed. It just aligns better to what the business needs. We manage this on a quarterly basis.
You mentioned that marketing is a different vertical - what is the hand off b/w product and marketing to ensure you are getting the right set of users?
We work closely together. An important metric that most gaming companies track is the payback period. This encompasses both marketing efficiency and product experience. It forces the marketing and product functions to work together. Quality of the users can be measured by LTV. Usually, the issues arise when marketing teams are purely focused on lowering CACs, and not tuned to the whole CAC to Retention to LTV equation. If you work closely and look at the same set of metrics, these problems don’t arise. So for example, I work very closely with the business folk to make sure we are all looking at the same metrics and look at them together.
What’s the interview process like for MPL?
We typically test for 3 areas during interviews: (1) bias for action / execution chops (2) data orientation / customer obsession (3) product sense / inventing mind-set. We use a combination of Amazon’s and Facebook’s interview process to test for these areas. Usually candidates will be asked to speak about a product launch in detail to demonstrate (1) and (2). We test product sense via a jam session i.e we will solve a case study along with the candidate.
Appendix
Here’s a great report on the state of gaming in India: Lumikai state of gaming
Article talking about TripleDot Studios: Triple dot studios forbes
Retention Benchmarks for B2C companies: Lenny’s Benchmakrs
Private investments in the Gaming sector: Naavik
gawdaps