This is a cross post from CheeseTalks.net, where I publish long form articles like this one. The original version has interactive charts and better navigation, but the text has been posted here in full for anybody who would prefer to read it on IndieDB.
Hive Time is a bee themed management sim in which players build and maintain a sustainable bee hive. It's a niche game that aims to provide a sandbox experience in which players can explore and experiment role diversity with a generational community once they've mastered the sustainability requirement of producing enough Royal Jelly to spawn a new Queen. It's punctuated by humour and zany characters who from time to time upset the balances that players work to achieve.
This article dives into Hive Time's budget, costs, marketing, and revenue between the game's release in December 2019 and reaching a sales milestone in July 2020.
- Before we begin
- Budget: None, Costs: Some
- Finding a price point
- Overcoming launch hurdles
- Spreading the word
- Money for the honey
- Player platforms
- What worked and what didn't
- If wishes were horseflies
- Direct reflections
Before we begin
Before diving in, this article needs to be prefaced with a few important notes:
- I don't mind talking about money, but I absolutely despise fixating on it. This has been a difficult article to write.
- So as to stay on topic, I'll avoid diving into what worked/didn't work about the game itself (as opposed to its marketing, its identity, and its "performance") as much as possible.
- My overheads and living costs are significantly low and are not indicative of typical development costs.
- I'm very happy to have worked on Hive Time and am glad that it has been played and enjoyed by so many people.
- Every game is different, every audience is different, and the specifics of the context in which every game exists are different.
- Hive Time's budget and expenses involve multiple currencies. Where unspecified, assume Australian dollars.
- From one perspective, I did everything discussed in this article myself, but realistically, every player, every journalist, every youtuber, every streamer, every tester, and my four collaborators contributed to Hive Time's post-release identity. Without them, my game would have sat unknown, unloved, and (so far as what's relevant to this article goes) unbought.
Budget: None, Costs: Some
After completing the 10 day jam that Mim and I initially prototyped Hive Time for in May 2019, I took most of the rest of the month to weigh up continuing development against getting back onto other outstanding projects. Based on positive responses to the progress I'd been sharing and tester feedback, I decided to move ahead and plan out a zero dollar budget and rough six month timeline for development.
Initially, my budget was comprised of my own labour donated in-kind with the expectation of recouping my labour costs after release. Mim was undecided about contributing further art, but preferred to donate her contributions if that happened (it did). With this approach, I'm free from having to worry about money as a tangible development resource, which is generally how I prefer it when I'm working solo. It also means that if a project doesn't sell a single copy, I don't have any debts or other obligations that I'd need to invest more unpaid work into fulfilling.
For the sake of similar simplicity, I also considered things like plant and equipment, utilities costs, and exhibiting costs as out of scope for the budget since all of those were things that I already owned or could justify as being covered by other projects. I bought my primary workstation and travel laptop years ago, we would be paying for home power and internet regardless, and I have a big tub worth of spare mice and keyboards, cables, adapters, tablecloths, A4 table stands, etc. left over from exhibiting other games in the past. Since I was working from home, I didn't have office rent to worry about (but I did have to ignore my mortgage for a little while). These are all real overheads that should not be ignored, but are outside the scope of what I will be covering here - in this case, there's so much overlap with my personal living expenses and they're low enough that I could comfortably pay them out of Hive Time revenue if it makes enough that I can pay myself the equivalent of Australia's National Minimum Wage.
Me recording bees at R. Stephens Apiarists in Mole Creek, Tasmania (photo courtesy of Mim)
I use Free/Open Source software for all of my development tools, and though I donate and contribute upstream to some projects, I've chosen to not include that in the budget since that's something I'd be doing regardless. The tools I used for developing Hive Time were the Godot Engine, Blender, the GNU Image Manipulation Program, Inkscape, and Audacity.
I was hoping to submit Hive Time to the 2020 PAX Australia Indie Showcase, which I would have had to have allocated a travel budget for (exhibiting costs are typically covered for showcase winners), but since PAX Aus didn't happen this year and PAX Online's Indie Showcase was only available to games on Steam, that became one less expense to worry about.
I did spend about $175 on some business cards, but given that a global pandemic is not the best time to be handing things around to people or circulating in contexts where one would normally hand things out, they've sat unused. Since some very extenuating circumstances prevented me from using them, we can pretend for the purposes of this article that I didn't buy them.
Although I had planned to do everything myself and only rely on Mim for things she was interested in contributing along the way, I found myself bringing on board other collaborators who helped elevate the project in ways that I wouldn't have been able to on my own.
When Peter expressed interest in composing a soundtrack and demonstrated that he understood and had enthusiasm for the project, I decided to give him the opportunity. I wanted to be sure I could pay him adequately, so I dug into Patreon funds that I'd planned to use for Winter's Wake with the expectation that based on general enthusiasm for the project, I should be able to make that back as a minimum from Hive Time's revenue. As the soundtrack expanded and took shape, we decided to produce a human-voiced version of the Beatles homage "Beeing Together," which carried some extra costs for a vocalist and recording time.
Several days before release, I had resolved to cut a feature that adorned the Throne Room with portraits of passed Queens in order to keep my workload manageable. During an unrelated conversation, Aubrey offered to help out, painting the initial portrait that appeared in 1.0-x releases as well as the eight portraits that were added in the Informational Update. Aubrey has doggedly refused to let me pay him, but he has kindly given me an indicative rate to use in this article.
I also gave out around a hundred or so keys before release to people interested in giving feedback on/playing early builds. I've had to rely on volunteer testing for projects in the past, and had a solid core group of experienced testers who've enjoyed working with me in the past enough to want to contribute their invaluable attention. Ideally, I'd like to be in a position to do paid QA/testing on my games and give back to these people, but sadly, that wasn't an option for this project. As I didn't audit their work or demand a minimum amount of attention, I have no way of gauging even a rough figure for the amount of hours these people have invested in Hive Time (it could be as little as ten, or it could be hundreds or thousands), and so I'll be leaving QA out (aside from my own testing as part of development/bug fixing, but as with proof reading, you're much less likely to find the holes in your own work).
Prior to release, but after the bulk of development was completed, I was contacted by an arts funding agency that was getting into games. After some discussion and reading terms, I decided to decline the recoupable investment they were offering. Since I'd already spent all the money I was going to spend, and paying back that investment pro-rata would just erode my revenue until it was paid back, it didn't feel like a good fit. I could potentially have used that funding to produce more content, but I had a release date more or less locked in, and it didn't make sense to push things back just so that I could get a loan of money I didn't need. I also considered using it to cover localisation, but with some rough post-release development making the game's content a moving target, I didn't want to commit to localising into languages I don't speak until after the game has made enough money to support itself (otherwise the risk is that I'd just end up shipping the game in more languages that it could be unsuccessful in).
