Last week, Google announced it was shutting down its gaming service Stadia. This came as a surprise to approximately no one. Google had already deprecated the service, but the company still drew a lot of negative attention on-line, with Ben Thompson justifiably calling the whole affair “lazy and stupid.” This came despite the fact that it surprised no one, in fact much of the world’s ire about the closing was centered on the fact that it comes as no surprise from a company with a really bad track record of starting projects only to lose interest and move on. Going even further back to 2019 when Google launched Stadia, we noted that:

The problem with Stadia is that it is going to be very dependent on getting games ported over to run on Google’s servers. 

We were being polite. What we really meant was that Stadia reeked of an engineering solution in search of a problem to solve. We commented around the launch that much of the Stadia marketing material and business proposition felt like it had been done last, as in the engineers said “We built this thing.” and then left the commercial team to figure out how to make it into a business. This seemed to play out as the company introduced one pricing scheme only to change it down the road and a whole series of other strategic pivots.

If Google did not have such have an awful track record here we would want to give them credit for at least experimenting. Stadia solved some important engineering problems around online game streaming. This remains an important subject, and we think that in the hands of a better-run organization Stadia could have been an interesting product.

However, to achieve that it should have been sold as a service to developers. We have written at length about the challenges that networking poses for game developers (and here and here). Any discussion about the “Metaverse” and our hot 5G Future has to contend with the fact that today’s networks are barely capable of maintaining consistency inside the very controlled environment that is a single game. Before we can have Augmented Reality goggles that overlay all the world’s data on top of the real world, someone is going to have to sort this out. Spoiler alert – the telcos can provide only a part of the solution, at best.

So Google gets full credit for identifying the problem. Their mistake was to sell it to consumers. Instead, Stadia should have been a full blown Google Cloud Platform (GCP) API. This would have allowed developers to plug into the service rather than be forced to build their own global CDN network as Riot has done. This would have been a compelling offering for GCP. AWS has something similar (that also suffers from its own lack of attention) but Google’s ownership of Android offers some very interesting potential for synergy, coupling phones and Chromebooks tightly to the service. The ultimate irony is that Google could have monetized it with ads, one of the things they have done very well for a very long time. A simple Stadia API with an attached ad platform could have been a huge hit. We are not mobile ads experts, so we struggle to understand how people who are did not see this opportunity.

We understand the difficulties. Gaming is a complex ecosystem with many stakeholders all pulling in different directions. The developers, in general, do not have a great grasp of networking, this is an area that is still beyond the core competency of many of them. The publishers have financial incentives that work against this, and are likely to see a networking service only in terms of the added cost. The platform owners, like Google, just have too many other things going on. Small start-ups trying to tackle this problem have to swim among these proverbial sharks and confront thorny issues like who is hosting the code, and how to get developers to trust and embed their service.

Our goal here is not to castigate Google further, others have done that better, already. Instead, we want to once again draw attention to this significant opportunity. We recognize that it is not an easy problem to solve. Google is not the only company to fail here, there are a half dozen failed start-ups in the space, as well as several that are still soldiering on. Someday one of those will connect. In the interim, if Google has any interest, we would like to make an offer for Stadia – $1 for the service, the associated IP, a support contact at GCP and an introduction to Google Capital.