Probe: Post Mortem Review 2020 (Full Version)
It's finally time to do a post mortem review and tell the story of my first solo developed, solo published PC game: Probe. I will cover everything in detail from it's conception to original release as well as all changes made in updates and generalized sales/statistics graphs (within Steams NDA restrictions) up to current date to see overall what was successful and what was not about Probe. Most importantly how to improve my development process for future game releases. If you want a faster read with just the bullet points and hard data check out the abridged version here: Abridged Post Mortem
Probe: Complete Post Mortem Review 2020
Release date: April 14, 2020
Table of Contents
From Concept to Prototype
I originally came up with the concept for Probe over the holiday season in 2019. Over the past few years most of the projects I worked on with teams were never completed but a couple projects I did finish my duties on, mostly art and the current game I had/have been working on with a friend was held up at the time so I wanted to try and make a game where I only had to rely on myself to get it done and wanted to do so within 6 months to a year with very little to no budget. In a sense it was a self-imposed challenge to improve my own skills in all aspects of development, especially learning how the self publishing and release process works on Steam, but I also wanted to make a game I knew I would enjoy and play. Having some experience in game development and having done a few game jams I knew scope and scale would be important to keep to a minimum.
I was heavily inspired while re-playing some classic arcade hits and Atari games I grew up with. Atari was my first console after all and I still have it. I didn't want to just do a clone of retro hit games though. I wanted something more unique and even more challenging to beat and master. I began playing with the idea of a hybrid or fusion of different classic retro titles. Most of the inspiration and elements came from Galaga, Space Invaders and Breakout which created the initial gameplay concept. Making a retro game like the classics would also help keep the graphics and sounds to a minimalistic style and help keep the overall development scope small and achievable in the timeframe I set for myself. With the concept in my head, I didn't create any design documents. I just dove right in to making a prototype and went from there.
Choosing a Game Engine
First thing I needed to do to get started on the prototype was to determine what game engine would be best to develop it in for me. I was fairly familiar with Unity at that point but not confident enough just yet to be able to do it efficiently and within the timeframe I set so I decided to hold off on using Unity for future projects since I was already super familiar with Clickteam products and already had a pro-license for Fusion 2.5. I was even using similar Clickteam products back when I was a teenager way before indie development and publishing was as easily achievable as it is today. I had also used Clickteam Fusion 2.5 in all previous game jam games I worked on at the time and the majority of them were very easy to complete as intended by the deadline because of my knowledge of the program. So Clickteam Fusion 2.5 it was.
Prototype Visual Design & Gameplay Features
I started on the assets keeping it to basic shapes with mostly only black and white objects and what little colors I did use were neon green, blue, and yellow to try to stick to that old-school arcade look with only a black background. For some extra flair I implemented some minor white pixel particle effects on the destruction of objects and had a particle trail for the ball. I also wanted to make the play area look more like an old arcade screen with an area blocked out to represent it. In the beginning the prototype looked like this:
Most of the mechanics remained the same all the way to the release version. The prototype had the 3 enemy types, ball only breakable blocks, explosives, power-ups and even the speed boost function. Once I got around to deciding how to design the win/fail menus (and then later the main menus) I thought perhaps it would be best if I was able to release this on Steam in every language so anyone around the world can enjoy it. Also maybe it would help add more sales more than just doing it in all English. I knew it would be extremely time consuming to do this with text and translating it all for each region so I took inspiration from some mobile app menu designs using only a visual icon language instead so that no text was involved in the game at all (aside from a couple silly level designs created later). This way nothing had to be localized except the Steam store page descriptions. Now there was only one thing lacking...sound effects.
Prototype Sound Design
When it came to the sound design, to save time I purchased a large library of royalty free sound effects (& music) off Humble Bundle for around $25. However, I didn't want to stick to the fully traditional 8-bit sound effects old arcades had. I wanted to have it a little more modern sci-fi sounding but also mix it up some by making the sounds almost rhythmic/musical in nature as you broke blocks changing pitch with each hit to add a bit of randomization to how each level might sound and be more satisfying for the player while also providing audible cues of how much of the block you have left to break combined with the visual gradual change in color. Since music composition isn't exactly my strong point I also did this an attempt to make the sound effects be the game soundtrack, similar to Everyday Shooter, and so it wouldn't need a full soundtrack playlist. The only full song I used was for the logo and main menu screen. I also kept the sound settings very simple, music and sound effects on/off were the only options in the alpha version.
The Prototype: Reveal and Reactions
I had the basic prototype up and running within a couple weeks or so outside of my day job and decided to reveal it to my friends over the holidays and get their opinions...they loved it! Even their 1 year old loved watching it being played and the sound effects always grabbed his attention. Only suggestions involved tweaking a few minor things like the speeds of movement, size of the bullets and ball, and the physics fixes for the ball. Also found a few bugs that needed hammering out but that was to be expected that early on. We also agreed the black background was too plain so needed something like stars so I added a simple black and white starry background that would twinkle slightly. Tried making the stars move but was too distracting when taking up the full screen. I also asked if they thought it needed a soundtrack/background music but they insisted it was fine without it because of the sound effects. After all games like Space Invaders didn't originally have music either and they were still a blast.
