How to segment your back catalog for ad insertion

If you are moving your podcast onto Publish and you plan on using Dovetail for dynamic ad-insertion, keep in mind that Dovetail relies on segmented audio instead of time stamps, so if you want to sell midrolls you will need to segment all future episodes and your back catalog. 

It can be a daunting task to segment all your old episodes, and we recommend getting started ASAP. Keep in mind that you can import your podcast episodes as they are (make sure to disable ads before import) and then work backwards to replace the audio as you segment the episodes. 

When replacing your audio files, make sure to update your episode to the appropriate template and then upload the new audio files.

Here are some guidelines for segmenting your audio.  

How many segments should I have?

  • This should be determined by you and the advertising operations team before you segment the audio. Your number of segments will depend on whether or not your episode will include midroll ad spots
  • Create a separate segment for each episode teaser/preview, apart from the main show audio, so we can insert the pre-roll ads between the teaser/preview and the main show audio
  • Create 2 files for each episode if you will have mid-rolls (Segment 1 will play before the mid-rolls and Segment 2 will play after)

What should I listen for or change while segmenting?

  • Remove sonic IDs from the end and/or top of episodes (they will be injected)
  • Remove sponsor ads
  • Remove surveys and cross-promos of other shows
  • Remove dated material (e.g. fundraising, live shows, special announcements, holiday references)
  • Listen for inconsistent audio levels and poor audio finish (ex: 20 seconds of music)
  • Make sure all segments use the same audio specs (channels/rate/bitrate)
  • Listen for places that might need an added transition between segments  
    • You may want to have 3-5 seconds of music fading out of the first segment and 3-5 seconds of music fading into the second segment in case there are no ads stitched into a download
  • Insert at least a half-second of silence at the end of each episode segment

Naming Conventions for segmented files

PRX can run a script to swap in your newly segmented audio if you have a large back catalog. If you are using this service,  you must use the following naming conventions for your files:

For single-segment:
Format: SeriesTitle_EpisodeTitle_Episode#.mp3
Example: MyPodcast_LeftRight_Ep68.mp3

If mid-rolls: 
Format: SeriesTitle_EpisodeTitle_Episode#_Segment#.mp3
Examples:
MyPodcast_LeftRight_Ep68_Seg1.mp3
MyPodcast_LeftRight_Ep68_Seg2.mp3

Adding ID3 tags

ID3 data is how you add meta data like a cover image, title, and author to an audio file. This is not required, but is recommended.

Most podcasters add this to their files, and most do it by adding the file to their iTunes library, then right-clicking on it, choosing "get info", and filling out the data.

Here's a pretty good explanation:
http://www.prestopod.com/how-to-add-id3-tags-to-your-podcast-mp3-file/

Please note that you only need to add ID3 tags to the first segment you’ll be uploading to Publish.

 

 

 

Was this article helpful?
0 out of 0 found this helpful
Have more questions? Submit a request