Leaked documents relating to Microsoft’s business strategy for Xbox show the company eyeing XR technology but continuing to keep it at arm’s length.

While Microsoft has previously taken considerable steps into XR with both HoloLens and the Windows Mixed Reality platform on PC, the company’s flagship gaming division, Xbox, has notably not joined the fray.

Over the years Xbox leadership has repeatedly pushed back on XR interest, saying the tech doesn’t yet have a large enough audience to warrant investment. And while it doesn’t look like we should expect anything relating to XR from Xbox in the near future, the company is at least continuing to eye the tech as a potential opportunity.

Road to VR reviewed the entirety of a trove of documents that leaked this week in connection with an ongoing Federal Trade Commission v. Microsoft court case. The documents, which reveal a significant portion of Microsoft’s long-term plans for the Xbox brand, show the company is still skeptical of XR but not discounting it in the long run.

In a mid-2022 ‘Gaming Strategy Review’ document, Xbox pointed to “AR / VR” as one of a handful of “opportunities” the company was mulling as part of its “early thoughts on [the] next generation of gaming.” In the same section the company pointed to tech like cloud gaming and ML & AI as potential areas of strategic focus.

In another section of the same document the company highlighted Windows Mixed Reality, OpenXR, WebVR, and HoloLens among many platforms and services that Xbox can leverage to build its “next gen platform for immersive apps and games.” Given the context of the document, however, it doesn’t seem that Xbox is specifically referring to XR when using the word “immersive.”

While Xbox has mentioned XR as a future opportunity, the company’s tone is still significantly skeptical that the tech has achieved a meaningful addressable audience.

In another section of the same document which overviewed Xbox’s competitors, the company pointed to Meta’s billions of dollars of investments into XR, but concluded by saying, “we view virtual reality as a niche gaming experience at this time.”

SEE ALSO
Meta Quietly Rolls Out 'Horizon Worlds' Premium Digital Currency in US, UK and Canada

Another document from mid-2022, which overviewed the company’s long-term plans for Xbox all the way through 2030, noted that Microsoft wanted to expand its hardware portfolios to include new hardware categories, but nothing on that long-term roadmap pointed to any XR hardware.

While the leaked documents did focus on long timelines, business is always dynamic and priorities can shift quickly, so it’s important to remember that the documents are just a snapshot of Xbox’s view in mid-2022. With the more recent introduction of devices like Apple Vision Pro, it’s likely that Xbox is looking even more closely at how important XR may be to its future portfolio.

Newsletter graphic

This article may contain affiliate links. If you click an affiliate link and buy a product we may receive a small commission which helps support the publication. More information.


Ben Lang
Ben is the world's most senior professional analyst solely dedicated to the XR industry, having founded Road to VR in 2011—a year before the Oculus Kickstarter sparked a resurgence that led to the modern XR landscape. He has authored more than 3,000 articles chronicling the evolution of the XR industry over more than a decade. With that unique perspective, Ben has been consistently recognized as one of the most influential voices in XR, giving keynotes and joining panel and podcast discussions at key industry events. He is a self-described "journalist and analyst, not evangelist."
  • Arno van Wingerde

    I can fully understand their interest and their hesitation. Yes: VR is nice, but the amount of VR players and VR games sold may not justify an entry at this point. Having the PS5 in the market would split an already niche market into two, so it may simply not be worthwhile to get in now, let’s see first how PSVR2 is doing. At the same time though, they should be ready to join when it looks that this is successful and by then PS5 is the console to be for VR. here’s hoping that Apple will be able to unlock the VR market at last in a few years with a non-pro $1000 device… because Meta’s strategy seems to be throwing billions at hardware development and forget the software, particularly AAA games other than for a Metaverse without Vision (sorry, couldn’t help myself)… curious as to what happens next week: “here is the new device, somewhat lighter/faster/higher res now go and buy it” or big games, streaming service etc.

    • Nevets

      I get your point. As for Quest 3, GTA San Andreas alone may be a system seller to some degree if they finally make good on it, but here’s hoping.

    • silvaring

      All those billions and we still don’t even have a simple VR headset that lets you lie down on your couch and watch a movie comfortably…

      pretty surreal

      • Andrew Jakobs

        Yes we have, its called the Vive Flow…..