For a couple months the only input I had was from close friends and family. I hadn't intended to show it publicly till it was more refined and polished. So I began working out all the gameplay bugs and tweaks to create a level template to make levels with and tested, tested and tested some more. The game still didn't have a name though. Didn't want to just call it Space Block Shooter since that sounds more like a genre than a title. I've always been a big fan of the early Star Trek series, The Next Generation series being my favorite, and while re-watching the series on Netflix I realized the ball you are launching could be considered a probe much like what they used in Star Trek to collect or transmit data. Then I realized the blocks could be the blocks of data the probe needs to collect. So I decided Probe was the perfect title that fit the theme and even sounded a bit like a game you'd see in an old arcade. This even gave it a subtle story of collecting data as you explore the galaxy and fight off hostile alien races along the way.
Deadline and Marketing Plan for Full Release
While working a full time day job I just focused on refining the gameplay making sure it was fun, challenging, and bug free in my free time. Wanted to be sure the mechanics worked as intended and were enjoyable. I never set a specific date for release in that time nor did I plan to market it anytime soon until I had some levels to actually play designed and done. Figured I would just keep working on the game as a hobby in my spare time till it was complete or presentable to the public...but then something unexpected happened....COVID 19 and economic shutdown.
While COVID negatively affected many people in many ways, including myself, I saw it as more of an opportunity to hunker down and get as much as possible done on the game since there was no telling when the shutdown was going to end. I also thought it was the perfect time for a game release. People would be at home bored wanting new games to play. So over the next few weeks of sleepless nights designing all 50 levels from start to finish, I finally got my beta build ready. Then I was ready to get my Steam "Coming Soon" page up and a release date set, April 14th 2020, a little over a month from launching the "Coming Soon" page. I didn't have a demo, and didn't consider making one at the time, but I sent out around 15 beta keys or so to friends that were interested in beta testing/playing and they helped pinpoint some bugs that I was able to hammer out before release. A few snuck by after launch but were quickly fixed and updated immediately upon discovering them.
Aside from the coming soon page being live, the most marketing I did was getting my trailer and screenshots up and sharing them on my Nate Bit Games social media accounts (didn't have a website just yet). A good friend of mine was able to get Kindred (Developer of Swords 'n Magic and Stuff) to test play it live on his stream pre-release and he had some good feedback and suggested volume controls and maybe some different power-ups that would make things easier in the first few levels to ease players into learning the mechanics. He admitted it wasn't his style of game and knew he was pretty bad at it but I gave out some keys to his viewers and they had the same feedback so I went ahead and added volume controls and better power-ups in early levels.
Majority of the rest of my time in lockdown leading up to launch went into bug fixes and implementing/testing steam achievements so I didn't do a lot of marketing outside of social media posts and the coming soon page but again, I was just trying to get the game playing great and finished before shut downs ended which when that would be was still unknown at the time. While I wanted to make it the best product possible for what it was, I knew I could always do updates after launch as well if it needed it...which turns out it did.
Here is the original release trailer and screenshots. Honestly wasn't much different in terms of aesthetics and sound from the early prototype:
Pricing was the last thing I struggled with. I didn't want to price it too high or too low. Price it too high and people could feel ripped off and with a lot of people out of work it probably wasn't the best idea either. Price it too low and people think you don't take your game seriously. My Friends said they thought it was worth $2.99 to them but again a lot of people were facing financial hardships at the time due to COVID. With Probe being an arcade style game I played with the idea of how many quarters you would have had to drop in it to beat it but turns out that was a lot of quarters so I threw that idea out. Did some research on pricing games from development books I own. Games used to be based on a $1 per hour of gameplay back before/around the indie boom in 2008. Inflation has changed that to around $2 now but also that rule really doesn't apply anymore since there are many 100+ hour games and online experiences that change constantly based on player interaction allowing infinite re-playability. Since this wasn't either kinds of those games though and it was a one time beatable game with a lot less/no re-playability I decided to stick to the old rule of $1/hr of gameplay and timed myself beating it as quickly as I could which was a little over 2 hours. For new players unfamiliar with the mechanics it would take longer though. Had reports of 4 to 5 hours for some new players to beat it. I programmed it though so I knew all the tricks and strategies and it still took 2 hours to beat. So using that method I decided on a good middle ground of $1.99.
I finally released the game on April 14th, 2020 with a 10% discount. I released it Steam (my main focus) and on itch.io and gamejolt.com as well also at 10% off. I may as well have not released on itch and gamejolt though due to very little to no sales. Probably won't publish on those platforms in the future unless it's a game jam submission/free game. Steam was much more successful, not amazing but still a good number of units sold and wishlists added.