Initial release costs
Informational update costs
Finding a price point
There are a huge number of potentially relevant (and not so relevant) concerns to weigh up when it comes to putting a price tag on your work. These might include current, historic, or projected common price points in a game's market, cultural perceptions of what a price suggests about a work's qualities and desirability, what's needed for expected sales to cover existing costs, what's needed for expected sales to cover future work, or what resonates with a specific audience. Even what resonates with the work itself can be relevant. For example, a 90s themed hacking game with a price of $13.37 is probably going to stand out a tiny bit more than one that's $13.00 if all other factors are equal because the price itself becomes a talking point and a part of the game's identity. Price also doesn't need to be rigidly fixed, as shown by Alan Hazelden's A Good Snowman Is Hard To Build, which indexed its price to the daily temperature for a month or so after release.
In my mind, finding a good fit is very much a balancing act, where tweaking a price to favour some axis shifts what that means along some other axis. Dropping a price point to reflect market norms might in turn detract from the perceived value of the product, and will put it in the context and company of a whole different range of works at the time of release.
For Hive Time, I decided that my best approach would be to derive the game's price from a minimum amount of money I would like to make from the project and an estimated number of sales. This boils down to the following formula:
(hours * hourly rate + additional costs) / expected sales * (1 + estimated fee rate)
In this case, I spent 7 months working full time on Hive Time between the end of the initial jam in May and the game's release in Beecember (along the way, I took a month off for personal health and other projects). That amounts to 1050 hours at 35 hours per week. Rather than taking an industry wage (which would be somewhere around $30 - $50 per hour for a designer/programmer), I felt willing to invest most of my labour value and just take the Australian National Minimum Wage, which was $19.49 as at July 2019.
Additional costs that I needed to recover were limited to Peter's work on the soundtrack, which was invoiced at £2,097, coming out at $4,174.52 after currency conversion fees.
The most difficult part of the equation is how many sales I should expect. For many projects, research on the performance of similar games with similar release windows and launch contexts can start to give an indicator of what's reasonable, but many other factors can also play a big role. Since the pricing and availability models I've committed to gives me a fairly unique context, I couldn't find much in the way of relevant data points and instead relied on a bunch of best guesses based on my visibility in game and game-adjacent spaces, existing enthusiasm for the project, and observations of the attention that my past work (independent and contracted) has received. The number I ended up with as a reasonable first year target that could optimistically be a first quarter result or pessimistically be lifetime result was 20,000 players/downloads.
Hive Time's Pay-what-you-want (PWYW) pricing means that players doesn't necessarily mean payers, and so I needed to also come up with a reasonable conversion rate for full price equivalent (FPE) sales per download. I've studied and written a little about Humble Bundle's PWYW promotions, and while a lot of that research and analysis is old now (for better or worse, the industry and market definitely aren't in the same state they were 8 years ago), I felt like an estimate of 10% FPE wasn't unrealistic and still fell within the bounds of the kind of more contemporary discussion I'd seen around other PWYW projects in different spaces.
Last, but not least, my estimated combined rate for currency conversion fees, payment processor fees, platform fees, etc. was 20%. In reality, this fluctuates a lot, and my initial thoughts had underestimated the amount that payment processors gobble up, making for some fun surprises the first time I tried to collect a payout. 25% would have been a better fit, so in the interests of keeping things balancing out, I reduced the Itch storefront's revenue share by 5% - something I don't feel good about, and plan to undo later, but which represents the easiest knob to fiddle width in all of this.
So breaking that down, that becomes
(1050 * 19.49 + 4,174.52) / 2,000 * 1.20
Which gives me a magic number of $14.78, which when converted to US dollars at a rate of 0.68697 comes in at USD$10.16.
As a rough barometer, I also put that against my gut feeling of Hive Time being more than a USD$5 game and probably less than a USD$15 game in terms of the way I value games and experiences, and a USD$10 feels like a reasonable reflection of the amount of effort that went into creating it.
Overcoming launch hurdles
At 12:12 on the 12th of Beecember, I pushed Hive Time out into the world. Unfortunately, I didn't have time beforehand to do any press outreach. I managed to roll out a quick and dirty sing-along release date announcement trailer at the beggining of the month, which focuses primarily on construction and doesn't expose any of the game's cute or zany aspects. It also doesn't do anything to introduce the game's PWYW identity.
While it's silly to expect that discourse and coverage will perfectly align with your preferences as a developer, initial trailers, press releases and press/influencer outreach gives you an opportunity to make a first impression and highlight some of the things that you think are important about your project. I found that a lot of early coverage framed the game as being "free" or "free to play," and my impression based on the referring pages that lead to purchases is that this was more likely to lead players toward not paying for the game than coverage that mentioned that the game could be paid for and/or its default price.
As soon as I noticed this correlation, I added a FAQ item to the game's website and store page explaining why we had decided to commit to a pay-what-you-want model, and framing the game as a USD$10 game that's accessible to people who can't afford that. It's been difficult to measure the impact of that, but I feel that it was worthwhile regardless.
Worried that a few critical bugs and missing assets made for a bad experience and would set negative expectations, I held off on shipping builds from the end of the jam before the release version of the game was ready. As initial launch support started to settle down, I invested some time into forking the 10th day jam version and adding some quality of life (QOL) fixes and addressed missing assets. I wanted to make sure that these changes didn't reflect anything beyond the scope of what was achievable during the jam itself, so it still feels like a nice snapshot of early development. I really enjoy the way that the core of the game is present and identifiable - even though everything is rough, the game's direction still feels clearly readable. I had initially anticipated that this would get maybe 1% of the full game's downloads, and hoped it might generate some discussion of Hive Time's development history, but so far, it's only had 50 or so downloads.
I ended up finishing what I'd intended to be the game's launch trailer in February. It opens on a bee called Penelope, and focuses more on presenting the bee population as being made up of individuals while exposing some of the game's humour. Alongside this, I sent out emails, DMs and PMs to my existing press contacts, anybody who had previously covered Hive Time, and anybody I could find who seemed likely to cover management sims. I took care to pay attention to submission guidelines and tailor messages to individuals/publications. I also briefly explained that I'd been unable to do proper outreach at the game's launch, in the hopes of making coverage so far from release feel easier to justify. This generated the largest concentration of coverage that the game had from traditional press, but didn't generate much in the way of video content.
I also published an article and video discussing games that strongly influenced Hive Time's design. This went out a couple of days before the Penelope trailer. Like the jam prototype builds, this didn't seem to have significant impact on the game's visibility or sales, but it's something I'm very glad to have out there because it helps contextualise and express behind-the-scenes nuances of Hive Time's identity that wouldn't otherwise be visible.