        • ViRGiN

          shut your mouth

    • Christian Schildwaechter

      TL;DR: there is no need to rush for Microsoft due to the rather slow growth of VR, and any disadvantage they might have from a later start could be compensated through Xbox VR releases of the massively popular gaming franchises they have bought over the last few years.

      By now nobody expects XR to suddenly become a huge success, growth will be somewhat slow, but hopefully steady.

      – PSVR 2 success will depend on how many AAA developers Sony can persuade to go hybrid and include VR modes. GT7, RE8, and the RE4 remake have now validated the concept, but seeing the fruits of that labour will take years due to the long AAA development cycles.

      – The Quest 3 will be more of an enthusiasts device sold at a 67% premium compared to the Quest 2 still sold in parallel. The rumored 2024 “cheap” Meta HMD for USD 199 will probably keep the performance levels of the Quest 2, and raise the Quest 3 premium to 150%, unless they drop the drop the price. It is unlikely that this will cause sales explode, as according to Meta, the important “instant buy” threshold is USD 300. Going down further has less impact, they sold more 64GB Oculus Go at USD 249 than 32GB Go at USD 199. They are basically setting the Quest 2 performance level as a baseline, which will drive consumer expectation for VR. More impactful improvements like eye tracking or significantly higher resolution/FoV that would enable new types of application and could attract new user groups are postponed to the Quest 4, at least for the basic consumer HMDs.

      – Apple will only be win a very tiny share of the market with the first AVP, both due to the USD 3500 price and display production limitations that make it impossible to manufacture more than 500K in 2024. The AVP is technically a head worn MacBook using tons of new, exotic and expensive tech, so prices won’t go to consumer levels anytime soon. And even if the second version sells at half the price, it will still be rather/too expensive for a consumer device. People may be willing to pay USD 1000 for a phone, but today we carry and use these almost 24/7, while the AVP has much more limited use cases.

      – Valve doesn’t see any need to keep up with Meta’s release cycle, or compete in the sub USD 300 market, so even a Deckard release wouldn’t send user numbers through the roof, as it will most likely again target the limited enthusiasts market similar to the Index.

      – Everybody postponed true AR glasses projects for years, and these were expected to really drive user numbers up. Meta currently planing for 2027, but with them not being able to create more than 1000 internal prototypes for 2024 due to technical limits and costs that make the AVP look very cheap, 2027 is probably still very optimistic.

      There really is no need to hastily throw products onto the market for a good pole position, as most potential customers will still wait for several XR device generations before hopping onboard. Today’s market leaders only reach a minuscule audience. 20mn Quest 2 is 0.3% of the current smartphone market, so lots of room for latecomers. The Pico 4 is technically very similar to the Quest 2, and thanks to pretty much everything being OpenXR compliant, lots of developers ported their games to the Pico 4 with very little effort. So even with apps being the first doesn’t provide a unique advantage. New entries like the Bigscreen Beyond seamlessly integrate into SteamVR.

      So if the PSVR 2 becomes very successful thanks to a future large catalogue of VR capable AAA games, and Meta keeps growing the consumer market with the quality of games getting better and better (which they already do due to devs learning how to design and optimize for the Quest 2), Microsoft can still release a HMD years later, and instantly benefit from the then current state of hardware and software.

      And it will be much cheaper for them then, because a larger number of manufacturers have already established the basic technology and driven down the prices through mass production. They probably won’t have much trouble getting devs to support them, who today already try to design for cross platform to increase sales numbers by using open standards like OpenXR that minimize the effort for (later) support of additional platforms. And, last but not least, Microsoft now own Zenimax with Bethesda and id software, so if you want to play a future Minecraft or Elder Scrolls or Doom or Halo or Call of Duty in VR, you will have to go through Microsoft. And these established game brands will have a much higher draw to the undecided masses than 20% more FoV or pixels will ever have. Content was and always will be king, something that may throw a(nother) wrench into Meta’s plans to dominate VR/XR.

      • Dragon Marble

        It may not be as simple as you think. Social media is a mature market. Why doesn’t MS build its own version of FB? Console gaming has existed for ages, and it’s still the same old big players in the market. MS Excel is terrible, and yet no one has been able to produce an alternative.

