Charts, charts, charts. I love charts. Don’t you? Methodologically-believable ones, in any case.
Edison Research and NPR published many a chart in the Spoken Word Audio Report that they released earlier this week — I guess we’re going back to “Spoken Word” as the umbrella category! I remember folks making fun of me for using it back in the early days of Hot Pod — and you can find the whole package here. Because it’s partly a marketing tool, you’re gonna have to insert some identifying information before you can download the report, so, data transaction and all that.
Anyway, I’ll do the right thing here and say that there’s a lot to pull from this report and that you should totally check out the whole document. But I’ll partly renege on completely doing the right thing here, because daddy needs content, by pointing to the three findings that stuck out to me the most:
(1) Among the key findings: that Americans are spending more time listening to spoken word audio… at the expense of music. Which is something that I’ve long heard postulated, but never quite believed. (If anything, I thought the two things would grow in tandem.) I guess we now have some data suggesting that this is the case.
(2) There are a bunch of data points in there about demographic splits between digital-first listening and analog-first listening: how the split pretty much falls along age lines, how the gender split is pretty much the same, and how digital-first listening is noticeably more diverse than analog-first listening.
But I’m really drawn to the finding that digital-first listeners appear significantly more oriented around the ability control of the spoken word audio content they’re consuming.
(3) Note the decline in AM/FM Radio share:
What I’m not saying: AM/FM Radio is going to fall out existence.
What I am saying: it’s pretty hard to hold onto the position that AM/FM is going to be the dominant medium distribution of spoken word audio forever and until the end of the time, which is pretty much something I’m still being told, aggressively and derisively, by certain circles.
Rev, Under Continued Scrutiny. On Tuesday’s newsletter, I linked to a developing story surrounding Rev, a popular transcription services platform that depends on contractors, which had begun to come under fire for its decision to alter its freelance payout structure — ostensibly to better account for work-value spread across different audio types — seemingly with little warning given to the community that provides services on its marketplace. Gizmodo had the first report on this development, along with subsequent follow-ups, which you should check out here and here.
In case it isn’t clear why I’m dedicating this much newsletter space to this story: Rev is a fairly widely used transcription option in the podcast and radio community, as well as the journalism and media industry more generally. Not always, of course, as there are more alternatives these days for producers to try out and cycle through — including various speech-to-text solutions, like Trint and Descript — but Rev, as far as I know, remains a consistent option for many.
Anyway, this story, as originally catalyzed by Gizmodo, is essentially another instantiating case in an increasingly prominent reality: the new volatility and difficulty of life in the age of the gig economy as mediated by technology platforms. But this particular case took on a slightly sharper edge yesterday when OneZero’s Sarah Emerson published a report highlighting severe security issues with the way Rev handles customer data on its platform.
Here’s Emerson, detailing out the specifics: When customers submit audio to the service, it goes into a database that is accessible to all of Rev’s 40,000 transcribers (or “Revvers,” as the company calls them). The database that Revvers use to select or “claim” jobs, screenshots of which were viewed by OneZero, lists full names and business titles for customers and permits any transcriber to listen to an audio or video file, so long as it is unclaimed. Though the company touts its “strict customer confidentiality policy” and claims “your files are private and protected from unauthorized access,” the reality is that your audio is essentially up for grabs during its time spent in Rev’s database.
Until a shift in security policy last year, Revvers could also claim and download files, then “unclaim” and return them to the queue. In October 2018, Rev emailed freelancers, saying it was removing the option to download files “due to recent breaches of our confidentiality agreement and an overall effort to bolster our efforts to protect the data and privacy of our customers.” These breaches were not publicly announced. Yesterday, Rev CEO Jason Chicola sent out an email to the company’s freelance contractor community declaring that Rev had gotten “a few things wrong” with the pay policy change, but did not discuss the security issues. Again, Gizmodo was on that.
Because I’ve seen a certain kind of chatter on this topic online, I feel compelled to say that the following statements can all be true at the same time:
- Providing episode transcriptions remains an important responsibility, at the very least for accessibility reasons
- Providing episode transcriptions can be costly and/or laborious in general, perhaps to a point that some, particularly independents, may feel prohibitive
- Rev is one of the more affordable options for transcription
- Rev is not particularly morally sound to use; let alone sound in terms of data security
- You should probably still provide episode transcriptions
Quick hat tip to this piece Caroline wrote last November.