Now that it was live I began sending press keys to curators to review the game as well as to youtubers both big and small. Only about 3 or 4 curators reviewed it around launch. As for youtubers, only one newer indie game channel I found picked it up and made a video review about the game: Hank Indie Games. Here is his original review video for Probe:
He has some great content and entertaining videos. I highly recommend checking out his channel and subscribing. I loved the hilarious intro and everything was very well presented. He brought up what his pros and cons of the game were and many of them were well founded and he had some great suggestions. Some of which I had considered implementing myself before launch but I also felt it worked fine without it too...until I got more outsider feedback.
In his review he said it was overall fun, very challenging (as intended), and a lot of content for the money. Hank said he would have easily pumped more than eight quarters into it if it were an arcade game and not feel ripped off. Gameplay itself was excellent but the game was not without issues. Graphics were okay as they were but not very visually memorable or exciting. Suggested some new enemy sprites, more colors and have the backgrounds change in some way at least from level to level. Sound effects were good but he also felt it would have been much more exciting/memorable if there was music or a soundtrack. The controls were the biggest complaint, however. He didn't realize it had controller support because there was no option on the main menu for switching controls but once he realized it had controller support it played better than with a mouse and keyboard. He also suggested some alternate key mappings for keyboard only use.
Once I heard these suggestions from an outside unbiased source, and after reading the few reviews on the Steam game page that had similar suggestions/complaints I decided to take this advice and get to work on a new major updates to really enhance the game. Over the course of the next month during lockdown I did the first 2 major updates before my place of employment was allowed to reopen and I had to go back to work but they were completed and live just in time.
Updates Post Launch
Update 1 – Controls update
Released: April 30, 2020
I decided to address the biggest issue first: the controls. Here were the changes made in that update:
Implemented switching to/from Xbox Controller and Keyboard from Main Menu/Settings
Implemented ability to choose from 1 of 3 Keyboard/Mouse control configurations including 2 keyboard only
Updated "Tutorial" to reflect the currently selected control configuration
Removed forced Steam controller input being set as Steam's default WASD configuration on boot (would remap the controller to the wrong keys displayed in game)
I tried fixing the issue of being able to click outside of the game window if you have multiple monitors but unfortunately that is a big issue/bug with the engine I used to create the game that still has yet to be fixed (I tried again recently). My game needed DirectX to run smoothly and properly and removing DirectX 11 could have been a work around but overall quality would have suffered so I just hoped that the new keyboard only mapped controls would work better for anyone without an Xbox controller. I wanted allow players to map/customize the controls but since I had already decided to run with a visual icon language instead of using text this made that impossible to do without having to use text and spend hours putting in translations for 25+ languages. I simply did not have the time for that if I was going to continue with the other improvements.
Update 2 – Soundtrack and Visual Improvements
Released: May 26, 2020
Changes made in the update:
New Enemy Sprites
15 Song Soundtrack
New colorful visual FX
Once the new control changes were finished I moved began trying to give the game a more memorable and enjoyable ambiance starting with the soundtrack. With the sound effects pack I purchased to use in the game came a lot of music as well which is where a majority of my music came from. I also sifted through some online sources like opengameart.org. Eventually I had a good playlist going that really fit the theme. Played the game several times while listening to the playlist to make sure everything fit or went with the flow of the game and if it didn't I cut it until I had my final 15 song soundtrack which I also released as a free DLC even for non-owners of the game. Check out the soundtrack here
I also decided to add some more colorful visual effects to the blocks when they were hit and when anything is destroyed which was easily implemented. Then I designed 5 new enemy sprites to alternate between on levels. Here are the ones I finalized and implemented:
With that said and done next big thing to do was improve the backgrounds on another update.
Update 3 – New Backgrounds & Website
Released: June 25, 2020
Changes in the update:
10 New Colorful Space Backgrounds
Minor Bug Fixes
When it came to the backgrounds I struggled with deciding how to approach it. I knew it needed something more like planets, asteroids, moons etc. Just didn't know what kind of style I wanted to do them in. Since the game was originally supposed to be minimalistic in design I decided to try that approach but with a retro 60's style with black/white minimalism.
To try and get live feedback from people as I played with new designs I decided to start streaming on my twitch channel (twitch.tv/natebitgames) that I hadn't touched in years hoping my old viewers might return. That did not work. Only my close friends showed up aside from a couple lurkers and was not the best way to go about it. It would take a long time to grow a recurring audience again. Not to mention once I was working again I wouldn't have the time to stream on a regular basis if at all. I still continued to work on some designs while streaming despite the lack of feedback and these were the designs I came up with during the stream:
Most people I did get feedback from didn't care too much for the style and quite frankly neither did I. The minimalistic look wasn’t working or easy to make work as backgrounds for the game. With or without color, it just wasn't gonna cut it. It clashed too much with the foreground. Luckily, because it was minimalistic, it didn't take long to do most of the designs I did so I had no problem scrapping it.