        You are ignoring an important fact. The reason tech companies tend to grow so big and maintain their monopoly so easily is due to something unique to tech: the peer effect. Everybody is using Excel because everybody else is using it. This makes barriers of entry naturally high for any new comers. Even Meta is finding it hard to take on Twitter.

        • Christian Schildwaechter

          There is no “peer effect” in that people buy MS products because they are just following others, “nobody was ever fired for buying IBM” has not been valid for decades. Companies don’t switch from Microsoft Office, because the costs of retraining their employees would by far exceed the license costs, as most have used MS products for years and now often grew up with them. And this causes a cycle, with schools and new users starting with MS Office again.

          It is much simpler to find someone that can deal with MS Word than LibreOffice Writer. New players like “Google Docs Editors” therefore have to offer significant improvements like “live online collaboration with automatic versioning” or “running everywhere in a browser” to compete. Rule of thumb: a new product has to be at least 50% better to get people to switch and relearn, and that is hard with products like office suits, where most users only ever scratch the surface.

          Compared to that, switching from a PS5 to an Xbox comes with basically no necessary investment besides the hardware itself, at most you have to rebuy titles you want to continue to play. Even when switching from Android to iPhone or vice versa, you will find most popular apps are available on both, and work pretty much the same way. And people have switched consoles a lot, it just takes a while due to now seven year release cycles. Atari, Sega and Nintendo once dominated the market, until first Sony spoiled the fun by turning the publishing model on its head with games pressed onto cheap CDROMs (they co-invented) instead of expensive cartridges. And then Microsoft messed that up again by taking gaming PC hardware and a reduced Win2000/DirectX and turning it into a console running , which Sony now had to follow. Most new attempts to enter the console market fail, Ouya tanked, Google quickly buried Stadia. But NVidia’s game streaming service Geforce Now launched in 2020-02 with 1mn users, had 10mn by 2021-04, 20mn by 2022-08 and is now at about 25mn, so they got to more users than Xbox Series X/S or Quest 2 (very apples to oranges) in 3.5 years.

          Gaming is very different from productivity use, which comes with a significant investment to learn workflows, or from anything driven by “network effects” like social networks. There the value is the people on that network, not the network tech itself, and switching means losing contact to these people. Google failed to establish Google+ and Orkut, Facebook failed with photos and instant messaging, so they bough the more successful Instagram (USD 1bn) and WhatsApp (USD 19bn) instead. Apple slipped in by making iMessage and Facetime the default for their user base, which they now leverage on the AVP. Microsoft bought Skype for billions, probably not a good investment. If they one day decide to create a new social network, it will most likely be based on Minecraft, because going directly against an established large social network without any significant user base is stupid and will only have a chance if that network itself utterly messes up like Twitter/X currently does.

          You are currently lucky if some of your friends use VR at all, so it is unlikely that many people will pick their HMDs based on that. The only part of gaming that is seriously impacted by peer/networking effects are massive multiplayer online games like Call of Duty or Fortnite that people play to hang out with their friends that gather there, even if they’d actually prefer to play another kind of game.

          • silvaring

            Hey Christian, only read about 70% of your above comment, and I wondered do you or would you be interested in posting a blog? I feel like a lot of these kinds of comments of yours (deep, well constructed analytical arguments) are a bit wasted, since they are in the bottom of random vr threads and not in centralized articles or websites etc.

            Let me know what you think of that, sorry didnt mean to come across offensively I genuinely think your comments are super helpful when I come across them.

          • Christian Schildwaechter

            I regularly consider it, not necessarily a blog, but something that allows for internal references. I am very aware that adding comments to a public new forum isn’t really the proper place, both because the form doesn’t match a discussion, and the inappropriate length of the resulting posts due to squeezing lots of background information into them.

            It’s sort of on my the ToDo list. But to get anything done, I should first prune my projects lists by 90%, so it’s not likely to happen anytime soon. My usual answer whenever someone asks me if I would be interested in writing articles or blog posts instead of comments, to make the invested time a bit less wasted, is “yes in theory, but I am somewhat afraid that the required extra effort would somewhat spoil the fun”. Currently I kind treat these comments as sort of a mental exercise, as I like to engage with complex, technical matters and doing research and analysis for them, and blissfully ignore that the results will mostly be lost in the void. In reality most of them are actually triggered by procrastination. Whether/when this will change towards a more fitting/systematic approach depends on how some other projects develop.

