Written by a no-hype software developer we managed to borrow for this analysis. All of the estimates, numbers, technical considerations and ideas are based on the author’s previous work experience and common software development practices. If you enjoy this type of opinionated content make sure to let us know in the comment section.
Let’s be honest – Pokemon GO launch was a wild ride. The game had some technical ups and downs, servers were all over the place and features were taken out. Bugs and technical problems were introduced – some on purpose, some by accident.
Things are currently moving in a positive direction: the communication is better, the latest version re-introduced the Poke Tracker (albeit a bit disappointing) and the game is a lot more stable.
There is a new problem for Niantic: there is simply not enough content in the game currently to keep the player base engaged. Engaged players are important, as they spend time and money inside the app.
Niantic has confirmed that three new content features/patches are coming to the game, and we believe we figured out when these updates are coming:
- Pokemon GO Trading releases in October 2016
- Pokemon GO Trainer Battles releases in December 2016/January 2017
- Pokemon GO Gen II releases in May 2017
Let’s dive in.
Pokemon Go Trading releases in October 2016
The Trade system is one of the most requested in game features, as there is a long standing tradition of trading in the Pokemon franchise. The technical implementation of Pokemon GO Trade should not be a difficult task with only a few possible bottlenecks.
We strongly believe that the Trade system will support trading with nearby trainers only, at least in the first version. There are a few possible versions on how to implement a trade system:
- #1 Optimistically – let the mobile apps perform the trade, then commit/rollback the trade after validating with the backend infrastructure
- Better UX as mosts trades will be successful
- Asynchronous, doesn’t block user actions while the server validates – user can be notified afterwards that something went wrong
- #2 Pessimistically – request the backend infrastructure to perform the trade and notify both trainers after the trade is completed
- Safer as the backend orchestrates and validates the entire trade flow
- Synchronous, blocks the app while trade is in progress (similar experience to buying Poke Coins)
Both options are valid and should not present a technical challenge, however it is more likely that Niantic will go for the option #2, as it’s more conservative and in line with their current technical solutions. Will it scale? That remains to be seen.
The estimated time required to develop, test and deploy this kind of feature is a bit more than 2 months. That estimation includes the following:
- 1 week for supporting current version, interviewing, meetings etc
- 3 weeks of development time
- 3 weeks of testing and bug fixing
- 2 weeks of regression testing and preparing the infrastructure
With that estimation in mind, it’s highly unlikely, if not almost impossible, that we’ll see the Trade system before October.
Pokemon GO Trainer Battles releases in December 2016/January 2017
When compared with the Trade system, the Trainer Battle system is more complex and expensive to develop. With that in mind, the estimate for this feature is quite bigger. The estimated time to develop, test and deploy Trainer Battles is around 4 months:
- 2 weeks product discussion and validation
- 1 – 2 weeks of technical Proof of Concepts and architecture overview
- 4 weeks of development time
- 4 weeks of testing and bug fixing
- 3 weeks of regression testing, architecture preparation and User Acceptance Testing
Unfortunately, as most of the development time is caught up with current features and bug fixing, only a small portion of the estimated work load can go in parallel.
We believe that there are already meetings being held on product planning and technical plans for this feature. It’s also quite likely that the trainer system will be implemented similar to the gym battles system:
- Most, if not all, of combat calculation and validation will happen inside the mobile app, not on the backend infrastructure
- Lag, server rollbacks, bottlenecking and other problems that online games face will ruin the experience if it’s implemented on the backend only
- Trainers will be able to fight only nearby trainers using Bluetooth or NFC as communication channel
- Local combat helps reduce server load
- It’s unlikely/impossible to have real time server-client combat communication at Pokemon GO scale
The Trainer Battles will have a profound impact on the game and will radically change the way we play the game, as players will be incentivised to meet face to face. Thus, launching this feature during Winter and in time for holiday get togethers makes perfect sense.
Pokemon GO Gen II releases in May 2017
When comparing to previously mentioned features, the introduction of a new Pokemon Generation to Pokemon GO should be rather easy, right? Well, it turns out Generation II brings in quite a few changes that need to be addressed and implemented according to the current product we have.
So, if you expected that introducing Gen II only means to “slap another 100 Pokemon in the game”, you’re wrong. Dead wrong. Here’s a list of changes and considerations that Generation II will bring into the game:
- 100 new Pokemon and 86 new moves is introduced
- Baby Pokemon
- New mechanisms for PokeStops and PokeBalls
- Diverging evolution lines and possible introduction of Evolution stones
- Example: Gloom -> Bellossom
- Moves changes and meta updates to accompany the new types and Pokemon
Overall, Pokemon GO Gen II should come conceptually after the previous two feature are introduced. That will enable the content to be multiplied and it’s longevity will increase.
We don’t have an estimate for the development of this feature, but we strongly believe it is currently not in the works. We expect it to go in product discussion somewhere in December or November, and in full production in February.
More considerations for the dates
If you think about the current state of the game and the easiest way to keep the players engaged, these dates have even more sense.
Launching Pokemon GO Trade in October gives the development team to test and prepare Trainer Battles for holidays. As most of the population stays indoor during winter, launching a social feature like Trainer Battles during that period is perfect.
Players do not like standing in the cold and battling, so while we expect the gym battle engagement to dwindle during winter, the Trainer Battles could give the game enough steam to roll through the holiday season. Of course, if they launch in December 2016.
And finally, the release of any major exploratory content should be delayed until late Spring or Summer, placing Gen II release somewhere around May 2017.