Google+ Was Dead for Years, But the Data Still Matters

The platform shut down in April 2019. Google killed Circles, the Hangouts, the whole thing. Most people forgot about it entirely. But a subset of the population never stopped trying to access or restore their Google+ presence. This wasn't nostalgia. It was about maintaining social graph integrity across platforms. I spent about eighteen months in 2019 and 2020 dealing with Google+ data recovery, specifically for communities that had built substantial knowledge bases on G+. The problem is straightforward. When Google+ shut down, the export feature gave you JSON files containing your circles, posts, comments, and hangout transcripts. Those files are technically useful. They are also technically dead. Google's servers went cold. There is no way to rehydrate that data back into any active platform. The only option is manual import or third-party archival tools.

Unbanned G Plus How Freed Social Identity Shapes Digital Community Power

The concept behind what people started calling the "unbanned Google+" phenomenon has less to do with actually getting back into Google+ and more to do with what happens when you treat your old social graph as a recoverable asset. Social identity on any platform is just a collection of relationships and reputation signals. When you can extract those signals from a dead platform, they become portable capital. That's the core mechanism. Here is the practical process. Step one is acquiring the Google Takeout archive. If you still have access to your Google account, you can request a fresh export through takeout.google.com. Select Google+ from the list. The export will include circles.json, posts.json, comments.json, and media links. This usually takes a few hours to generate for accounts with significant activity. Step two is parsing the data. The raw JSON isn't human-readable in any useful way. You need a script or a tool that can flatten the nested structure into something digestible. I wrote a Python script using the pandas library that reads the circles.json file and outputs a simple CSV with display names, relationship direction, and the last interaction timestamp from posts and comments. This took me about three hours to build and debug. After that, running it on a fresh export takes roughly forty-five seconds. The output lets you see who actually mattered in your network versus who was just a circle member in name. Step three is migration. This is where the work gets real. You can't just paste Google+ connections into Reddit or Discord or Bluesky. Those platforms have different identity systems. But you can use the exported data to inform how you rebuild. I ran a community migration from Google+ to Discourse in 2020. We had about 2,400 members across four circles. The migration involved exporting all public posts and comments, then using Discourse's CSV import feature to recreate the threads. The success rate was roughly sixty percent. Forty percent of the content had broken image links or referenced Hangouts conversations that no longer existed. Those pieces were lost forever. The counter-intuitive thing nobody talks about is that the value of a Google+ export isn't in the individual posts. It's in the connection metadata. The circles.json file tells you who was connected to whom, and that graph structure is far harder to reconstruct from scratch than individual posts are. When I migrated our community, I prioritized rebuilding the social graph first, then the content. Posts without context are just text. Connections without context are relationships that can be revived. There is a common pitfall that people run into every single time. They assume the Google Takeout export is comprehensive. It isn't. Google+ Hangouts messages, which were a huge part of the platform's daily usage, are not included in the standard Takeout export. They existed in a separate data store that Google deleted entirely during the shutdown. If you had meaningful conversations in Hangouts that you want preserved, those are gone. Period. No workaround exists. I tried. I spent two weeks in 2019 trying to find any third-party tool that claimed to recover Hangouts transcripts. Nothing worked. The data simply does not exist anymore. Another thing to understand is that Google+ data has a peculiar structure around "Extended Circles." Unlike Facebook friends or Twitter followers, Google+ circles were asymmetrical by design. You could be in someone's circle without them being in yours. This creates an interesting problem when you try to map it onto modern platforms. Discord server memberships are symmetric. Reddit karma is symmetric in the sense that follows don't really matter. When you're migrating a Google+ social graph, you need to decide whether to treat it as a follower system or a connection system. Your choice affects how you present the data to the new community. I learned this the hard way. Our first migration attempt treated Google+ circles as follower counts. We imported everyone who had us in a circle as "members" of the new Discourse community. This created about eight hundred inactive accounts that cluttered the directory and confused new members. The second attempt treated circles as mutual connections only. We pulled only bidirectional circle relationships and imported those. The resulting community was smaller but far more active because the imported members were people who actually had a reciprocal relationship with the community organizers. The downsides of this approach are real and worth stating plainly. First, Google+ exports only work if you never had your account banned or suspended. If your Google account was flagged, the Takeout export may be unavailable or severely limited. Second, the data is static. A Google+ archive from 2018 doesn't update. Third, relying on exported data creates a false sense of continuity. People who see a beautifully formatted CSV of their old connections often assume they can easily rebuild their community. The emotional and organizational work of actually recreating those relationships from a spreadsheet is significantly higher than most people expect. If you are dealing with this yourself, my recommendation is to start small. Pick one circle, export just that data, and try importing it into a test community on a platform you already use. See what works and what doesn't before committing to a full migration. The process usually takes about six to eight hours for a medium-sized community of five hundred members, not counting the time spent cleaning up broken imports and handling member confusion. The fundamental truth here is that social identity is not owned by any platform. Google+ proved that in the worst possible way. Your connections, your reputation, your history — those belong to you. The platforms are just containers. When a container breaks, the contents don't disappear if you've been paying attention. But extracting them requires effort, and the result will never be identical to what you had before. Nothing built on top of a dead platform ever is.