Omny Studio provides measurement analytics service for podcasts originating from the Omny Studio service. Triton Digital, the parent company of Omny Studio services, also offers measurement analytics through its hosting platform-agnostic Podcast Metrics (PCM) solution, which is the source for Triton Digital’s Podcast Rankers. As a result, publishers using Omny Studio through Triton Digital have two analytics options to choose from. This document describes the main differences between them.
Measurement analytics are the backbone of any audio monetization solution. You can’t monetize if you can’t measure. The Triton Digital and Omny Studio measurement and analytics solutions are fair and accurate and are both IAB certified. However, some differences exist between the two, so to get the best overall measurement picture you might consider using both solutions, with a clear understanding of the subtle differences between them.
When/Where to Use Each Solution
Both the Triton Digital and Omny Studio solutions share similar features, but they were built to serve different needs. You can decide which solution is best for you based on the information below:
Podcast Metrics (PCM):
Designed as a measurement tool, best for use at the corporate level and for financial transactions on baked-in advertising.
Measures downloads across multiple hosting platforms and can be used by sellers and buyers to get apples-to-apples comparison numbers from publisher-to-publisher and across hosting services.
PCM data is used to produce Triton Digital’s publicly available podcast reports (rankers).
Omny Studio Analytics
Designed as a best-in-class analytics tool, best for content creators and program directors who are already using Omny Studio to manage their programs.
Used by the content creator to understand how their shows are being consumed (i.e., drop off points, various players, etc.).
Overview of Differences in Metrics
The main difference between the two systems is how the resulting data can be used. Podcast Metrics allows for custom queries, and data can be exported for further custom analysis such as creating pivots. This requires the user to have their own expertise in data analysis and form building in Excel or whatever data analysis platform they are using. Omny Studio Analytics, on the other hand, are presented visually and are generally easier to understand right out of the box but are otherwise fixed and can only be filtered by date.
Also remember these important details:
Podcast Metrics data is based on UTC and Omny Studio Analytics is based on the timezone of the user viewing the reports (taken from browser settings). (See more on this, below.)
Data processing times can, in some cases, be a factor. (See more on this, below.)
The table below outlines the main differences between the two systems.
Feature | Omny Studio | Podcast Metrics |
|---|---|---|
Ability to analyze data with multiple dimensions and filters | ✅ | |
IAB 2.2 certified | ✅ | ✅ |
Schedule recurring email report based on custom query | ✅ | |
Graphs and visualizations of data | ✅ | |
Metrics | ||
Total downloads per clip/episode, program/station, | ✅ | ✅ |
IAB listener metric (IP + User-agent) | ✅ | ✅ |
Unique IP metric | ✅ | |
Subscriber metric per program or organization | ✅ | |
Total downloaded hours per clip/episode, program/station, | ✅ | |
Gross Downloads, showing quantity of downloads before IAB v2 filtration, to be used as a base to understand the filtered portion | ✅ | |
Data (GB), to understand your Triton Digital | ✅ | |
Player/Device | ||
App used to download per clip/episode, program/station, network or organization/publisher-levels | ✅ | ✅ |
Geography | ||
Country and region (state) | ✅ | ✅ |
City-level granularity | ✅ | |
Other Dimensions | ||
CDN | ✅ | |
CMS | ✅ | |
Category/sub-category | ✅ | |
Data source | ✅ | |
Device | ✅ | ✅ |
Device Family | ✅ | ✅ |
Episode ID | ✅ | |
Episode Title | ✅ | ✅ |
Player | ✅ | |
Player Family | ✅ | |
Published Date | ✅ | |
Downloads per feed | ✅ | |
Compare downloads relative to release date | ✅ | |
Consumption information | ||
Per-episode consumption graph for supported players | ✅ | |
Per-program/network/organization consumption | ✅ | |
Consumption analytics API | ✅ | |
The Importance of Timezone
Podcast Metrics data is based on UTC and Omny Studio Analytics is based on the timezone of the user viewing the reports (taken from browser settings).
UTC data provides a consistent reference point when calculating downloads from multiple listeners in different locations. The effect of UTC on download data depends on the duration of the measurement period and the day of the week when it was measured. The longer the duration, the smaller the differences in download numbers. Weekday vs. weekend numbers are affected by variations in listener habits on those days.
The chart below shows an example of how UTC vs. local timezone, and day-of-the-week differences, can create download variances between measurement in Omny and PCM. The chart shows actual download data for a single podcast pulled from both Omny and PCM from a user in the Eastern Standard timezone (EST), measuring downloads from late April into May 1, 2022.
Source | Saturday 4/30 | Sunday 5/1 | 4/28 - 4/30 | Week of 4/25 |
|---|---|---|---|---|
Omny | 2316 | 1932 | 8313 | 16771 |
PCM | 2553 | 2007 | 8615 | 16884 |
Variance | 110.2% | 103.9% | 103.6% | 100.7% |
Saturday 4/30: UTC is four hours ahead of EST, so when pulling Saturday 4/30 in PCM, the local time will include some downloads from Friday 4/29, which will cause a larger variance due to the different listening habits on a weekday versus weekends.
Sunday 5/1: As above, the PCM report is run on Sunday, but it brings in part of Saturday 4/30. Both are weekend days, so the variance is less (weekend to weekend) than when a weekend measurement includes some data from a weekday.
4/28 - 4/30 (3 days): The variance declines, due to reduced volatility over a larger measurement period (three days).
Week of 4/25: The variance declines even more, due to even less volatility over an even larger measurement period (one week).
Processing Times
Data processing times for Omny Analytics and PCM are different. Whereas Omny Analytics processing is in near real time, some PCM processing is batched on a daily basis. As a result, you could see a difference in the two results if you run a report for the previous day and the PCM data for that day has not yet finished processing. Obviously, the more days you report on, the lesser this effect is visible, and if your report's end date is more than a day ago there is no processing time effect at all.
Terminology Differences
In Omny Studio Analytics, the term “clip” refers to a unique episode of a podcast series, where Podcast Metrics uses the term “episode.”
In Omny Studio Analytics, the term “playlist (podcast)” refers to the set of clips to bring together in a RSS feed, where Podcast Metrics uses the term “podcast (feed)”
Different Filtration Methods
Triton and Omny use different static lists of IP addresses. Triton uses the Tagtoday.net official list of Data Center IP addresses. Omny built a list from the public information provided by the large cloud-based organisations (GAFA).
Triton and Omny use different user agent deny lists. Triton uses the IAB user agent Spiders and Bots list and Omny uses the open-source “UA-Parser” user-agent database enhanced with additional proprietary data.
Triton and Omny use different lists for app/device parsing.
Triton and Omny use different methods to group HTTP requests into a session. When there are multiple requests, and the IP address changes between requests while listening to an episode, there will be a variation in the numbers for analytics between the two platforms.
For more info on Triton’s filtering techniques, see the IAB Podcast Measurement Guidelines v.2.2.
For more info on Omny Studio’s filtering techniques, refer to Omny’s “Analytics filtering techniques” Help article.
What to Expect
Considering there are many sources of difference, it’s expected that the percentage of difference will vary by programs and time. Therefore, it’s expected that, for example, Program A could be 5% higher on Triton, while Program B would be 5% higher on Omny. It’s also expected that Program A could be 5% higher on Week 1 and 5% lower on Week 2.
If, for popular programs, one sees recurrent differences of over ±10% on monthly numbers, please reach out to Triton Digital for deeper analysis.