120. Category 4 Monetisation Models

My fellow here on Substack has called me out on LinkedIn regarding some assertions I made there about World of Warships. I will take that as peer review and an opportunity for a deep dive.

My colleague Catalin Alexandru has called me out, using language like “nonsensical”, “weird”, “grandiose”, and my descriptions of my products as “mythical”.

https://www.linkedin.com/feed/update/urn:li:activity:7323630423056564224?commentUrn=urn%3Ali%3Acomment%3A%28activity%3A7323630423056564224%2C7326532705679024129%29&replyUrn=urn%3Ali%3Acomment%3A%28activity%3A7323630423056564224%2C7326838991561900032%29&dashCommentUrn=urn%3Ali%3Afsd_comment%3A%287326532705679024129%2Curn%3Ali%3Aactivity%3A7323630423056564224%29&dashReplyUrn=urn%3Ali%3Afsd_comment%3A%287326838991561900032%2Curn%3Ali%3Aactivity%3A7323630423056564224%29

This comes just a week after calling out Matthew Emery, also on LinkedIn:

https://www.linkedin.com/posts/catalinmalexandru_founder-how-to-maximize-ltv-wo-hurting-activity-7323027278664228864-p7fS?utm_source=share&utm_medium=member_desktop&rcm=ACoAAAArO78BdvOJZ4f1Z2rIkXJIWsvg_BKxMsk

My initial concern is that this is a pattern of merit transfer from potential competitors to himself. If he can successfully “neg” more successful experts, he can place himself above those experts.

For the purposes of this article, I am going to accept his taunts as a peer review challenge for me to “put up or shut up”. I can’t address his repeated wild claims about my World of Tanks model, because I did not design World of Tanks and it is not my model. I have repeatedly told him this and it seems to have no effect. I did repair the WoT model, with the permission of Serb (Sergey Burkatovskiy) the author. To Serb’s credit, he never put his ego ahead of the wellbeing of the company (Wargaming). For this and many other reasons, I consider him one of the greatest game designers of the 21st century.

World of Warships

I consider World of Warships to be the first Category 4 monetisation model. I describe 6 categories of monetisation models here, including this one. Project Drakara was designed earlier but never made it to market. Getting new tech like this to market is extremely challenging. Drakara was designed as a mobile game but WoWs is a PC game (that later got ported to other platforms).

For those who would like to know a lot about the politics and personal challenges involved in bringing WoWs to market, and some of the obstacles that had to be overcome, this 125 minute interview with Josh Bycer has a lot of information. A lot of things (and people) had to come together to make that possible.

The backlash described in the video took place weeks prior to the interview in August of 2021. Catalin seems unfamiliar with it and presumably has not watched the video. It wasn’t imaginary or mythical. The wiki describes it briefly. It is very rare for the most loyal influencers to rebel against the game that essentially sustains them financially. The wiki also states that Serb was fired from the company he founded because of his unfortunate views on the second invasion of Ukraine. Wargaming had huge studios in Belarus, Russia, and Ukraine. Apparently they abandoned Belarus and Russia in 2022. I would assume this was at least in part to avoid another period of crippling sanctions. It’s hard for me to imagine how difficult that split must have been.

Serb’s exit is critical because due to the secrecy around the WoWs project, only two people got training from me about the proper care and feeding of the WoWs model. One of those two people was Serb, and the other was a Russian quantitative designer at the Lesta (in Saint Petersburg) studio. This makes it likely that no one now at Wargaming understands how the model is supposed to work.

As the video describes in detail, I was the target of a blackmail campaign by a personal friend of the CEO, Viktor, in 2015. I refused to agree to the secret demands of the person I call “The Catfish” in the video, and was terminated. It’s entirely possible that no one still at WG knows why I left the company, unless they have watched the video. I did notify my boss immediately upon the threat but he left the company some years ago.

These key departures could be problematic for WoWs as it’s likely that no one at the company would know how to repair it if something in the metadesign failed. Fortunately, it seems to still be going strong.

What Makes it Category 4?

This is the part I’m most reluctant to discuss because it is still a trade secret after 11 years. The game appears to be subscription based with a common and a premium currency and no pay to win elements. The lack of pay to win distinguishes the model from the WoT and WoT Blitz models. That’s not how I distinguish Category 4, however.

I define Category 4 as being a F2P Game as a Service (GaaS) with a multi tiered subscription system. WoWs looks like it only has one standard subscription. The second subscription is hidden. It kicks in automatically depending on which ship you play. If you pick a big ship, it kicks in. If you pick a small ship, we may actually pay you to play the game (with in game currency). As with WoT, play gets more expensive also as you advance in tier.

This is intentionally sneaky and makes “cloning” the game more difficult. Catalin keeps trying to impress readers by saying every successful monetisation model ever has been cloned “trillions of times”. I’m going to interpret that as drama, not a realistic number. The numbers in WoWs are designed to be difficult to extract. If it was possible to do so, the recipient of all that data might have a huge data dump but not know what to do with it. I’m sceptical, especially without any evidence, that the model has been replicated in a way that operates like the original.

The original model has no microtransactions. But it was designed to have premium ships added to the game later and charged for as a microtransaction. Very strict instructions/parameters were left on how to do this without screwing up the underlying model or making the game pay to win. As I’ve already pointed out, it’s possible no one now at WG is aware of these instructions, or that they ever existed.

Reactivity

Catalin also seems upset/incredulous that I say publicly that I intentionally brick my incomplete designs in case they are stolen. I’ve had my work stolen 8 times since 2011. I blame Tom Cruise, because he popularized the idea that if you can find a “Rain Man” and exploit them, you can get rich quick. Since this has been normalized in my lifetime, I defend my designs by always scrambling them when not in use, but making them look good. My lawfully deployed designs are of course never bricked. I want my designs to be successful and they are. The possibility that they could be “cloned trillions of times” suggests they are really quite valuable.