After that failed attempt I went back to the drawing board but still didn't know quite what to do. I took a little break from the game for a bit and came back with fresh eyes. I then began to look at several arcade machine designs and Atari box art for inspiration and that's when I realized the screen space around the play area doesn't have to be minimalistic to make it work with the overall style. All the arcade cabinets I looked at had a lot more detailed and colorful art than the games themselves. So I started drawing up some mimics of arcade machine art like Space Invaders using a stylized but colorful retro depiction of an alien world or setting.
Still didn't quite feel right but it was a step in the right direction. I decided I wanted it to be a little more modern. I've always loved doing photo-manipulation and using custom brushes so I found a ton of great NASA public domain images and began piecing them together with rock formations and landscapes to create my own galaxies and alien planets. After the first design I did I knew this was the best bet. I made a few and got some feedback from friends and family and it was all positive. I also was very happy with how they were turning out.
I spent the next month or so working on making 10 unique backgrounds. I realized 50 would have been A LOT to handle so I decided to settle on 10 and have them cycle every level with10 mirrored versions to add slightly more variety at least. I also began running out of ideas and resources towards the end of making all 10 so it worked out perfectly. Here are some of the images of the game with the new backgrounds: ->
About half way through working on the backgrounds my place of employment was allowed to reopen and I went back to work. Luckily, I got all the new backgrounds done just in time for the Steam Summer Sale when I launched the update. I also made a new trailer to reflect the changes:
About Half way through the summer sale though my place of employment got shut down...again. With the new free time I took that opportunity to go ahead and set up an official website with Press Kits for my company and Probe. After that was finished and a little thinking I still felt like there was something I should be easily be able to implement in the game that could give it more re-playability value and make it more competitive like an actual arcade game...a scoring system, hi-scores and Steam leaderboards.
Update 4 – Leaderboards
Released: August 18, 2020
Changes in the update:
Steam Leaderboards for All 50 Levels and Overall Score
Another New Trailer
Creating a fair and balanced but competitive scoring system was definitely a challenge but one I had fun doing with the help of my good friend/tester David Rosler. Initially I tried making all the blocks and enemies a set point value and for every time your probe bounced it would add a point as well. There was just one issue with that...someone could just keep from destroying the last block as long as possible and keep racking up points by bouncing the probe around instead but without that the scores would all be the same for every player. So I turned my attention to more of a time based scoring system. Over time a meter (on the left side of the screen) would reduce the amount that blocks are worth in points so the faster you beat the level the higher your score got. Enemies, power ups, and extra probes were still worth set point values. Unfortunately, that made it to where too many people could tie for a score still if they all beat it quickly so something else needed to be tweaked.
Eventually I implemented bonus points that would be added to the points of the blocks value based on how quickly your breaking blocks consecutively. Every block broken added a set number of points to another meter (on the right hand side of the screen) and the value of the meter is added to the points of each block when broken maxing out at an extra 500 points. The meter would quickly/constantly reduce in value so the faster you broke blocks one after the other, the higher the meter would get/stay making bonus points an important role in determining a proper score based on skill. This seemed to work really well for varying scores and was what the final scoring system became. I then created leaderboards for all 50 levels and an total overall high score.
Once that was in place and leaderboard scores were uploading to Steam successfully I added in game leaderboard displays for each level and the overall scores on the main menu. This was the most difficult and frustrating part that I thought would be simple but unfortunately there was no documentation for Steam leaderboard integration in the engine I was using or any tutorials on how to do it so I had to program via trial and error. This meant building and uploading a beta build every single time I needed to test the new code to see if it worked based on my guesswork. Many many times it did not. Took me a few weeks of brutal trial and error to get it to finally work properly but eventually I figured it out. To add a little more flare/reward I also added a special celebration display event when you beat a level with a new high score that shows the new score and your new leaderboard rank in each level. Once again, I updated the screenshots and trailer to reflect the new changes:
Shortly before I released the scoring and leaderboard update my place of employment once again was allowed to reopen (and stay open) so work started taking up a lot of my time but was still able to finish the update shortly before I started having computer problems. I had to replace some parts before I could continue working on the game. There were still some issues that needed to be fixed and updated that I discovered. After a month or so, I finally got my PC fixed and had more free time so I started to fix the issues for the final update and even made a playable 10 level demo.