After taking some time to work on other projects, I returned to Hive Time to work on a large update focusing on quality-of-life features, polish, and an in-game "Beepedia." I timed this to release on the 20th of May to coincide with World Bee Day. In an attempt to further address the perception of Hive Time as a "free" game, I added a screen that appears when quitting the game explaining its PWYW identity and inviting players to consider buying the game and/or its soundtrack. Players can indicate that they've already purchased the game and be shown a thank-you screen, indicate that they'd like to be reminded again later, or indicate that they'd rather not see the message again at all. The Informational Update resulted in similar views and downloads to the Penelope trailer, but a hugely disproportionate spike in purchases compared to views and downloads, suggesting that the majority of purchasers were existing Hive Time players as opposed to new players coming in.
Once the Informational Update shipped, I felt much more confident that Hive Time was capable of representing itself. With the Beepedia providing a way to review tutorials, get hints, and dive deep into game concepts. It's still the same game, but I think it's easier to identify for players who might be outside the game's core audience or might need a hand. Although I don't think I'd made up for the initial launch setbacks the game had had, I had at least partially recovered from them in a tangible way.
Spreading the word
In the lead up to release, I reached out to Itch to ask if they had any initiatives in place for working with developers to promote their work. I didn't expect that the storefront would have any sort of opportunities for paid placement, and I'm not sure if I'd have tried to take advantage of them if they existed, but the response I got back was advice to make sure that my store page had screenshots, video, and to make use of Itch's custom CSS features. I made sure to check these boxes, and though the store page isn't super exciting, I think (and hope) it's good enough.
At the beginning of March, Hive Time found its way onto the Itch front page banner, accompanied by a YouTube video from the official Itch account. My understanding is that this was prompted by its climb through Itch's various ranking metrics (top selling, popularity, reviews, etc.). Unfortunately, the game was flagged as both free-to-play and in-development, in spite of the game's store page and store metadata being configured otherwise (these were corrected once I drew attention to them). Front page placement resulted in a notable jump in views, and a small increase in downloads, but didn't meaningfully impact sales.
When Itch's Bundle for Racial Justice and Equality was announced, I submitted Hive Time. I didn't expect to get much visibility (it's difficult for an individual game that isn't already significantly prominent to stand out among 1,700+ items) and I figured that it would erode potential sales, but I wanted to help contribute to the bundle's perceived value and show solidarity. Unexpectedly, Hive Time's addition ended up getting specific coverage on Rock Paper Shotgun, and it's been showing up here and there in people's videos and reviews of games from the bundle. It also corresponded with an increase of maybe a dozen or so sales, though I have no data on whether that's bundle supporters who decided to pay for Hive Time or not.
Prior to and after release, I'd done my best to show up and engage with chat whenever somebody was streaming Hive Time. Sometimes that would mean setting an alarm for 4:00 AM or putting aside productive goals for an afternoon, but it has generally been a good morale boost for me, and seems to have typically resulted in at least a sale or two per stream. Typically if I come across a streamer who's expressed that they're interested in management sims or whose community I'm already a part of, I'll offer a key. With the game being PWYW, that becomes a bit of a token gesture, but that small effort seems to be appreciated.
Throughout the game's history, any time I'd seen people talking about the game on social media in a form suitable for sharing, I tried to engage or signal boost in some way. Typically, I'd avoid boosting if someone mentioned it in a conversation that others were tagged in, but I'd favourite and/or comment if appropriate. I had hoped that players would share hive screenshots, and prompted that a couple of times on the game's social media accounts, but that didn't seem to get any traction. I also made an effort to leave a thank-you message on every video I'd found of the game, offering some advice, trivia, or other commentary where relevant. These comments seem to have been well received, and while they don't seem to have resulted in sales that I can see, I feel that it's nice to acknowledge people who're signal boosting my work.
A piece from Albuffalo's "swarm of bee fanart"
My favourite pieces of social media interactions have come from people who wanted to share something that characterised their own experiences. Albuffalo did some wonderful fan art that I really appreciate, which shows attention to some of the game's smaller details. Another player shared a photo of their cat trying to catch their hive's bees that is absolutely adorable. The screenshots of proud hive builders who've built out the entire play space or who have drawn cool patterns with their hive's cells are treasure to me.
Releasing in the middle of December made it difficult for Hive Time to appear in many "game of the year" lists. Hive Time was nominated for several categories in GamingOnLinux's Linux GOTY Award 2019, but only made the top five for Best short game of 2019. Later in 2020, outside the period that this article focuses on, I submitted Hive Time to the Australian Game Developer Awards, but sadly, it wasn't selected as a finalist for any categories (which was disappointing as I expected Hive Time's amazing soundtrack to be a pretty safe bet for Best Music). Hive Time was also nominated to be Ask Gamedev's Community Member Game of the Month for October 2020, but didn't secure enough votes to win (it was featured in the video announcing the winner though). I had hoped to submit Hive Time to more awards for consideration, but I've not had much luck finding suitable ones (the ones I have have typically been delayed due to events that they're attached to being cancelled or postponed due to the current pandemic).
While I don't feel that I have enough solid data points to do any meaningful analysis, the interplay between different types of coverage and their impact on views, downloads, and sales has been interesting to observe. High profile coverage can put the game on a lot of smaller people's radars, but smaller coverage can sometimes make things visible to larger people - my impression is that there are people at both ends of the scale spectrum who're keeping track of what's getting attention outside of their own spaces in order to find new things.
In general, this feels healthy, though I came across a disturbing amount of plagiarism over the past ten months while looking for coverage. I made an effort to weed out plagiarists and aggregators as I went, so I haven't kept full records, but it appears that a lot of these have since been taken down. That said, it's still not hard to find Jody Macgregor's words about memories of playing SimAnt or Sin Vega describing joy at watching Hive Time's development in places where their names aren't mentioned and the original articles (neither of which are published under licences that permit redistribution) are not linked to.
On some levels, it doesn't feel as problematic to see for larger publications though - those organisations have more resources to tackle these issues and seem to have a little bit less to lose than smaller sites or individual bloggers. In particular, seeing non-English coverage lifted verbatim without attribution or licence distresses me the most (orcsbase's Chinese coverage and CryptoNick's Russian coverage in particular were both very difficult to identify as the original sources among a deluge of copies). Obviously attitudes and legalities surrounding these things varies from region to region, but there's still extra effort and skill involved in learning both another language and the cultural context required to appreciate a reference-filled game like Hive Time that I believe authors should be recognised and respected for.
With Hive Time not being localised to any languages beyond English, I've avoided reaching out to non-English press and influencers unless they've previously covered the game. Surprisingly, somewhere around a third of Hive Time's coverage has been in languages other than English. I've primarily had to rely on automatic translation to get the gist of what people have been saying about Hive Time in non-English languages, and broadly it's been positive recommendations. I feel like this reflects that although the game has a moderate amount of text content in vignettes and events, the management and construction gameplay that make up the core of the game largely leans on iconography to communicate elements and concepts, which makes these aspects of the game a little more accessible.
