No AI was used in the preparation or writing of this post.

So there is a lot of discussion about playercounts in the comments here, and discussions elsewhere about this game and these discussions almost never put the playercounts we see on SteamDB into their proper context.

As the rest of this post will attempt to illustrate, even with our current (fairly disheartening) playercounts, there is likely still enough ongoing revenue coming in to support the operating costs of a fairly large (by industry standards, not necessarily by Bungie standards) development team.

This is not a short post, but it is not as long as it first appears due to the use of a code block and a table. I've also put some of the less important bits behind spoilers to make skimming past them easier. The majority of this post's content is just providing the calculations and estimation methodology. If you don't care about any of that you can skip to the 'Conclusion' section at the bottom.

>! I do not attempt to make a case that the game could become profitable both because we do not have sufficient information about its actual total development and marketing costs prior to release, and because that's not how the decision about whether or when to cease the live development of the game will be made. That decision will be made primarily based on operating revenue compared to operating costs and the opportunity costs arising from tasking those resources to some other project. !<

Monthly Operating Cost Estimate

Assume a nominal salary of $115,000 for Bungie employees, then multiply by 1.6 to estimate^(1) nominal salary + employer costs (healthcare, retirement, FICA, etc.) for a yearly cost per employee of $184,000. Then divide that by 12 for the monthly cost per employee (commonly referred to as 'burn rate') which works out to $15,333. Then multiply that by the team size (commonly assumed to be ~400 though no one actually knows), for a monthly team cost of ~$6.1 million. Working from the base $184,000 per employee cost for the year a team of 400 would cost $73.6 million per year.

>!Now this excludes things like office space rent, computer equipment, software licensing, etc., however these are going to be a small minority of the studio's total costs relative to the labor expenses (costs of employees). This math is also assuming every single employee is a developer being paid at that salary rate when in reality that won't be the case with some of the employees being in relatively low paid positions like secretaries, or other less well paid positions than that of a software developer, as well as some employees being in more highly paid positions like directors and other forms of leadership, we can ignore that in this case and just assume that the average of those other positions more or less nets out.!<

---

^(1)>!While no doubt different, I do work in a capacity where I see the direct payroll expenses for my (very large) employer and a 1.6 multiple for factoring in employer costs relative to nominal salaries is the current ratio. It was more like 1.36x as recently as 2020 but US Health Insurance costs have grown very rapidly, I would not be surprised if it grows to 1.75x or 1.8x by the end of the decade.!<

Per Player Spend & Monthly Revenue

The exact portion of players that buy microtransactions varies quite a bit from game to game so we'll instead work with 'average per player monthly spend.' While hard figures on this are virtually completely unavailable there are a lot of grounded estimates out there. I'm going to assume an average monthly per player spend of $7 including battlepasses and skins^(2) but ignoring up front price of the game, factoring in the platform cut of 30% that works out to about $5 of actual realized revenue per player on average (most players spend nothing, some players spend something, a few players spend A LOT).

In order to estimate monthly revenue using this figure we need an estimate for the playerbase size against which to multiply the estimated average monthly per player spend. Now it is important to note this is Monthly Active Players (MAU) not Peak Concurrent Players (SteamDB counts), Daily Active Players, or even Weekly Active Players.

---

^(2)>!This is the part of this analysis I feel the least confident about, so if anyone has some sources that would allow me to refine this estimate (either up or down), that would be much appreciated. !<

Extrapolating DAU, WAU, and MAU

I'll be using the below calculations to estimate the specific values. The resulting estimates are highly sensitive to the provided input (given) values. I encourage you to mock up these calculations on a spreadsheet yourself and adjust the inputs to see just how sensitive, especially since most of these inputs require me to make reasonable assumptions about their true values.

    _Given Values_
[Daily Average CCU] = ?
[Average Hours Played per Day] = ?                    per Player
[Average Days Played per Week] = ?                    per Player
[Average Weeks Played per Month] = ?                  per Player        
[Percent of Active Users on Consoles] = ?             as a decimal value; 70% = 0.7
[Average Monthly Realized Revenue per Player] = ?
    ---

[Total Hours Played per Day] = [Daily Average CCU] * 24

    _PC Estimates_
PC_DAU = [Total Hours Played] / [Average Hours Played per Day]
PC_WAU = (DAU * 7) / [Average Days Played per Week]
PC_MAU = (WAU * 4) / [Average Weeks Played per Month]

    _Total Estimates_
DAU = PC_DAU / (1 - [Percent of Active Users on Consoles])
WAU = PC_WAU / (1 - [Percent of Active Users on Consoles])
MAU = PC_MAU / (1 - [Percent of Active Users on Consoles])

    _Revenue Estimate_
Monthly Revenue = MAU * [Average Monthly Realized Revenue per Player]

Using the only hard data we have of ~4,000 per day (eyeballing the chart for the past week) for [Daily Average CCU], and the above established estimate of $5 for [Average Monthly Realized Revenue per Player] and using several sets of what I believe to be reasonable assumptions for the other inputs we arrive at the below.

Value	Set 1	Set 2	Set 3	Set 4	Set 5
[Daily Average CCU]	4,000	4,000	4,000	4,000	15,500
[Average Hours Played per Day]	1.5	1.5	1.5	1.75	1.75
[Average Days Played per Week]	2	2.5	3	2.5	2.5
[Average Weeks Played per Month]	3	3	3	3	3
[Percent of Active Users on Consoles]	0.3	0.3	0.3	0.35	0.35
[Total Hours Played per Day]	96,000	96,000	96,000	96,000	372,000
PC_DAU	64,000	64,000	64,000	54,857	212,571
PC_WAU	224,000	179,200	149,333	153,600	595,199
PC_MAU	298,667	238,933	199,111	204,800	793,599
DAU	91,429	91,429	91,429	84,395	327,032
WAU	320,000	256,000	213,333	236,308	915,691
MAU	426,667	341,333	284,444	315,077	1,220,922
Monthly Revenue	$2,133,336	$1,706,664	$1,422,221	$1,575,385	$6,104,608

Sets 1-3 vary only the number of days played per week. Set 4 is my own guess for what the actual given values look like, and Set 5 is a hypothethical case identical to Set 4 but adjusting [Daily Average CCU] to reflect the minimum value necessary for revenue to exceed the estimated monthly cost of $6.1 million.

Conclusion

Based on this analysis it would appear that a sustained 4x - 5x increase is needed to cover the ongoing costs associated with the game. The last time we were in that range was approximatley the week of Sunday 4/12 - Saturday 4/18.

The current playerbase is only sufficient to support a development team size of ~100. While this is only a fraction of the current (presumed) team size, it is still a very significant number of developers, and at most studios, more than enough to support, or indeed even develop a game of Marathon's scope and size. ID for example stated recently that they developed Doom 2016 with a team size of about 50 people, though I'm sure some amount of outsourcing was used as well which has its own costs.

Add some details