Final Update – Demo, Bug Fixes and Other Improvements
Released: November 24, 2020
Changes made in the update:
Whitelisted/Fixed the exe file being flagged as malware by antivirus software
New Save Data Structure
Load Screen on the Main Menu
Hotkeys for In Game Menus
10 Level Free Demo w/ Leaderboards and Achievements
Enhanced Achievement Icons
First major issue that popped up after the last update was that some people were reporting that their antivirus software was flagging the games exe file as malware when downloaded from Steam. Unfortunately, a curator reported this issue in a negative review before I was able to fix it. I never had the issue before during the development of the game, nor did my friends/testers. Eventually my computer started to do the same thing though. Apparently the engine I was using had a default project property function called “compress at runtime” that would cause any and every exe made with it (even empty ones with no code) to be flagged falsely as malware. Other developers reported the issue, including the developer of Distraint (view his forum post about the issue here). Finally, after years of the problem in the program, the feature was removed from the engine in a recent update and my exe was no longer being flagged. Just to be safe I also submitted the exe to Avast.com (as the forum suggested) to have the game whitelisted with several antivirus software programs and so far I have had no reports of false flags from anyone since.
The next major issue came to light when I had to reformat my computer. I noticed when playing on a fresh new save or new computer every new score was marked as a new high score in game but technically it didn't mean you had a new leaderboard high score. So I set the leaderboard data to always download and save high scores locally and compare them to the current local saves and always save/upload the highest of either. This way all in game saved scores would reflect the players leaderboard scores and if any new high scores were achieved offline they would be updated when coming back online. As a side effect of this I had to add a short load screen on the main menu.
Another big issue I realized was the way I had set up the save data and stats. It would share that save data with any other Steam user on the same computer. Luckily no one discovered this and exploited this issue for leaderboard scores or achievements on alt accounts before I could fix it. I changed the local save data to be tied to the users Steam ID so it would no longer be shared. Unfortunately, I had to leave level progression/unlocks as shared save data as this would have reset any players current progress when trying to continue where they left off or that had already beat the game which would be extremely unfair to the player. It should have been something I considered before launch but unfortunately missed due to my self-imposed crunch time. (In the future I will be doing this from the start and even allow cloud saves)
Another feature that was suggested by Hank Indie Games in our direct communications was to add hot keys or menu navigation for the in game menus when beating a level/losing a life/failing to improve the flow of gameplay instead of having to move the mouse or use the joystick on a controller to get to a button and click it. I thought this was a great idea but since the player ship would move when trying to navigate the menu using the directional movements between probe launches I decided on hotkeys for each button with an icon next to each button to indicate which hotkey does what for both keyboard and Xbox controller.
One last bug/issue was with the music. When retrying a level it would skip a song and throw off the flow. Was an issue with how the engine is set up with restarting a frame but I found a easy work around that allowed the songs to continue playing in sequence. However, sometimes when the soundtrack was over it wouldn't loop back to the beginning and start playing again like it should but that was also a quick easy fix. Minor oversight in the code.
Once all the fixes were done it was time to make a free playable demo for people to try it before they buy it. Initially I only had the first 5 levels of the game included but those levels were designed to introduce just the base mechanics and then the following 5 levels were to introduce the gradual increase in difficulty with new obstacles and challenges like enemies. So I felt it was more important to include the whole first 10 levels so players could get an idea of just how much the difficulty will begin to increase with progress. I designed the game to be very challenging and I understand that it is not the type of game for everyone so I needed to make sure that was portrayed properly in the demo so people don't buy it if they would consider it a waste of money when they get stuck half way through the game and rage quit also possibly leading to a bad review. I never want my players to feel cheated. In fact, if they did buy the game, did not enjoy it and are able to get a refund still I highly encourage that. I don't want to charge someone for something they didn't like.
I even added leaderboards and achievements for the demo to give it more of the arcade feel that is in the full version and allow anyone to compete for the number 1 spots. More importantly I allowed all of the saved data (level progression, stats, and hi-scores) to carry over to the full game should they decide they want to buy it and they can then continue where they left off in the demo.
I also enhanced the achievement icons to be more varied by adding numbers to achievements that related to specific stats like 500 blocks broken, 10 probes launched, 100 shots fired, etc.
Finally, after more extensive testing on multiple computers and accounts I came to the conclusion it was finally finished. To celebrate I released 10 free Probe desktop wallpapers of each levels background for download on the Probe page of my website (here).
From this point on there will be no updates, at least in the near future, unless there is a new bug that is reported I need to/can fix. However, if I hit a personal milestone in units sold I will probably add some new free content. I can't disclose specific numbers of this milestone in units sold due to the Steam NDA but I will post an announcement of the update coming soon when/if I do hit that goal. I can say currently I am about 2/5 of the way to that goal.
Speaking of sales, lets break down the month to month generalized sales statistics from launch to current date.
Generalized Sale Statistics Graphs
While under NDA with Steam I am unable to share specific numbers on sales statistics, I can do visual generalized graphs showing spikes in sales and comparing them to events such as updates and discount events. Here are my unit sales peaks and falls in a generalized graph from release to current day (December 3rd, 2020). I marked special events that contributed to each sales peak. It is a bit inaccurate data in some areas but the spikes are definitely correct for when sales were the best.