Distribution of Hive Time sale locations
Interestingly, this correlates with about a third of Hive Time's sales coming from countries where English isn't the primary language. It doesn't seem reasonable to assume that one third of the game's player base does not have English literacy, but it does feel like it hints at non-English coverage having the power to make the game more visible and more accessible to audiences in regions that aren't dominated by the English language.
Money for the honey
I've wanted to write this article since before Hive Time launched, but I felt like it was important to wait until I'd either crossed a revenue milestone that was more meaningful than just making back the money I'd spent on the soundtrack. It also felt important to make sure that enough time pass to feel like the game's launch performance was visible and in the past enough that the presence of this article itself wouldn't shift things dramatically (or if it did have a big impact, that that wouldn't undermine what's discussed here). After the press coverage that the Penelope trailer generated had died down, the game had sold about 350 copies, and I decided that 500 sales was a good point to sit back and retrospect.
Views per day over time, annotated with milestone details.
Downloads per day over time, annotated with milestone details.
Sales per day over time, annotated with milestone details.
In July, shortly before my birthday, Hive Time hit that 500 sale point (note that this does not include the Itch bundle's 815k owners since those represent additional owners rather than sales). At the time of writing in September, it's just shy of 550. Unfortunately, that only amounts to a 1.36% sales conversion rate rather than the 10% I was anticipating if the number of downloads is taken at face value. The game has been downloaded 40,018 times, which is double my initial projections, but an unknown number of users has pulled down an updated version of the game, either automatically via the Itch app, or manually after seeing an in-game update notice (if those are enabled).
It's enormously difficult to estimate, but my impression is that it's going to be a distinct minority of users that download updates. For the purposes of having a baseline to consider values in relation to, let's generously halve that 40,000 to a round 20,000, bumping our 1.36% up to 2.72% - a number still a long way short of the 10% I've been aiming for.
It might be easy to assume that the Itch bundle eroded that significantly, or that the global economic depression that is dominating 2020 is responsible. While these have both likely had a strong impact on how Hive Time has sold, numbers have remained close to that 1.36% figure since release, so it's difficult to point the finger at either or both for being responsible for that ratio.
In fact, I was able to leverage the economic situation to my advantage, drawing the first chunk of Hive Time's revenue during a period where the Australian dollar's value had dropped dramatically against the US dollar. This drop resulted in a difference of about $500, allowing me to reach a point where the game had made back the money I had spent on commissioning the soundtrack from Peter around two months sooner than I would have otherwise.
The average payment price has more or less hovered consistently around USD$9.15 after the first couple of weeks following launch, which I find heartening. That means that for every sale below FPE, I've also seen corresponding sales above FPE that comes close (within 8.5%) to making up the difference. Ignoring my own finances for a moment and that some of that might be motivated by a desire to support me, I like to think of this as people with more subsidising people with less in a way that suggests that somehow, there are people out there who'd rather equalise wealth distribution.
Looking at the rate at which things have slowed since launch, the Informational Update, and the Itch bundle (as the three major spikes in sales across the period being discussed), I'm not feeling particularly optimistic about reaching that 2,000 FPE sales target. A typical game sales curve will project between 50% and 60% of a game's income as arriving within the first 12 months. With Hive Time only 25% or so of the way there at 10 months in and the game's average sales per month trending downward in general, my revised estimate for a 5 year lifetime revenue would be more or less double what I've made so far, leaving me 50% behind.
In tangible terms, that means not only am I delaying paying the bills I put on hold while I was developing Hive Time, I have to wait 5 years to pay half of them. In spite of my efforts to minimise overheads, avoid debt, and significantly underpay myself, I've still been unable to approach anything near covering development costs.
There is one further wrinkle that's not immediately obvious, and is overlooked by the above. Between release and World Bee Day in March 2020, I did significant work on Hive Time for the Informational Update (at this point, I've also likely exceeded my initial allowances for bug fixes, support, etc., but since this update embodies the concept well enough, let's assume I haven't for the sake of simplicity). Thanks to the fluctuating exchange rates of multiple currencies being involved, and the game potentially continuing to make sales that it would have made without the Informational Update, it's hard to gain a true sense of how things have progressed. Within the 500 sale window we're looking at, 108 sales (105 FPE) were made after the update launched, representing both a bump in average sale price, but also the frequency of sales. The sales spike accompanying the Informational Update's launch is 20% higher than the game's peak at launch. The game only made back one third or so of the update's development costs within the period being looked at (which includes two months' worth of post-Informational Update sales), but looking at a slightly larger picture at the time of writing the boost prompted by the update looks like it will pay for itself early next year and extend the game's lifetime sales curve by a little, which longer term looks like a net gain.
Sadly, Itch doesn't expose any information on what platforms sales' downloads are attached to. For games with a minimum price, all downloads are made by paying players, but for a game with a $0 minimum, there's no distinction between people who have paid and people who haven't.
Pre-release downloads per platform
When Hive Time launched, I already had 7 months' worth of download data from testers and pre-release players, at which point, Linux accounted for over 70% of downloads, Windows a touch over 20%, and MacOS making up the difference. Even though I have a higher number of Linux users in my core community, this ratio is exaggerated by those Linux users being more likely to download updates and do thorough testing, and less an indication of actual platform representation. By a week after launch, Linux had dropped to a little over a quarter, and Windows was approaching two thirds.
v1.0 downloads per platform
For a while, Linux support remained above 25%, but by the time the Informational Update shipped, Windows had grown to three quarters, and Linux had dropped to under 10%. If Hive Time had a minimum price, I would have expected Linux to stay above 10% representation, but with the game more accessible to people who aren't necessarily going to be interested enough to buy it, the ceiling for Windows users is just much, much higher.
Interestingly, Mac downloads had grown to nearly 15% by this point. After finally having enough of Apple's nonsense in October 2019, I started advertising system requirements of "10.14 or lower," which I expected would likely alienate Mac users who'd moved up to 10.15.
v1.1 downloads per platform
For the period between the Informational Update shipping and the 500th sale being processed, Windows growth had started to slow, Linux had recovered to 10%, and MacOS had dropped below 9%. Assuming no dramatic events that put the game in front of large audiences biased towards any given platform, I expect these ratios to more or less hold moving forward.
These ratios likely look different from most other projects. My visibility as a Linux-based developer, Linux porter, sometime contributor to a popular Linux gaming news site, and a community organiser put me in a position where myself and my work are just more likely to be on Linux users' radars than many other developers. The goodwill I've generated by supporting both the projects I've directly worked on, and other developers' Linux-supporting projects means that my fellow Linux users are a little bit more likely to be inclined to buy the game to support me than they might be to buy another developer's game. As a niche game, Hive Time's not going to be every Linux user's cup of nectar, and I'm very confident that its sales and downloads don't reflect the full market size.
