You know I love looking at Discovery metrics, from streaming signups to the ratings in Canada! So here's a new angle on it that I discovered today:
A bunch of internal Amazon Prime documents leaked a few weeks ago, which led Reuters to take a very interesting look at how Amazon measures the success of its shows. The whole thing is worth reading, but here's the key passage:
The article goes on to say that The Grand Tour Season 1 had a "cost per first stream" of only $49, which is well below the cost the streamer will pay for Amazon Prime, so that's money in the bank for Amazon, and probably explains why Prime keeps promoting it to me even though I've never watched a car show in my life and never will. This made Grand Tour Amazon's most successful show. By contrast, Good Girls Revolt, which was cancelled after only a few weeks despite critical acclaim, had a "cost per first stream" of $1,560, so its viewers would have to subscribe to Amazon Prime for more than 15 years before Amazon would start to show a profit on Good Girls Revolt.
Using the rest of the data in the article, we can reconstruct a pretty interesting chart showing the "cost per first stream" of several shows in Amazon's arsenal. Better yet, using other publicly available data, we can also come up with some estimates the "cost per first stream" of Star Trek Discovery. (Of course, our conclusions will vary enormously based on our assumptions; see below.) We can put that all into one chart, and... voila!
Finally, with this metric, Star Trek Discovery's financial logic begins to make some kind of sense.
The "optimistic" estimate for Discovery assumes a cost-per-episode of $5.2 million (which is the lowest figure I could find, from back when it was a 13-episode series) and that all 500,000 users who signed up for CBS All Access during Discovery's run were there for Discovery and streamed Discovery first. As you can see, if those generous assumptions are true, then Discovery is a considerable success, not quite paying for itself, but performing dramatically better than many streaming shows that have gone on to run for multiple seasons.
The "pessimistic" estimate for Discovery assumes a cost-per-episode of $8 million, as reported by the Hollywood Reporter. That would make it the most expensive show on this list, which is saying something, because the leaked figures from Amazon include the shows' marketing budgets, while the only figures we have for Discovery are for the (smaller) production budget. So if Discovery's production budget alone is nearly double the production plus marketing budget of The Man in the High Castle, that makes Discovery an expensive venture indeed. Nevertheless, there are indications that this is indeed closer to the true figure. The other assumption in this estimate is that only one-quarter of the 500,000 people who signed up for All Access during Discovery's five-month run streamed Discovery first (so DISCO is credited with 125,000 first streams).
As you can see in the chart, if these negative assumptions are close to correct, then Discovery is clearly a troubled program, which will probably need steep budget cuts in future seasons in order to continue to justify itself. On the other hand, it's still doing much better than Good Girls Revolt, and Sneaky Pete (with a similarly high cost-per-first-stream) got renewed for a second season, so, at the very least, we can say that CBS's decision to renew Star Trek Discovery was justifiable, not an insane move driven by a fear of losing face (which is a suggestion I have made previously).
The "average" estimate for Discovery simply takes the midpoint of the optimistic and pessimistic figures, assuming a budget of $6.6 million per episode and that half of CBS AA's subscribers during its run (aka 250,000 new users) were "first streaming" Discovery. On this average estimate, Discovery looks like an average streaming show. Which is, y'know, not terrible for fans and probably would ensure a run of at least 2-3 seasons even without steep cuts.
Based on everything I've read, I'm inclined to think that Discovery is more toward the pessimistic end of this, but it is worth bearing in mind that, as CBS struggles to build its streaming platform, it may be less sensitive to money-losing shows than more established competitors, simply because CBS AA needs all the content it can get right now to be an attractive proposition to consumers. Another factor to consider (and I have no idea how large a factor this is) is that Discovery comes with licensing deals and toy sales, whereas shows like The Man in the High Castle pretty much have to make their own profits.