Here is a heatmap of how many copies were sold in which countries. Brightest colors are highest sales and the darker it gets the less were sold in that region. I released in 29 languages so I did get some good numbers from a few foreign countries. I will be using this data to determine the top 5 or 6 countries it sold best in with languages other than English if I decide I want to localize games in the future that include text.
Sales Performance Review
So the best successful sale period was during the Steam Summer Sale. This was also right after my major visual update. The more colorful and exciting screenshots, I suspect, got more people to click on my game more frequently than the minimalistic black and white version as it isn't as attention grabbing.
Worst performance was during September/October. This was probably due to the fact that I wasn't working on any new updates or promoting the game at all during that time while I was focused on my day job as well as having to deal with computer issues. As far as discount events and updates performance goes, the Autumn Sale was the worst. This is probably because of Black Friday/Cyber Monday deals on bigger name titles that people were more interested in.
The Demo release however has (since I generated this data) brought more positive user reviews from people who did play the demo and then bought the game. Currently it is at a 90% positive rating out of 12 reviews. Technically only 8 are counted and displayed at the top of the game page because 4 people that did review it received the game for free as curators, youtubers, or testers. Steam still says the game needs more user reviews to generate a score so if you own the game, received it as a gift, or are going to buy the game please remember to leave it a review after playing it for a bit with your honest feedback. Good or bad, it all helps me know where to improve in the future and get a ranking with Steam.
I do continue to see wishlists for the game rising now that the demo is out, which is nice but not sure I can disclose those numbers so will keep that to myself for now. I can however share any data from steamDB.info including showing when the game has had an influx of new followers and averages/influxes of concurrent players/twitch viewers.
User Statistics To Date (December 3, 2020)
Maximum Daily Active Users, according to SteamDB.info, was in October with 4 players.
It should be noted that I've found that SteamDB.info and Steam data doesn't always match up so with what I can share from Steam I would assume that data is more accurate and current.
Daily Active Users Average (Lifetime) - 2
Maximum Daily Active Users (Launch Week) -11
Median Time Played - 8 Minutes
It should also be noted that these have varied over the months since release. Median Time Played over the first few months was between 15-20 minutes. The leaderboard statistics that can be viewed on the community page on Steam (HERE) will also show how many have played since the newer updates and you can even see how far they progressed before putting the game down by looking at the individual level scores as well as global achievements to see what percentage of players have achieved. This also applies to the demo version as well. That data ended up being greatly informative.
Global Leaderboard & Achievements Evaluation
Full Game Global Leaderboards can be viewed HERE
Demo Global Leaderboards can be viewed HERE
Currently if you look at the leaderboards there are a good amount of scores for the first few levels but then they gradually drop off as you go through each level. These were most likely players who bought the game before I had a demo to try and did not realize how difficult it can be. I'm hoping that now that I have a demo that it should help curb that player retention and narrow my audience to people who enjoy hard arcade style games and are determined in overcoming the challenges the game presents.
As of December 3, 2020 the Overall High Scores for the full game had 24 scores logged. 15 had overall high scores listed above 0 but 9 players had 0 for their scores meaning they bought/received, downloaded and booted the game but never played or never got past level 1 (post leaderboard update). I noticed similar patterns in the demo leaderboards as well with 47 overall scores but 26 had 0 as their score. As to why this is baffles me a bit. Never had reports of crashes or bugs that would prevent players starting the game. Perhaps the visual Icon language threw people off and didn't understand what the icons meant. Either that or they just could not figure out the controls/mechanics and beat level 1 so they quit after several retries. About 2/3 of players beat level 1 though and had a score for it in the full game which leads me to think overall difficulty played some role in it.
I did find a video on youtube of someone trying the demo who didn't realize you could fire bullets and took them several attempts to figure it out. Was very hard to understand them in the video so not sure if perhaps that was the issue for some players who had 0 scores. You can watch the video and see for yourself HERE. I thought it was very clear in the trailer what the gameplay included, especially firing bullets, so I don't see that being the issue but again, I think it has to do more with the games difficulty.
I noticed players scores started to drop to 0 on the level leaderboards after level 13 meaning they stopped playing the game and never beat it, at least after the leaderboard update. Every level or 2 after level 13 players with scores would continue drop by 1-2 players until it hit 6 and stayed at 6 through level 25. Around level 40 there were only 5 people on those level leaderboards and only 4 after level 46. So, technically, since the leaderboard update and from my current data, only 4 people beat the game or bothered to replay it to get scores. Those last 25 levels do get extremely difficult so perhaps I did make the game a little too hard or not clear enough on how hard it was in it's advertisement.