          • silvaring

            I feel the same way about my comments, a lot of procrastination mixed with habit, and a bit of intrigue because I love video game technology.

            Do you capture your research or analysis anywhere else? Like on manuscripts or documents of any kind?

          • Christian Schildwaechter

            Lots of notes and folders, usually of the “should be finished some day” kind, that didn’t even make it to a ToDo list, and a log of pretty much everything I’ve posted for about the last decade, much of which is about VR. Printing out that log would currently end in a document just short of 2500 pages.

            The perspective of wading through these and turning parts of it into something cohesive and usable for others [which is part of the “plan”‘] doesn’t exactly match up with my fun/procrastination driven approach, so I mostly use it as a reference for new posts, so I don’t have to look things up again. And occasionally I include extra informations in comments mostly because that way I will later find them more easily by searching the log of my posts.

          • silvaring

            CTRL+F is the greatest invention isnt it : )

            I have a similar method, but not via comments. Its a good idea though, to use a comment thread as a space for providing research… Id find it so hard to balance with the discussion aspect though, and having to explain x3 what vergence accommadation conflict is…. nah I’ll pass :)

            be well.
            d

          • Christian Schildwaechter

            Without access to full text search in the external, digital part of my brain, my perceived IQ immediately drops by 20 points. Though this turns out be both a blessing and a curse, as I am no longer forced to go through notes and properly summarize and categorize them to have a chance to ever find the information again, instead I can create large dumps of unstructured text and still find things later.

            Which I unfortunately do a lot. The lack of structure sometimes leads to bizarre results, where I google for some information only to find a post I’ve written myself explaining exactly what I am looking for, only I completely forgot that I already researched it and therefore never bothered to look into my log.

            I’ve recently become obsessed with the free Obsidian note taking app, which allows to still just dump lots of text into it, but also some very light formatting with MarkUp and linking between articles. I’ve hassled with Wikis and custom databases allowing the same before, but these all introduce way too much friction and/or restrictions.

            But since Obsidian stores everything in plain text files, all my previous workflows still work, but it is much easier to introduce some structure. There is a genius shortcut that shows a graph of text node connections based on links/backlinks, and it can be extended by using tags. So even without any extra effort to structure information, it becomes much easier to get an overview, avoiding the situation where I simply forget that I already found and stored the information somewhere. And of course it has an extremely powerful search function and plugins that basically turn it into a database that can be queries in an SQL-like fashion.

            Something in that direction is also what I’d prefer instead of a blog, which are usually about some current events, with older information being mostly ignored. I am mostly interested in long term development and fundamentals, and a lot of the things I posted about e.g. the physics of optics and the advantages of different lens types for VR years ago are still valid. So a form that allows to just link to a more in-depth explanation would work better, and would also avoid having to explain VAC over and over again.

          • silvaring

            Wheres AI when you need it :) I can see a Language Model in your future…. and the future sounds very bright. Just keep me in the loop when / if you decide to publish or want to share your work : ) My own VR research is very limited technically, just a bunch of notes on important (for me) milestones, like the Oculus lawsuit, DK1 / DK2 release, hand tracking updates etc.

          • Dragon Marble

            Yes, “network effect”. I knew there’s right word but just couldn’t remember it. VR doesn’t have much of it YET because it’s not mainstream yet. It doesn’t yet have a large, integrated ecosystem. In the future, your VR headset will talk to your TV, your phone, and other people’s headsets. That’s when VR goes mainstream. If you wait until then — which is also when there’s serious money to be made — then it’s too late.

  • Jeremiah Tothenations

    It is niche and it would remain that way if all had that attitude, if and when it does break out into the mainstream, Microsoft will again be left playing catchup. They can count on me not to support them at that time.

  • ViRGiN

    FUN FACT: Microsoft estimated Valve revenue in 2021 at $6.5bln.
    Company that has zero expenses other than slightly-above-average salary for its few hundred employees and cost of electricity.

    • Jason McDonald

      As the owner of 2 VR Arcades I can assure you my friend running a biz has a few more overheads then you seem to be aware of.