I'm still a little puzzled by the representation of Mac users. It's possible that my announcement that I would not be supporting MacOS 10.15 onwards generated some goodwill, but since I haven't had time to sit down and get PowerPC builds of any of my games together (turns out both of my PowerPC Macs are running Linux and I'm not sure where my OS discs are), I don't really feel like I've provided anything of value that I said I'd try to aim for in that announcement. It's more likely that this is instead just organic Mac traffic from users who haven't bothered to read the system requirements (the bulk of Mac support requests I've received coming from people running 10.15 supports this), but all of my recent projects have significantly lower Mac representation. Maybe The Away Team, Honeycomb CRUNCH, and Supply Chain are just specifically disinteresting to Mac users?
What worked and what didn't
It's easy to immediately view this as an entirely negative situation, but while Hive Time may not have made as many FPE sales as I'd hoped, it seems to have exceeded the number of players I had expected. In terms of building a community and an audience for future work, there are a lot of positive aspects to be considered. Still, it makes sense to look at and think about possible factors that might have hampered Hive Time's sales along the way.
There are so many factors involved that it's not possible to meaningfully find one or a small set of critical "failings." While I have thoughts and ideas on how to mitigate some of the pitfalls I've faced, I also think that some of them reflect room for improvement in the industry and the culture that surrounds it. I think back to my friend Daniel West's article Good isn't good enough - releasing an indie game in 2015 in which he talked about doing the full checklist of common shallow advice that makes up the dismissive, "You didn't do X? It's your own fault your game failed!" narratives that dominate a lot of failure discourse in the industry, and still not having guaranteed success (success in terms of making back development costs) for his award winning game. The vast majority of responses that I saw to the article and the discourse it generated were primarily weak arguments focusing on single aspects aimed at pushing blame back onto the developer instead of engaging with the article's core focus of trying to discuss the way that the dynamics of the industry and its markets are fundamentally volatile and difficult to effectively wrangle, and the way that people approach success and failure narratives isn't healthy. In the years since, the notion of an "indiepocalypse" has coalesced, and that definitely feels positive to me (from the perspective of acknowledging that some of the problems we face are industry level problems), but the discourse I'm exposed to still doesn't seem to give much weight or value to finding ways to maximise stability and safety for creators navigating these unpredictable waters, and often positions those who're left behind as being deserving for fairly flimsy reasons.
The thing that I most frequently regret surrounding Hive Time is dropping the ball on release marketing and press outreach. There were definitely obvious mistakes on my part that cost the game momentum and exposure, and prevented me from being able to inform/direct launch narratives. Since the game's had significantly more players than I had initially projected (and significantly more than I expected given that I did no press outreach at launch), it doesn't seem rational to fixate solely on that, though. In addition, Hive Time received traditional media coverage spanning print, traditional web media, TV, and radio that so many other games and developers just don't get - to the extent that I've had people congratulate me on my launch marketing even though I didn't do any. Hands down, I could have done better on this front, but I'm not convinced that it's been a critical aspect - stronger marketing would have helped offset Hive Time's poor performance, but it's probably not meaningfully responsible for that poor performance from just a raw numbers perspective.
The thing that I am most frequently proud of is that I've been able to commit to a pay-what-you-want pricing model with a zero minimum. I'm certain that it's very easy for a lot of people to say that this was a problem for the game, but within the scope of this article, that feels equivalent to, "The worst thing about a pay-what-you-want model is pay-what-you-want pricing." If every download resulted in a $1 payment, I'd be at more than double my target, but it is reasonable to assume that a distinct majority of players who haven't been interested in paying for the game just wouldn't play if the game had a minimum price. The high average purchase price means that a $5 minimum applied to current sales numbers would only result in an increase of about 35 FPE sales. Instead, I'd rather frame and embrace the combination of commitments I made for Hive Time as a constraint rather than a problem. I'm glad that people can still play my game without paying for it, I'm humbled that someone paid USD$82(!) for it, and I'm gratified that the average payment sits pretty close to the USD$10 that I consider the game to be worth.
That said, I feel like I've considerably underestimated the portion of a player base willing to share some level of support if it's optional. As discussed earlier, the 10% FPE player-to-payer conversion rate that I had based my projections on has been off by a huge amount. If I re-run my calculations based on the 2.72% FPE conversion rate we identified earlier, that would price the game at USD$54.35, which is beyond what I'd be reasonably able to or comfortable with selling the game for.
Obviously, just pumping up the price isn't the only or even the best lens through which to view this particular concern. The next most obvious approach to consider is to increase the player base to a point where 2.72% is capable of providing the target revenue. While 75,000+ players seems achievable, that starts to represent a user base that's pushing at the limits of what a solo developer can provide end user support to (not to mention the labour cost of that end user support) in a scalable manner. With the game being more than halfway there at just over 40,000 downloads, I already feel like I'm responsible for a potentially overwhelming number of players. I'd rather find that smaller audience of people willing to pay for the game than specifically appeal to a wider audience that isn't interested. In my mind, it makes more sense to see access to global audiences as being about allowing people with niche interests to come together rather than expecting every person on the planet to care a tiny bit about something.
The framing that makes the most sense to me is that I wasn't successful in encouraging the game's player base to feel like it was worth paying for or important to pay for. The Informational Update pushed along this axis to a degree, and the spike in sales it generated suggests room to expand here. The payment reminder screen prompts players to think about paying, but the game itself and how it's presented will strongly affect the value that players ascribe to the game. Positive feedback on the Informational Update's new features leads me to believe that the improvements it brought played a role in the higher payment average too.
And this leads to the downside of my "do as much as possible with as little as possible" development mantra - it's easy for areas of low polish to manifest and remain so long as they don't dramatically affect the game. On the upside, this leaves room for post-release updates to have a high impact on perceived quality without requiring large investments of design or production. The Queen animations added in v1.1-46 early in September feel like a good case study. I don't think I once got feedback or saw players talking about the Queen being static. It's something that's easy to not think or care about. The new animations draw some attention and give the impression of more content and greater polish, and coverage since then seems to be generating a slight increase in sales. Of course, even small amounts of work add up, and while there is definitely room to further refine the game, the overheads involved make it something that I feel I need to navigate thoughtfully.
The full set of commitments/constraints (pay-what-you-want pricing, being able to buy/play without requiring a storefront account, being able to buy/play without requiring a storefront client, being able to play without forced updates) that I've embraced significantly limit the storefronts I can sell the game on. Itch.io currently represents the best fit, but I'm open to considering other storefronts that offer services that meet these requirements, and at some point I would like to set up a direct purchase option. Steam, GOG, and Epic are all immediately ruled out. Of these, Steam is the one that I see requests for most frequently. Anticipating this, I tested to make sure that Hive Time would run through the Steam client when added as a "non-Steam game" as a way of allowing Hive Time players to have the experience of running the game through Steam if that's what they want.