As for the Global Achievements statistics (View Here), it also reinforces the difficulty theory. 90% of players started the game but only 50% got past level 5 with the percentage dropping over every 5 levels to only 12% of players beating the game (from launch to present). 50% or more of those players that played through level 5 also achieved many of the other statistic based achievements so those were probably to easy to obtain and not enough consideration was taken into account on how many times a new player might have to replay levels which would increase their stats much quicker than expected like number of blocks broke and shots fired, each time they retried. There may have been too much reward too soon not giving enough incentive to continue playing for the rarer achievements, much less for beating the game if they have a high achievement completion rate that early on in the game.
After this last final update I did make some changes to the wording on the store page and updated the Steam tags to reflect and prioritize Probe as a "Difficult" game to complete. Actually, I redid all of the steam tags to make sure it accurately describes the game and it's features in the proper priority to ensure it is displayed/displays similar games appropriately in the "Other Games Like This" columns. Time will tell if that will help with new player retention or sales. If I see improvements in the near future I will post an update on the topic.
As it is though, would I call Probe a "Success"? First let's take a step back and review briefly what was and was not successful before coming to that conclusion.
What was successful?
Was Finished and Released - Though it took some time to put in the final touches via post-launch updates, I did finish and release the final product.
Fun and Challenging Gameplay - Though I may not have reached the right audience initially or reached enough players for the genre or difficulty, the gameplay was challenging and as difficult as I intended. It was the game I wanted and found fun to play.
Creating a Retro Arcade Aesthetic - Initial release was more of a design from the standpoint of the screen on an arcade cabinet. When I updated the new backgrounds it really felt like an old arcade game with just the play area being the screen and the backgrounds the cabinet art. Very happy with how they turned out.
Community Engagement - I thoroughly enjoyed reading, watching and engaging with players who did play my game and provide feedback. Even the negative feedback. If there were any issues a player had I responded promptly letting the player know I heard their concerns and worked to fix the problems ASAP. Soon as bugs were reported I got straight to work on the fixes and made the patch live as soon as it was done (sometimes within minutes of the bug report). Also took all suggestions into consideration and even implemented the vast majority of them in the game with updates.
Frequent Updates & Improvements - While the game that Probe is today should have been the state it was in for full release on launch, continually updating and improving the game elements seemed to help boost sales accompanying those updates and slowly build a small following. Also they let my player base know I'm taking the development of my game(s) very seriously and that I want to make it the most enjoyable experience possible for what it is.
Setting & Reaching Realistic Goals - I kept all updates and improvements to a reasonable scale to be able to finish them on a regular basis and within a reasonable timeframes.
Sound Design - While I did not make the sound effects and music, I did choose them, tweak them via pitch/tone and chose the soundtrack playlist. This really enhanced the overall production value and feel of the game. Plus the free soundtrack DLC was a huge hit in terms of downloads.
Scoring System & Leaderboards/Re-playability - Adding a high score leaderboard system added a higher re-playability factor than it originally had for players who did enjoy the game and it's challenging nature allowing players to compete with each other and show off how skilled they are or have become at the game. Plus it was a valuable tool in determining player retention.
Avoided Burnout - Taking occasional small breaks/time off during development, I've found, is pretty important. Even if it's to work on a different game. I took time off from working on the game and updates to do 3 game jams over the development of Probe since March 2020. Taking those steps back helped prevent any burnout and I was able to come back at the game feeling refreshed. I honestly believed this also helped increase my productivity when working on the updates.
Learned a Lot - While making a fun and challenging game that I wanted to play was one my main focuses and helped improve my overall game design skills, the other main focus was to learn what all steps need to be taken to develop and release a game on Steam. I especially improved my knowledge of Steam integrations like leaderboards and achievements. Also learned I need to switch to Unity 3D for future projects to ensure better quality control.
What was unsuccessful?
Financially a flop - I didn't expect Probe to succeed financially in the first place as it is rare that a developers first release is a big financial success, at least from the from sources I've reviewed. Thankfully I was not relying on the income from the sales to live off of and am doing this as a hobby for now.
Marketing and Promotions - I did little to no marketing before release outside of a month of a coming soon page which should have been up 3-6 months before release instead of just a month. I did have some community engagement after release but definitely needed more than what I did do if I want to get my game in front of the right people who would want to buy and play it. I also don't think I targeted my audience properly when doing promotional marketing.
Released Too Soon - I admit it was a rushed release but since it was I should have released it in early access instead of a full launch. That way I could have gotten the feedback I needed/received sooner to make the changes in the game you see today before releasing as a full finished product.
Player retention - The game was too difficult as the levels progressed for the average players that were stumbling upon my game. This could also be a failure in marketing and caused by the lack of a demo before this last update. I intended for it to be hard, so in a sense that was a success, just may have been TOO hard for most who did get it pre-demo or was not properly reaching the target audience that enjoys difficult games.
Majority of Achievements were TOO Easy to Obtain - With statistics like blocks broken, shots fired, etc. there was not enough consideration taken as to how many times a player would be retrying the first several levels, let alone the whole game, making a majority of the achievements easily obtainable early on. This gave less incentive to keep playing for level progression achievements since they already had a majority of all the other achievements.