A bunch of internal Amazon Prime documents leaked a few weeks ago, which led Reuters to take a very interesting look at how Amazon measures the success of its shows. The whole thing is worth reading, but here's the key passage:
Basically, "cost per first stream" is the amount of money Amazon has to spend on a show in order to acquire a new Amazon Prime subscriber.
The article goes on to say that The Grand Tour Season 1 had a "cost per first stream" of only $49, which is well below the cost the streamer will pay for Amazon Prime, so that's money in the bank for Amazon, and probably explains why Prime keeps promoting it to me even though I've never watched a car show in my life and never will. This made Grand Tour Amazon's most successful show. By contrast, Good Girls Revolt, which was cancelled after only a few weeks despite critical acclaim, had a "cost per first stream" of $1,560, so its viewers would have to subscribe to Amazon Prime for more than 15 years before Amazon would start to show a profit on Good Girls Revolt.
Using the rest of the data in the article, we can reconstruct a pretty interesting chart showing the "cost per first stream" of several shows in Amazon's arsenal. Better yet, using other publicly available data, we can also come up with some estimates the "cost per first stream" of Star Trek Discovery. (Of course, our conclusions will vary enormously based on our assumptions; see below.) We can put that all into one chart, and... voila!
Finally, with this metric, Star Trek Discovery's financial logic begins to make some kind of sense.
The "optimistic" estimate for Discovery assumes a cost-per-episode of $5.2 million (which is the lowest figure I could find, from back when it was a 13-episode series) and that all 500,000 users who signed up for CBS All Access during Discovery's run were there for Discovery and streamed Discovery first. As you can see, if those generous assumptions are true, then Discovery is a considerable success, not quite paying for itself, but performing dramatically better than many streaming shows that have gone on to run for multiple seasons.
The "pessimistic" estimate for Discovery assumes a cost-per-episode of $8 million, as reported by the Hollywood Reporter. That would make it the most expensive show on this list, which is saying something, because the leaked figures from Amazon include the shows' marketing budgets, while the only figures we have for Discovery are for the (smaller) production budget. So if Discovery's production budget alone is nearly double the production plus marketing budget of The Man in the High Castle, that makes Discovery an expensive venture indeed. Nevertheless, there are indications that this is indeed closer to the true figure. The other assumption in this estimate is that only one-quarter of the 500,000 people who signed up for All Access during Discovery's five-month run streamed Discovery first (so DISCO is credited with 125,000 first streams).
As you can see in the chart, if these negative assumptions are close to correct, then Discovery is clearly a troubled program, which will probably need steep budget cuts in future seasons in order to continue to justify itself. On the other hand, it's still doing much better than Good Girls Revolt, and Sneaky Pete (with a similarly high cost-per-first-stream) got renewed for a second season, so, at the very least, we can say that CBS's decision to renew Star Trek Discovery was justifiable, not an insane move driven by a fear of losing face (which is a suggestion I have made previously).
The "average" estimate for Discovery simply takes the midpoint of the optimistic and pessimistic figures, assuming a budget of $6.6 million per episode and that half of CBS AA's subscribers during its run (aka 250,000 new users) were "first streaming" Discovery. On this average estimate, Discovery looks like an average streaming show. Which is, y'know, not terrible for fans and probably would ensure a run of at least 2-3 seasons even without steep cuts.
Based on everything I've read, I'm inclined to think that Discovery is more toward the pessimistic end of this, but it is worth bearing in mind that, as CBS struggles to build its streaming platform, it may be less sensitive to money-losing shows than more established competitors, simply because CBS AA needs all the content it can get right now to be an attractive proposition to consumers. Another factor to consider (and I have no idea how large a factor this is) is that Discovery comes with licensing deals and toy sales, whereas shows like The Man in the High Castle pretty much have to make their own profits.
All I know is that I paid ten dollars a month after one free month. Watched Discovery episodes each week. It had its break and I kept my (Netflix) service then started up again. Cancelled Netflix last month. Saw about two or three movies and one episode of Designated Survivor and didn't like it.
