What Playrix’s Lead Designer Revealed About Gardenscapes Level Design

Gardenscapes level design example showing different game elements, boosters, obstacles, and a match-3 board layout.

Inside Gardenscapes level design: what a Playrix lead designer revealed about elements, boosters, randomness, and board structure.

In 2017, Alexander Shilyaev, then Lead Game Designer at Playrix, publicly explained how Gardenscapes levels were actually built — including rules for combining elements, rotating mechanics in and out of levels, using randomness, placing boosters, and shaping the player’s experience.

Nearly nine years later, that presentation remains one of the clearest public glimpses into Playrix’s level-design thinking. It shows that many things players may see as random — why certain elements appear together, why others disappear for a while, or why some boards seem designed around a specific chain of actions — were part of a much more deliberate design process.

Not every rule described in 2017 should be assumed to work exactly the same way in Gardenscapes today. But Shilyaev’s presentation gives us something rare: a direct explanation from a Playrix lead designer of the principles the team was actually using to build Gardenscapes levels.

Every Good Level Needed a Clear Idea

For Shilyaev, a level should not simply be a board filled with obstacles and a limited number of moves.

It needed a clear central idea.

That idea could come from the structure of the board, an unusual combination of mechanics, an unexpected use of a familiar element, or even the random behavior built into the game.

The level designer's job was not simply to create a challenge the player could win or lose. The level was meant to deliver a specific experience and create a particular feeling.

Not Every Level Had to Be Memorable

Shilyaev also makes an important point: not every level needed a spectacular or completely original concept.

Some levels could be simpler and more conventional. In fact, that helped the more unusual levels stand out when they appeared.

Variety, then, did not come only from introducing different mechanics. It also came from pacing — alternating more straightforward levels with levels built around stronger or more distinctive ideas.

When a Helpful Element Becomes Part of the Problem

One example in the presentation involves Lemonade and an element referred to at the time as Big Fireworks.

Fireworks normally helped the player. In this particular level, however, they were positioned in a way that blocked the Lemonades from moving downward.

The element itself had not changed. Its placement had changed its role within the board.

Something the player normally viewed as helpful could temporarily become part of the challenge simply because of the way the level was arranged.

Even the Shape of the Board Could Be Part of the Experience

Another example shows an Easter-themed level whose board was shaped to resemble a rabbit.

There was no complicated gameplay reason behind it. The layout was designed to create an immediate visual reaction and connect the match-3 level with the Easter theme running in the garden at the time.

The board, in other words, was not just an empty space where mechanics were placed. Its shape could be part of the experience itself.

The Objective on the Screen Was Not Always the Real Puzzle

In another level, Lemonades were placed in the four corners of the board.

The visible objective was to collect them, but the real challenge was figuring out the sequence of actions needed to reach them.

The player had to create large matches, build power-ups, trigger explosions, charge the Lightning, and use that chain of actions to finally free the Lemonades.

The interesting part of the level was therefore not simply the goal shown on the screen. The real puzzle was understanding how the board wanted the player to approach that goal.

Randomness Was Also a Level-Design Tool

The presentation includes a level in which the player can directly influence mainly the center of the board, while moves there also create changes along the sides.

Collecting the Fireflies depends to a noticeable degree on what develops as the board changes.

This example shows that randomness on the board was not always something the designers were trying to eliminate. In some levels, uncertainty itself could be part of the intended experience.

That does not mean every Gardenscapes level was based on luck. It means randomness was one of the tools a designer could deliberately use.

Playrix Already Had a System for Rotating Elements

One of the most revealing parts of the presentation concerns the way different elements were introduced and reused.

Shilyaev describes a system that divided them into several categories.

Basic elements: mechanics that could appear throughout much of the game.

Incoming elements: elements that could appear for roughly three to six consecutive levels and then disappear for around twenty to twenty-five levels.

Special elements: more unusual mechanics used when the team wanted to create a specific type of level or a more distinctive experience.

This shows that, even in 2017, Playrix did not expect every available mechanic to remain in constant circulation.

An element could appear for a short stretch of levels, disappear from active use for a while, and return later.

There is no reason to assume that today's designers still follow exactly the same three-to-six and twenty-to-twenty-five-level intervals. Gardenscapes is now far larger and includes many more mechanics than it did at the time.

What the presentation does establish is that rotating elements in and out of active use was already part of the game's level-design philosophy during its early years.

Not Every Element Could Be Combined With Every Other Element

The team created dedicated test levels to find out which mechanics worked well together.

They then maintained internal documentation that helped level designers understand which combinations were considered suitable and which ones should be avoided.