Dominant discourse holds that by not being on Steam, I'm limiting my audience significantly. Since I've gotten more players than I projected, that doesn't feel like it needs to be a critical concern. My impression from paying attention to comments and observing sales is that most people who take the time to ask whether the game will be on Steam end up buying on Itch once they understand that it won't be. It's anecdotal, but my experiences as a developer who's contributed to and worked on games that are on Steam lead me to understand that there are no guarantees of attention or revenue there (like Hive Time, many of these have been niche games and face some of the same hurdles with regard to finding audiences). It's possible that today, with the visibility and momentum that Hive Time currently has, launching on Steam with a significant marketing push would lead to a revenue boost, but I don't think that the story would be the same if it had launched on Steam to begin with. Doing so now would also mean abandoning commitments I've already made, and that's not a path I'm comfortable treading.
Something that's become apparent to me after observing players' responses to the game is that the 1989 SimCity style of play-as-long-as-you're-having-fun and enjoy-exploring-the-different-types-of-structures-that-the-simulation-can-support type of experiences that Hive Time aims to provide isn't something that seems to come intuitively to contemporary audiences. I've seen reviews and discussions talking about the game having little or no replayability, but who haven't played past the tutorial. I've seen players talk about the game being too slow at the start and too stressful at the end, but with no realisation that those two things are related. I've seen players talk about there being only one right way to play, but I've put special effort into making sure that different styles of hives and population distributions can be viable.
Initially, I'd intended for all cell types to be unlocked from the outset. The current initial research gating and the tutorial were designed to help ease players into the game so that after their first Queen, they'd be equipped with enough knowledge to play for fun and explore the simulation on their own terms. In addition to being a way for players to optionally start new hives without feeling like they have to abandon their existing game, I also wanted the new Queen sustainability goal to be a comfortable stopping point so that players who aren't a part of the game's intended audience can still feel like they've had a "whole" experience. It seems though, that many players have gotten to the end of the tutorial or unlocked all research and feel like they have exhausted the game rather than recognising that that's the point at which the intended experience becomes accessible.
In itself, this isn't problematic - people can approach and enjoy the game in whatever way fits their tastes and sensibilities. If the game's core identity ends up defined by that though, that feels troubling. In addition to not being able to have much influence over the game's launch narratives, I feel like I might not have done enough to lead players toward considering the types of angles I intended the game to be approached from. These things were on my mind when I sat down to plan out a tentative roadmap for Hive Time's post-release updates. The Informational Update partly addresses some of the onboarding aspects of these identity problems, and assuming I'm able to make them, the Monumental Update and Goal Update will try to address leading players into continued play for its own sake and surfacing different play styles respectively. This is a difficult balance to strike though. Ideally, I'd like players to experiment and discover things, and to enjoy that process. Too much hand holding and direction takes away the opportunity for that to occur.
Similarly, there are also aspects that might make the game more attractive to its intended audience from a thematic perspective that aren't immediately obvious. At its core, Hive Time is about the relationship between representation and diversity and the value of individuals within generational communities. It also expresses some things about labour, infrastructure, and redundancy. These concepts are all expressed emergently through the game's systems rather than narratively though, and I'm not really comfortable with the idea of presenting those explicitly up front.
Even though I mentioned previously that I was able to take advantage of the current global economic crisis, I have to imagine that for a significant number of people out there, opportunities for income have been affected, and support benefits have been little and slow, making unnecessary expenses more difficult to justify. Here in Australia, official estimates of unemployment rates (which are believed to be understated) have tripled in the past year. I can completely empathise with people wanting to be more careful with the money they spend, and this context combined with Hive Time's price being optional, must make it easy to decide to not pay or put off for later when things are more stable. In another timeline, I might have felt more comfortable pushing harder and being more forceful in encouraging players to pay for the game, but right now that feels selfish and inappropriate (and while that might be something I'm projecting, ignoring those feelings would lead to an unhealthy relationship with my work).
The coverage that Hive Time has received has generally been positive. Even though I was able to generate some press coverage with the Penelope trailer, I can't help but feel that front loading that at release would have led to higher visibility and a higher launch peak. I don't have meaningful metrics for the impact of print, radio, or television coverage, but eyeballing my charts suggests that any impact on sales was small. I'm super thankful for the attention though, and it's nice to have artefacts outside of web spaces (as an aside, writing this has made me realise that since I'd already been reducing unnecessary circulation ahead of Australia's late pandemic response in March, I sadly never made it to a newsagency to pick up APC or PC PowerPlay when they would have been available - disappointing, but a small sacrifice in the grand scheme of things).
It's not quite regular coverage, but Itch front page placement (which began around the same time as Hive Time's first print coverage) also resulted in very few sales. The game received a significant boost in views, and a modest bump in downloads, but March included both the smallest number of sales and the longest streak of zero sale days across the period that the first 500 sales covers. March was fairly slow for coverage too, and with the exception of one video, the few pieces that appeared in March were from people I'd contacted prior to releasing the Penelope trailer. The best interpretation I can come away with is that front page placement is good at putting the game in front of people outside my audience. Again, I'm very thankful for the attention, and it was nice to see people within my community excited about seeing the game featured prominently.
Streamer coverage doesn't seem to have meaningfully driven sales. Being present, offering community participation and development anecdotes seems to have frequently resulted in one or two sales, which might be understandable given that I'd typically been featured by streamers with modest audiences (500 - 1,000 followers, which for scale, exceed Hive Time's sales for the period discussed), but it doesn't seem to have really varied when I've been featured by streamers with 1,000 - 5,000 followers, that doesn't seem to have resulted in more attention for the game. If anything, it feels like smaller audiences tend to be a little more focused. My impression from participating is that viewers are generally there for the streamer rather than the games they play, which makes sense - after all, they're the streamer's audience, not mine. The largest streamer I've seen playing Hive Time had about 18k followers, but their and their audience's primary focus is a specific game (not Hive Time), so again, it's understandable that that didn't result in significant uptake. Once again, I'm very thankful for the attention and have enjoyed watching and interacting with people who've chosen to share their experiences of playing Hive Time.
Another thing that watching streams of the game has highlighted is that I didn't take webcam or chat overlay placement into consideration. Activity and notification indicators appear in the bottom right and left corners respectively, the resource, population and cell count HUD bars take up most of the top and the top corners, and the population and goal HUDs extend down the left and right sides. Initially, I wanted the game to have no HUD and have players rely on indicators within the hive itself, but this didn't scale well - it's not so bad for the first 20 minutes of play, but once a hive's population gets above 50, it becomes pretty much impossible to intuit role diversity visually. I'm not sure whether a more streamer friendly interface would have helped the game be more attractive to streamers or their audiences, but I think that for anything that is streamer focused (which Hive Time isn't), that's a pretty important consideration.
