Fox Studios logo Fox StudiosGame projects, devlogs and Scratch builds
Devlogs

Cutting Features From Our Game, a Devlog on Scope

Six weeks into our small Scratch platformer, the feature list on our shared board had 31 items. We had finished 9 of them.

A project board with sticky notes for game features, several crossed out and moved to a next game column

Six weeks into our small Scratch platformer, the feature list on our shared board had 31 items. We had finished 9 of them. At the pace we were going, the game would be done sometime next summer, and none of us believed we would still care about it by then. So we sat down for one long afternoon and cut 14 features.

This devlog is about that afternoon: what cutting features from a game actually looked like for us, how we decided what stayed, and what we learned about scope that we wish someone had told us on day one. It is not a story about a perfect process. Two of our cuts were wrong, and we will get to those.

How the List Got to 31 Items

The game started as a simple idea: a small character carries a lantern through dark rooms, and the light only reaches a few tiles around you. The first prototype took a weekend and felt good. That was the problem. Every playtest gave someone a new idea, and every idea sounded small.

A wall jump. A second character. Collectible gems. A shop to spend the gems. A boss at the end of each world. A level editor. Each one was "just a few scripts." Added together, they turned a two month project into a ten month one. Nobody made a bad decision. We made 22 reasonable ones without ever looking at the total.

Cutting Features From a Game Needs a Test

Arguing feature by feature went nowhere at first. Everyone had a favorite. What finally worked was one question we asked about every item on the board:

Does this make the lantern matter more?

The lantern was the core of the game. If a feature made the light, the dark or the limited view more important, it stayed on the list. If it was fun but had nothing to do with light, it went into a file we called "next game." That framing helped a lot. Nobody had to say an idea was bad, only that it belonged somewhere else.

We also gave every remaining feature a rough size in days. That turned a vague feeling of "too much" into a number we could check against the eight weeks we were willing to spend.

What We Cut and What We Kept

FeatureDecisionReason
Lantern fuel that runs downKeptMakes light a resource
Shadow enemies that fear lightKeptDirectly about the lantern
Wall jumpCutMovement, not light
Gem shopCutNeeded a whole menu system
Second playable characterCutDoubled the art work
Level editorCutA project on its own
Three bossesCut to oneOne good fight beats three thin ones

The level editor hurt the most. One of us had already spent four days on it. Throwing away finished work is the hardest kind of cut, and we almost kept it out of guilt. We saved the project file, wrote a short note on what worked, and moved on.

Why "Finish Everything You Start" Is Bad Advice Here

A lot of advice for young developers says to always finish what you start and never give up on an idea halfway. For whole games, that is mostly right. We wrote about it ourselves in why small games beat big unfinished ones. For individual features inside a game, it is the opposite.

Finishing a feature that does not serve the core makes the game worse in two ways. It eats time you needed for polish, and it gives the player one more thing to understand that leads nowhere. Our half built wall jump was fun on its own. In the actual levels, it let players skip the dark sections we had designed, which broke the whole point of the lantern. Finishing it would have cost a week and then required redesigning six levels around it.

The Two Cuts We Got Wrong

Being honest about mistakes is the reason we keep a devlog, so here they are.

Checkpoints

We cut mid level checkpoints to save time, thinking the levels were short enough. They were not. Playtesters died in the last room and had to replay three minutes of careful dark walking, and four out of seven quit. We put checkpoints back in week nine. It took one day.

Sound for the lantern

We cut the flicker sound when fuel runs low because it felt like polish. It turned out to be a key warning. Without it, players did not notice the light shrinking until it was too late. A small cue can carry a lot of gameplay, and we now treat sound for core mechanics as part of the mechanic, not decoration.

The lesson from both: cut by asking what the player needs to understand the core, not just what the core is.

How the Cut Changed Our Schedule

After the afternoon, the board had 17 items, with 9 already done. The estimated remaining work dropped from about 60 days to 21. We actually took 29, including the two features we restored and a week of bug fixing nobody planned for. That is still far better than ten months.

The bigger change was mood. Before the cut, opening the project felt heavy. After it, every session ended with something visibly closer to done. That alone kept two of us from drifting off to other projects, which is how most small team games quietly die.

What We Do Differently Now

  • We write the one sentence core of the game at the top of the board before adding any feature.
  • Every new idea goes into a "next game" list first and only moves to the main board after a week.
  • We give each feature a size in days and keep a running total against a deadline.
  • We run the "does this make the core matter more" test every two weeks, not just once.

If you are starting your first project, the same thinking applies even earlier. Our post on why your first game should be tiny covers how to pick a scope you can actually finish.

FAQ

How do you decide which game features to cut?

Write the core idea of the game in one sentence, then ask whether each feature makes that core more important. Features that do not connect to it move to a list for a future project.

Should you throw away code for a cut feature?

Do not delete it. Save a copy of the project with the feature in it and a short note on what worked. You may reuse it in a later game, and keeping it makes the cut easier to accept.

How often should a small team review scope?

Every two weeks is a good rhythm for a project of a few months. It catches feature creep before it piles up, without turning every meeting into a debate.

The Practical Takeaway

Open your project board today and write the core of your game in one sentence at the top. Put every feature through the same test, give the survivors a size in days, and move the rest to a "next game" list. You can read more of our honest project notes in the devlogs section.

MK
Mira Kovac

Mira teaches coding to beginners and writes the step-by-step guides, always testing each one on someone who has never coded before.

More posts by Mira

More in Devlogs

01doorganico 084goad 3hot8 4465456 after5internprogram ahandfulofblessings ahmadpc ahomcybersecurity allodialresources annodex antiquesandperiodcollectibles aotc asianworldkashmir beedlectrics bloggertemplatefree bluefeatheragency bodyweightsolution campaign08blog carfixadvisor catstreetcollective cbstvsports cnfanspurchasing committeeof500yearsofdignityandresistance covertopsdjs cs-kaiyuangame cuestacomunicaciontotal cxoutlet dancebetgame deskgearhub destinyvillastay detectmagazine di33 diamondsteelbuildings donmo dyanadropship emarsysfinance emimattl escapechan eskitarih evautoplay everythingweddings-orlando farmzone feedviewerapp gamevui123 gleasongrade greetingsday hacklabasu haskell iw-kaiyun jagoanproperti jilidskripsi joltbusinessfinance juegodepokergratis juegospucca kaeru-festival latestmehndidesigns lestanzediottaviano litemagazines localcommunityhero lustrsazan mariashebets menswheels newwebstechs1 nikkit ootheday oxigenvpn plurfitness prefetial probudgetcarpetcleaning psychicmrsmartin pyaten qvantor rejoignezlogis rexzbot riselayouts ruchin rwxqsgd5 sakaltime shibiaowang shop-bot skrtips spotlesshaven strcoding tetapatrium topselller uneplanteartificielle vipertechcr webtech1223411 wo-kygaming yamaha-mekar yeknava yoybuysheets zenunwa yamaha-mekar jilidskripsi ankertoto ankertoto ankertoto