Visual Icon Language and No Text - Some people may have had difficulty deciphering what some of the main menu icons were or perhaps it looked too much like an app ported to desktop and that may have turned them off from the game. Also limited the adding of certain features like custom keymapping.
Releasing in Every Language - Marketing and promoting in one language is hard enough. Not to mention Google Translate isn't exactly accurate. It took a long time to translate the store page in all languages but the data that did come from it helped me refine a localization list to the top 5 languages/countries the game was bought in that I should target and translate for first.
Using Clickteam Fusion 2.5 as the Game Engine - It has it's benefits and it has it's flaws. Unfortunately, at least for Probe, the cons outweighed the pros. There are a few issues that were impossible to fix because of the engines limitations but were luckily minor issues. The worst being the built in Physics engine. It has some issues that are hard to counteract since there is no hard code.
No GDD (Game Design Documentation) Prior to Development - If I had made and followed documents I would have been more prepared and able to set some things up much more properly to include future updates like leaderboards and saved data structures.
Demo Released as Last Update - This should have been on the "Coming Soon" page before release even if changes were made to the game later.
Not Enough Proper Feedback in Beta - Only using family and friends for feedback prior to launch meant I probably didn't hear what most of the players who were interested in my game might want out of it. Some of my friends are brutally honest though, and that's what I like about them, so that did help a bit here and there. After release though, I got some great outsider input that helped with post release updates and improvements. I do think overall sales could have been better from the get go if I had done some heavier Beta testing with way more outside sources first.
Releasing on Gamejolt.com and Itch.io - Was not worth the effort. Unless somehow it directed traffic to the Steam version, which I do not believe it did based on data. Overall game page traffic and sales on those sites were quite frankly...almost none.
Overview - The Takeaway
Overall...I did almost everything backwards and completely wrong when it comes to developing and releasing a game by far but I knew this from the moment I started the release process. I was given the opportunity to find out what it really means/takes to release an indie game solo on Steam and it was an opportunity I did not want to miss so I relentlessly pursued it not knowing how long I had to complete my goals. I've grown as a developer and learned a lot in the process. Now I feel I have a really good grasp on how to properly make and release other games on Steam in the future. In this sense, I feel the game was a success. I did what I set out to do...make and release a game on Steam and learn what to expect in the future. However, it definitely is not a success in terms of sustainable income/sales and, let's just say, Steam certainly wouldn't consider it a success.
In terms of the game design itself, it is as difficult as I intended, but also was not successful in player retention due to the difficulty curve or not reaching the proper target audience interested in difficult games. In the future I will probably add multiple difficulty levels to avoid this or make sure it is more clearly marketed as a really difficult game to complete if it is. I have also been improving my skills in Unity 3D over the past couple years so I will be switching to that engine for most all of my future projects, especially my next title, so I have more control over quality which I lacked in Fusion 2.5. I also will be posting development progress for my next upcoming title as soon as I feel it's in a good state to present.
Plans for the Future
Create detailed GDDs (Game Design Documents) for future game releases prior to development - This will help make sure I properly develop the game as intended and get information/demos out ASAP for feedback so I can alter things as needed before release
Try making different genres of games - I don't want to just stick to retro arcade style games. I have many ideas across many genres I would like to see made.
Switching to Unity 3D for development
Make and Release a New Game
Possibly a sequel for Probe - This won't be until after I release my next game which will be very different from Probe because I don't want to seem like a one trick pony or players to think of my games as shovelware and that I am only sticking the retro arcade genre. With what I did learn from Probe though, I do feel I can make an all around better experience to include a broader audience or range of players which makes me want to make a sequel just not right now. I want to try something different for a bit.
"Weekly Blog Updates" will become "Weekly Blog Posts" - I will still be posting weekly but will cover several topics or types of content for a while until I can reach the point of being able to start updating publicly my progression in the development of my next game. Posts will include things like games I've worked on in the past, game jams I participated in, little dives into me as a person or my history with games, my favorite games and genres, game reviews, development insight with tips and tricks I've learned, and possibly even some tutorials for getting started as a new developer with the knowledge I have learned and accrued from several other development sources. Don't forget to subscribe to the blog/newsletter to receive notifications when I make my weekly blog posts. Usually they are posted sometime between Sundays and Tuesdays each week but I will try to be better about doing them on a set day in the future, probably Sundays.
I am super excited to get going on my next project. Already got some stuff for it done and really looking forward to revealing it to the world and hopefully build a community around it of devs and gamers alike that want to be involved in its progress. I hope this Post Mortem review can be useful for other devs (new and veteran) or at the least was an interesting insight for my players and subscribers. Until next week, everyone stay safe, Happy Holidays, and thank you again for all your support!
Nate Bit Games