      • ViRGiN

        What? Of course you have more expenses.
        Valve doesn’t, since all their income is essentially passive. As long as the servers is running, people continue to buy the games.

        As an owner of 2 VR Arcades, did you buy the licenses to be able to charge for people to play at your place?

        • Jason McDonald

          Thanks for the question and if you need some info on LBE. Commercial / licensing options let me know and I’ll guide you in the right direction .

          In regards to ‘zero expenses’ – I share AT LEAST the following with Valve which I’ll outline to help you with your understanding of costs associated with business operations.

          Workers comp / rent / many different insurances / GST / marketing / accounting costs / tax costs / Super / Buisness compliance fees / communications / vehicles / advertising / registration fees / data main fees / IT support / cleaning / maintenance / security / growth costs / LSL etc

          No where close to zero expenses my learned friend but I appreciate your understanding as I have people everyday stating I must be making bank as they simply are ignorant to the extensive costs to run any Buisness – let alone something that’s cutting edge.

          Have a great VR day.

          • ViRGiN

            Are you seriously continuing to compare your little vr arcades to behemoth of valve?
            Valve surely items the building where they are located. Everything else is part of running business as small as hairdress saloon. My entire point is that valve has basically no expenses on anything outside employees and electricity. Where do they run their advertising? They don’t have any because “buying game on pc” buy default indicates buying it through steam, as they have full monopoly. They don’t have growth costs cause they don’t grow, just their revenue. They could fire 90% of work force and nothing would change. Money exchanges hands without human intervention. It’s so bad that when you actually need help from customer service, it takes forever for anyone to reply since there are so few people working there.

            I don’t know what are you even trying to imply for the scale of valve.
            And i would never want to run vr arcade – most of them fail quickly. Vr outside largest cities is something better consumed at home. Why would i pay you 30 dollars or whatever for an 1 hour of play when for 10x that i can own one forever and sell it whenever i want to? If you found success, that’s nice for you.

          • Jason McDonald

            Dcffff

          • Jason McDonald

            Thanks. I misunderstood your claim Valve had zero costs apart from well paid staff and electricity. In context of your clarity I would agree it’s a business mode which is highly automated and requires minimal humans (except trying to get some service which is a shocker). EBay is the same and I wish I had thought of it first.

            In regards to LBE – the social elements is what set LBE apart from home. Even if everyone has free roam on their own tennis court in every continent of the world .. after the VR is finished and the hmd removed .. you can’t pour each other a beer.
            That’s why my LBE’s will always have what a home doesn’t – the chance to also physically connect with your friends.
            Thanks for the conversation – it’s has been interesting

  • ApocalypseShadow

    Microsoft is doing what they always do: wait for others to create a market, jump in, then try and act like they were the leaders in that market from the beginning.

    They tried controlling the market with Windows headsets on PC. And I’d bet all those manufacturers got sweetheart deals on Windows software for their businesses in exchange for making them. Problem was typical Microsoft: they made no games to go with their push. Relying on others to get them there was why they failed. Made no Forza, Halo or Gears or any software for their own headsets.

    And as for Hololens, they tried to sell it at E3 with Minecraft on stage with John Carmack against VR but no one cared. Then they tried to play it off like they weren’t really trying to get gamers interested. So now, they push the tech on the US Military.

    Microsoft are never leaders unless it’s something negative that usually fleeces consumers pockets and makes easy money. When they bought all those developers and two publishers, they took Bethesda, Inxile, ID Software, Compulsion, Ninja Theory, Double Fine, etc with them that actually made games for VR. Another damaging thing to a growing platform because none of them make VR games anymore.

    Thanks Microsoft. You suck. Who cares if you do or don’t. Let others innovate while Microsoft imitates.

  • Well, basically this means that they are not lying to us: they truly believe that VR is currently a niche technology and are not investing much into it

    • ViRGiN

      Reminds me of one other company

  • Eagleman

    One of the documents (PX4066 Redacted) with the subject line, RE: Meta – Quest Event, discussed bringing XCloud to the Quest as well but said that Meta did not have the MAU that MSFT wants yet. Hoping for this announcement at Meta Connect next week, hopefully the full year since has brought the number of users necessary