In my continuing quest to become immortal, I've been maintaining a YouTube channel that displays newly produced videos on an ongoing basis (Figure 1). I figured that it would be interesting to see which of my productions did best: the home cooking videos, those dealing with car repair tasks, or even those dissecting and modding all kinds of electronic gadgets. Sure, YouTube provides basic stats in its Studio section, but because I like writing my own code with custom features, I thought it'd be interesting to see how to fetch YouTube stats with Go and process and visualize the data this month.

Register Before You Try

Before granting your homegrown applications access to the viewer metadata, however, YouTube owner Google requires you to register a project [1] to use their API. The idea here is to keep an eye on bot activities and, where needed, to restrict them by imposing a quota. Google distinguishes here between two different mechanisms for authenticating incoming API requests. If requests read private user data or even want to write to the server on behalf of the user, OAuth credentials are required. But if you just want to read data that is already publicly available (e.g., the number of views for a video), all you need is an API key for a project that uses the YouTube Data API (Figure 2), which you can quickly register on the Google Console.

To quickly generate an API key, select a key type from the Credentials menu (Figure 3). Once created (Figure 4), the API key opens a treasure trove of all kinds of public data across Google's offerings. To mitigate the impact of potential misuse with stolen keys, the recommended approach is to restrict the validity of any keys you generate to the areas you actually need – in this case, to the YouTube Data API (Figure 5). You can do this by drilling down on the Restrict API Key option displayed next to the new key.

[...]