Azur Games shared their experience of developing Sword Play: the game’s performance deteriorated with each iteration, and the previous publisher abandoned it. The first thing that the development team did was roll back the game to the very first game version, and it is almost immediately got top #1 in stores.

After that happened, Azur Games didn’t want to get involved with the main gameplay, so they had to caution later on. They made minor adjustments to keep the game starts as they were. Producer Dmitry Savchenko invites us to take a closer look at each step.

Before our participation, the game became less and less like itself for some unknown reason. It has been completely overhauled, and the gameplay has become more complex, with varied NPC behaviours, reimagined level design, etc. As we already know, it didn’t work.

Dmitry Savchenko

But the most shocking thing was that judging by the metrics, the first game prototype was ready for commercial launch even before all the updates.

The development team’s main goal was to keep the stats intact and improve the game, making minor improvements to the gameplay and level design without affecting the proprietary visuals.

What changed?

Nothing was removed from the original assembly; the game had its own rhythm, gradually involving players with new tasks. But the developers added to the gameplay:

  • New types of weapons, such as shurikens.
  • New types of enemies with unique patterns of behaviour.
  • An armour mechanic that can only be penetrated with a few hits.
  • Bosses.

We did it because we wanted to see how the medium and long-term retention was – the original mechanics weren’t enough for 50 levels, so we needed more content.

Dmitry Savchenko

Initially, there was no monetisation in the game. Azur Games has been embedding interstitial ads and a banner at the bottom of the screen since the very first iteration.

We have implemented skins as a progress reward, skin gallery, and in-game currency. At the same time, we tried not to change the user flow too much, as all previous attempts did not end well.

Dmitry Savchenko

Azur Games barely touched on visuals. It just concluded that users couldn’t see the same skyboxes every time they played. This is why they made a basic rotation of the colour settings: blue, orange, and green. It’s just a colour, but the result was tangible.

When it came to drastic changes, the team proceeded with caution. Then any mistake will cost them more than they are willing to lose. Once the project goes global and starts optimising acquisition based on CPI volumes, a 2-3% day 90 churn will be very costly. Therefore, the iterations were with a step of no more than 10-15% of users.

The game’s success among 10% of users did not mean that the game would be successful among the other 90%. This is a matter of confidence intervals and 10% representativeness. For example, Android complicates the task due to numerous devices. The fewer users with legacy devices in the sample, the better.

There were practically no unsuccessful iterations. Once, we rolled back the essential chests because the boost was less than expected. A couple of months ago, there was also a technical rollback due to unsuccessful bug fixes related to third-party libraries.

Dmitry Savchenko
What is the result?

Azur Games is now moving towards a subtle increase in difficulty by decreasing the intensity or completely disabling the time dilation effect, making the game more dynamic.

Later, they plan to design the levels, add forest and mountains, add more particles to each action, and try contrasting colours in places with enemies.

We discuss all future updates with the studio, break it down into iterations and implement. We get a lot of ideas from developers. They came up with an increase in complexity. We just need to test it out and see what happens.

Dmitry Savchenko