Sales per web referrer
Unfortunately, the data I've gotten from Itch regarding referral information is missing 16 sales, but lumping the missing items in with those that contain no referrer information shows that the largest individual contributor to Hive Time's sales has been YouTube videos. I have no solid data on the ratio of these that have come in from the game's trailers or any of the archives of dev streams that I've uploaded vs coverage, but I'm inclined to assume that it's the latter. The top 11 coverage videos by view count for the period discussed currently each have more views than all three of Hive Time's trailers. Nookirum, Dr_Horse, SplatterCatGaming, and El Escoces provided the most viewed coverage. Nookrium in particular seems to have an affinity for management sims and basebuilders, and I suspect that Hive Time has likely resonated with his audiences more than the other prominent video creators.
The next highest known referrer for sales is Google. Adding all search engine referrals together along with Itch's own search page amounts to something just shy of YouTube's numbers. Having a strong contingent of players arriving at the game's store page via searches suggests the game has good word-of-mouth propagation and/or that it's memorable enough to want to hunt for after previously hearing about it (assuming that most people will follow a link to the game's website or store page in a piece of coverage rather than just open a search engine).
Rock Paper Shotgun is the highest referring piece of conventional web press coverage, with all but two RPS-referred sales coming in via Sin Vega's coverage of the Informational Update. Sin had approached me before the game released to ask for press access to the game after following its development for some time. Unfortunately, the timing of the game's release aligning with some overwhelmingly distressing world events meant that launch coverage didn't happen. I can empathise with and relate to feeling overwhelmed by events outside one's control, and to the way that that can sap away all motivation. To whatever extent it offsets that unpleasantness, it's comforting to know that Sin's writing eventually ended up being one of the most significant pieces of coverage that Hive Time received.
While they might not have been referrals for immediate sales (if someone downloads the game and then comes back to buy it later, the original referring site won't be represented), I frequently see Tom Sykes' PC Gamer article, Liam Dawe's GamingOnLinux coverage, Emily Rose's Re:Bind article, and Paolo Pedercini's ranking of bee games on the Molleindustria blog show up in the store page referrers list. Unfortunately, this only shows 30 days' worth of history, so I don't have solid data to look at, but I feel like I've seen every piece of coverage that I'm aware of bar YouTube videos with very small numbers of views make an appearance at some point or another.
It feels like the game's best coverage and recommendations have come from people with whom the game itself resonates. Many of these people stumbled across the game on their own or via third parties that I don't know, but some of them were people that I reached out to directly. I think that for any future press outreach, I'll be putting extra effort into finding people whose interests and tastes cross Hive Time's domain. I'll still endeavour to cast a wide net, but giving extra care and attention to the people most likely to have care and attention feels like the right investment.
If wishes were horseflies
If Hive Time had exceeded the revenue target and the possibility of having some obligation-free "disposable income" was on the table, there are a few things that I'd have liked to have considered which would have likely influenced Hive Time's performance.
I mentioned earlier that I wasn't comfortable pursuing grant funding for localisation before release. There are three main hurdles, representing a mix of risk and redundant work. Hive Time grew out of a jam project, and as such, all of the game's initial design decisions were made without localisation in mind, and with the key development priority being to make the game as efficiently as possible, every decision on top of that built upon the presumption that localisation wasn't a specific goal. Because of the way that strings are loaded and generated, Godot's built-in localisation support doesn't really offer meaningful functionality in this case, and I'm left with needing to write a solution for exporting strings in a localisaton-friendly format, and then substituting with localised strings at runtime. It's not an insurmountable problem or a prohibitive amount of work, but it's still requires a bunch of effort on top of the cost of getting translations done. The risk of not making those costs combined back would be a lot easier to embrace if I'd already made back the game's initial development overheads.
The second hurdle is that I shipped Hive Time with an ideal minimum of content, and have continued to tweak and add content over time. The Informational Update is the biggest concentration, but if things go well enough, I'd like to do two further updates. My approach to QA, post-release support, and post-release content development leaves the text within the game in a state of fluctuation. This would mean that any patch with a content change would need a localisation pass. Often this would be small, but it represents an amount of wasted effort. On top of that, relying on third party localisation requires a content lock phase between sending out content to be localised and shipping that content, otherwise different localisations will end up out of sync, risking unfixed bugs sneaking into releases and typically making non-English versions the least-good. To do things properly, I'd need to re-submit for localisation, ultimately wasting everybody's time and incurring extra admin overhead for keeping track of which parts of which localisations are up to date. At a larger scale, the patches and larger updates themselves represent the same kind of problem. As a small example, The Informational Update changed the titles of a couple of tutorials and events. If those had already been localised, then the initial localisation work would be made redundant, increasing the cost. This is completely normal and part of the commitment/investment made when embracing localisation, but again, that's more difficult to justify when the game is still effectively "in debt."
The third hurdle is the risk of not selling enough copies to cover each localisation. Once again, that's easier to get on board with when there's surplus funds to invest as opposed to piling that cost on top of money that hasn't yet been made back. Having additional revenue makes these kinds of improvements less risky (both in that the severity of the risk is reduced, but also in that by a larger paying audience in the native language domain providing more points of propagation into localised domains, there's a higher chance of making further sales and making back the investment) and it helps provide a way of brute-forcing through the chicken-and-egg problem of "spending money to make money."
At the time Hive Time launched, Australia was undergoing an unprecedentedly extreme, but foreseeable bushfire season. Australia's long history and culture of volunteer emergency support is nuanced and fascinating, but the main thrust is that part of our national identity revolves around looking after each other, pitching in, and having shared responsibility for the welfare of all (it's just a shame that that doesn't extend outside of emergency situations as often as perhaps it should).
For this season in particular, our volunteer and employed emergency services were severely under resourced and undersupported. Since release, I had wanted to donate a couple months' worth of sales to support affected communities during and in the wake of this disaster. I felt that I needed to make sure that I'd made back enough to cover the tangible money I'd spent on the soundtrack first, and that didn't happen until the end of February, which was the start of a significantly low point in Hive Time revenue (this period included the game's three longest zero sale streaks). I figured that if I'd only be looking at $50 to $150 by the time fees, taxes, etc. were accounted for, it'd make more sense to wait until the Informational Update shipped, since that would also give me time to make some thematically relevant content (smoke and fire related events are still on my todo list) and hopefully drive more sales.
By the time May 20th rolled around, the world had proven itself unable to prevent the onset of a global pandemic, and the ensuing economic downturn made me feel like I needed to be as careful as possible with the resources I had. In retrospect, I wish I had've just donated March and April revenue and forced myself to be content with a smaller contribution.