Each step on that Category list is a huge step up in revenue generation. The design for WoWs is quite sensitive. It has to operate within narrow parameters to function properly. If anything gets out of whack, the model “steps down” to Category 3. If it is made Pay to Win, it can even step down to Category 2. This isn’t because of the existence of a “kill switch” (I never used that term). It’s because players will notice right away if the game is tampered with and will react as their stimulation is compromised.

This happened in late 2021 when the game got the MZ treatment with pay to win elements and gambling systems. MZ makes Category 2 games. So in order to feed the game into their marketing machine, they essentially had to step the game down to Category 2. The results of this are dramatic, and they don’t have to wait 30 days to see the metrics decay across the board. I have no doubt, as Catalin says, that if revenue went up the cries of the community would have been ignored and the changes would have stayed. But that’s not what happened. The changes were rolled back, and attributed to “listening to the will of the fans”.

And their lack of spending.

More advanced Category models are even more sensitive, like any progressively advancing technology. An electric car is going to be a lot more sensitive than a horse and buggy. It’s really not advisable to let an uneducated monkey in there to play around with your Sacred Spaghetti (that’s what they called those numbers at WG, before I even arrived).

So while it’s true that the design has strayed from the OG, it hasn’t strayed that much. It isn’t policed by me, it is policed by its players. With a web2/web3 Category 5 or 6 design (or any web3 game), screwing up one number could cost millions of USD in seconds. Thus the numbers have to be extremely solid.

Clones

As Catalin points out, War Thunder was an obvious attempt to copy the design of WoT and then later WoWs. I would not be surprised if they were successful in copying the visible subscription layer, the currencies, and the microtransactions added after I left. To a lay person or even a mathematician, this might look like a “clone”. But it is very difficult to clone what you don’t understand.

From my review of WT, the community seems well aware that WT is not a clone on the monetisation level. The more that MZ types tinker with WoWs, the easier it will be to clone, but they know not to play with the Sacred Spaghetti at this point. They will create microtransactions and special event offers while leaving the core numbers alone. A competitor will copy all of the visible numbers and even some of the hidden numbers through data mining and such. But if they don’t know what to look for, and don’t know what this or that number does, they are likely to just overlook numbers designed to hide in plain sight.

Of course if you are just trying to do your job and don’t know if your game will be a big success (that will be cloned trillions of times), you don’t put in this much work to obfuscate your data. I like to plan for success and that has paid off.

While it’s exciting to hear that WG has put out a new game finally, the suggestion that Steel Hunters is a clone of any of my models seems pretty far fetched. This is the top review of the game on Steam currently:

The game currently has a “Mixed” rating. I assume that once done, this will be a Category 2 game, a step back even from World of Tanks. It might end up Category 3, but with Serb no longer at the company that seems unlikely.

This less incendiary assessment of reality makes more sense because it includes the word “most”. I don’t know how you would measure that, but if someone is trying to clone something they will achieve “most” if they copy all of the data that is available visibly and via data mining. There is a vast difference between “most” and “complete” with something even as complicated as circa 2014 WoWs.

A Category 4 game does not need to be “clone hardened”. But WoWs definitely was, very intentionally. With WoT, years before I arrived, they still made a great effort at this. And they didn’t have the foreknowledge of success. When I predicted their success (and why) in Supremacy Goods in 2011, the game was not yet profitable.

To the Future

Star Garden was shut down last year after it ran out of funds. I describe it as Category 6. But the game is incredibly complex with at least 2 years of design input from me. That includes 120 “on chain” earnable currencies that are not sold by the developer (no microtransactions). They can be traded which can generate royalties. Primary revenue comes from a progressive multi-tier subscription system. To my knowledge progressive subscription systems have never been deployed.

The game’s complexity and cost to develop are certainly its Achilles heel, and I had warned that it should not be our first title. I got out voted.

REVENGE (Everreach Labs) is a Category 5 model. While less complex, I would consider it a more technologically advanced model. It is far easier to deploy compared to Star Garden. It currently has a “complete enough” model to power the pre alpha, including fully developed oxytocin reward loops. Players spend on each other, not on the developer. Under some conditions (especially if you are streaming) you can end up with more premium currency than you came in with.

The game is F2P and capable of running in “Web2 only mode”, making it legal on Steam and Epic stores. We aim for large scale distribution. The plan is to launch the token first before the game goes live, but market uncertainty has slowed that down. I am eager to get data on it’s performance. I have an entire team of neuroscientists that want to help me analyse the data.

I’ve already demonstrated to investors the model’s ability to pay out players at any participation level with between 4 and 1000+ participants in a match. Yes, we can have over 1k players in a match.

This is just a pre alpha. Key elements like the Echo system (an intense dopamine and oxytocin booster) and marketplace (for crafting) are not in yet. This is the closest I’ve gotten to deploying one of these advanced models, and it wasn’t easy. If successful it would open the way for successful web3 development, neuroscience breakthroughs, consumer friendly monetisation models, and higher revenue generation that could support more complex game dev projects than what humans have been able to attempt so far. It might even rekindle interest in Star Garden.

But getting really new tech to market is hard. For an investor, it appears high risk without a proven successful prior run with that same tech. So when Catalin says “what have you done lately”, that hurts a bit because there is truth in that. But it’s not for lack of effort. These complex products take a lot of time, money, and a really sharp team. None of that is easy in the current market.

Comments

If you have comments or questions, drop them here. Catalin, I would recommend if you honestly want to know how something I do works or does not work, talk to me directly or ask me here where I can give you a proper response. Flaming me on LinkedIn just isn’t the proper place for that.