So the elements used in a level were not simply mixed together without restrictions.

Ivy and Honey

One example involves Ivy and Honey.

Both mechanics compete for space on the board in different ways. When they were used together, they could interfere with the same tiles and create situations that became excessively difficult or simply did not play well.

The team's solution was straightforward: avoid that combination.

Water Lily and Honey

In this case, the problem was different.

When Honey covered a Water Lily and the Water Lily later moved, the resulting situation could become visually confusing.

For the design team, the player needed to be able to look at the board and quickly understand what was happening.

A combination could therefore be rejected not because it made a level too difficult, but because it made the board harder to read.

Every Playable Area of the Board Needed a Purpose

Shilyaev also shows a level involving Flowerbeds where certain corners of the board allowed the player to make matches that did little or nothing to advance the main objective.

He considered that poor level design.

If the game gives the player an area where moves can be made, that area should serve a purpose.

Otherwise, the level is not creating a meaningful decision. It is simply giving the player a place to waste moves.

An Objective That Completes Itself Is Not a Meaningful Objective

Another example involves a level with Apples and Gnomes as objectives.

To uncover the Gnomes, the player had to make so many matches and explosions that the Apples would effectively be collected along the way anyway.

According to Shilyaev, the Apples were not adding meaningful strategy. They were simply adding another counter to the screen and could even distract the player from the objective that actually mattered.

Interestingly, he says he found this exact mistake in a new batch of Playrix levels while preparing the presentation.

The rules existed, and the designers knew them, but that did not mean the team never broke those rules by accident.

Boosters Were Supposed to Help Without Breaking the Level

The presentation also reveals a very specific internal rule involving pre-level boosters.

Boards needed to begin with five suitable tiles where those boosters could be placed.

If a booster were allowed to replace an important blocker at random, it could interfere with the entire structure the level designer had built.

But the opposite problem could happen as well.

Shilyaev shows a case where two Lightning boosters could appear next to each other at the start of a level. The player could combine them immediately and remove almost the entire core obstacle setup in a single move.

That was also considered undesirable.

A booster was supposed to give the player an advantage without completely bypassing the puzzle the level had been designed around.

Big Explosions Could Be Planned From the Very First Move

In another level, the player started with two Lightning power-ups and could almost immediately trigger a major change across the board.

The goal was not simply to clear a large number of tiles.

The team wanted the player to feel, right from the beginning, that meaningful progress had already been made.

Similarly, boards could be designed to encourage chain reactions, cascades, and repeated explosions.

A spectacular sequence that looks completely lucky to the player may, in some cases, have been encouraged by the structure of the board itself.

The Designers Studied Other Major Match-3 Games

Shilyaev explains that a level designer should not simply play competing games. The designer should analyze them.

The team looked at how new mechanics were introduced, how objectives were presented, which elements were combined, and what kinds of ideas other games used in their levels.

He specifically mentions Candy Crush Saga and Candy Crush Soda Saga as useful examples for studying combinations of mechanics.

Designers then shared those observations with the rest of the team so useful knowledge did not remain with only one person.

How Different Gardenscapes Was at the Time

When the presentation was given, Gardenscapes had more than 1,650 levels and roughly 34 different modes or mechanics.

The level-design team had grown from three people to around eleven, while Playrix was releasing roughly 25 new levels per week in addition to extra content created for events.

Today's Gardenscapes operates on a completely different scale. Even so, Shilyaev's presentation shows how the production process was already being organized at a time when the game was expanding rapidly.

What Can We Actually Learn From This Today?

The 2017 presentation is not a rulebook for Gardenscapes in 2026.

We do not know which of these principles are still in use, which have been modified, or which were abandoned as thousands of additional levels and many new mechanics were introduced.

What it does give us is something we would otherwise have to guess: there was a structured design logic behind when an element appeared, what it could be combined with, how much of the board should remain useful, and what kind of experience a level was supposed to create.

That changes the way we can look at a Gardenscapes board.

It is not only about the objective, the number of moves, and the blockers visible on the screen.

Behind all of that is a more fundamental design question: what is this level trying to make the player notice, attempt, or feel?

Sources

Features of Level Design on the Example of Gardenscapes — Alexander Shilyaev

Alexander Shilyaev — Level Design in Gardenscapes, White Nights Moscow 2017

Join the Gardenscapes Strategy Community

If you enjoy discussing Gardenscapes levels, events, teams, game mechanics, and updates, you can join the Gardenscapes Strategy Facebook community and connect with other players.

Join the Gardenscapes Strategy Facebook Group