If Hive Time had've dramatically exceeded my target revenue threshold, I would have liked to have investigated sponsoring bee related content such as Extra Credits' Extra History series or Kurzgesagt. These would likely represent significant expenditures, but the idea of "giving back" by helping increase opportunities for discovering how fascinating our most important pollinators are and what their relationship with humans has looked like over the past 35,000+ years feels like a worthwhile use of a surplus.
I mentioned earlier that I already contribute and donate usptream to some projects, but if I'd surpassed my revenue target, I would have been very happy to share some of that surplus upstream by giving additional support to the Godot engine, Blender, the GNU Image Manipulation Program, Inkscape, and Audacity. Outside of Hive Time's development, these are culturally valuable tools that represent an incompresensibly large amount of labour, much of which is the work of volunteers rather than sponsored developers. It feels like they face similar hurdles to those discussed here, particularly in terms of encouraging users to value that labour enough to generate enough support to approach sustainability.
Reaching target revenue would also make it easier for me to invest more time in Hive Time's updates. I had to drop three planned items from the Informational Update in order to ship it on time, and having resources to hire a research assistant to help with Beepedia facts would probably have given me the space to cover those items and include more content. Financial realities have driven a lot of my decision making processes since Hive Time's release and are part of the reason that I haven't started production on the Monumental Update as yet - disappointingly, pursuing contract work and giving attention to other projects that are more likely to make money in the short term feels like the best application of my time right now, and if I do end up shipping the Monumental Update, it will likely be the minimal version of what it could be because of this.
In addition to making it easier to work on Hive Time, meeting target revenue would also put me in a better position to approach future/other projects. In particular, I find it difficult to justify giving time to things that have no ability to generate revenue like my in-progress Another World research project, Hidden Star, doing Linux ports of my old Half-Life 1 mods, working on a PowerPC Mac port of Icicle, and so on. Those projects are still very important to me, but starving to death while working on them won't help get them done. There's also a lot of prototyping for future projects, particularly VR work that I've struggled to be able to justify working on even though they're hugely exciting for me.
If Hive Time earned more than my target revenue, I'd also like to look towards engaging paid testing/QA for future updates. Ideally, this would be focused on providing opportunities to people who donated testing time to the game before release. Not only would this go some way towards thanking those people for their time and effort, but it would also benefit the game. Though I think I've nailed most outstanding bugs on the head by now, focused QA and rigorous testing would have helped identify and resolve those much sooner, and future bugs (of which there are always ideally none but in practice some) would be less likely to slip into public releases.
Last, but not least, if Hive Time managed to reach its target revenue, it would be nice to have a small amount of money for myself. The second hand jeans I'm wearing now had holes in them before I inherited them last year, and it'd be nice to replace them since they have now almost completed their circuitous journey to jorts and I can't really wear them outside without flashing my neighbours. Across Hive Time's development, I've avoided a lot of unnecessary expenses, and I have a list of about 200 games I'd like to buy in order to support other developers. The second hand desk that I bought in 1996 has served me long and well, but its structural integrity is on the decline and I've been hoping to replace it for some time. I'd love to get a new laptop, maybe buy one of those fancy mirrorless full frame CMOS cameras so that I can do more wildlife photography without disturbing animals, and while we're at it, a house that's big enough that I don't have to keep my lawn mower in my linen closet would be nice too. I'm OK with being minimalist though - I'd settle for less jorty jeans and being able to pay my bills.
So, I suppose at this point it makes sense to ask myself some direct questions. I'm really proud of Hive Time, and what I was able to accomplish both with its initial development, and the effort I put in after release. I'm glad I made it and I think the experience has been worthwhile, but it's clear that as a whole, it hasn't been "successful" (whatever that means) to the extent that it's covered its own overheads.
Do I feel that Hive Time could have done better with the same constraints and commitments? Yes, I think so. A better managed launch would likely have made a dramatic difference.
Would I embrace the same constraints and commitments for Hive Time if I could do things over? I think so. I'm very happy to have been able to explore committing to a user-friendly purchasing experience and a pay-what-you-want monetisation model. I feel like I've learned a lot, and it's been rewarding to put a lot of my past analyses and interpretations of pay-what-you-want models to the test.
Would I embrace the same constraints and commitments for future projects? Probably not. While writing this article, I've undergone some difficult life changes that put even more financial pressure on my life. When I decided to make Hive Time pay-what-you-want, the hope was that it would make its funding target back, but the primary goal was to make back those tangible soundtrack costs. My situation was such that in a worst case scenario I could absorb not being able to pay myself for the time I spent making Hive Time without that preventing me from being able to make games in the future. Today, that's not the case, and I think I'd need to pursue minimum pricing and begrudgingly leverage less user-friendly storefronts in order to push revenue higher and mitigate risk that I was previously in a position to embrace.
Regardless of how Hive Time has gone, the experience has equipped me with knowledge and experience that I'll be able to bring to bear on future projects. I'm still a proponent of PWYW models and the inclusivity that that can offer, and I hope that my experiences as I've shared them help others navigate paths similar to the one I've walked here.
A note from Cheese
Thanks for reading!
This article represents the state of Hive Time leading up to and including its 500th sale, with some extra anecdotal perspectives gained during the following three months. I'll continue doing what I'm able to grow Hive Time's audience, and what's written here may not reflect where things go in the future. If there are significant changes, I'll endeavour to either update this article or write a new one.
 To be clear, I strongly believe that cultural growth is stunted when culture can't freely own and iterate on its own artefacts. I am very much in favour of dramatic copyright reform and a transition to a culture more focused on supporting people who make stuff, but we're not there yet, and in the meantime it's important to respect creators' copyright.
 Sadly, I don't have a good reference for the type of "rule of thumb" curve that I'm using here. These sorts of tools are not really about predicting the future so much as they are about giving an idea of a common median path through sales over time that can be used to create a set of baseline expectations that real values can be compared to over time to gauge how "good" or "bad" things are going.
 May includes an anomalous upward spike coinciding with the Informational Update, and though the average following that has been raised, there is still consistent downward movement across the rest of the period covered.
 Incidentally, Daniel's first month of Airscape sales were within the ballpark of Hive Time's, and while the two games' contexts and hurdles aren't necessarily the same, they do share the reality of not being able to financially support their creators with sales.
 Australian firefighting history is well beyond the scope of this article, but here's a different article (the text, not the video) that gives a bit of an overview from a perspective relevant to the 2019-2020 bushfire season.
 A couple of planned new events, an expanded "Yam party" event chain, and main menu tooltips were cut from the initial v1.1 release. Most of those items were added in subsequent v1.1 patches, but at the time of writing, "Yam party" remains one event long, and there's a "Wax audit" event that still isn't in the game.
This article was originally published on the 24th of October on CheeseTalks.net