diff --git a/README.md b/README.md
new file mode 100644
index 0000000..c32a0eb
--- /dev/null
+++ b/README.md
@@ -0,0 +1,39 @@
+# Youtube-Archiver
+***Always** keep a local backup.*
+
+This utility allows you to define a collection of links to youtube-channels and/or playlists, which you want to keep synced to your local storage.
+All you have to do is to call `download.sh`, which will download all videos of your defined playlists that are not already saved locally.
+
+This basically means, after doing the initial big download, you could run it as a cronjob, since it won't take long when it's only downloading newly uploaded videos.
+
+## How does it work?
+Pretty easily.
+
+#### 1
+First, you define the stuff you want to download, in yaml files. For more info on syntax, see the [example file](https://github.com/Leonetienne/Youtube-Archiver/blob/master/linklist-example.yaml).
+```yaml
+# linklist-music.yaml
+
+# Eminem
+-
+  url: https://www.youtube.com/c/eminem/videos
+  folder-name: Eminem
+
+# DJ. S3RL
+-
+  url: https://www.youtube.com/channel/UCb6JTMjrHZCYFD9Y04CBk9g
+
+# Snoop Dogg
+-
+  url: https://www.youtube.com/watch?v=zV-Xcy_bF2w&list=PL75C4B43E5F60A82C
+  folder-name: Snoog Dogg
+```
+
+#### 2
+Secondly, calling `download.sh` will invoke the two python scripts responsible for downloading.
+
+#### 3
+These python scripts will invoke [yt-dlp](https://github.com/yt-dlp/yt-dlp) for each link supplied. Yt-dlp runs in download-archive mode,
+which basically means it won't download a video twice. It will save it's ID and check if it's already saved before downloading.
+
+Files will be saved to `./videos/<folder>/` and `./music/<folder>`.