This is called “maybe we can trick you into thinking our content is more interesting than it is.” Netflix’s gone lowbrow, more global but the new/old 500-channels and nothing to watch.
3
If you are not the consumer, you are the product.
1
So, like every other business, it suggests things based on what you have previously watched before. I get ads for gay movies all the time, or comedy specials. This is an overreaction.
2
When something is free, you're the product. When it's really cheap, you're still the product. Pay full-price in case to go to the movies, and nobody even knows you're there. Buy a DVD (a WHAT?) and, same.
Netflix is a business and they are out to make money. They do it by renting us with entertainment. The more we use the service the more money they make. I totally understand why they are sending targeted ads. If you liked something once, they are going to work hard to give you something like the thing you already liked. That's just business.
2
I can guess what happened here: Netflix doesn't have data on customers' race, true. It doesn't need it. Race, gender, income, geographic location, etc - they've all been used for decades now by businesses that slice and dice the data to try to predict behavior in order to influence behavior, whether it's buying something, voting for someone, re-subscribing to Netflix or whatever.
But now Netflix doesn't need to target by race, gender etc. It can target by the individual, something other companies can't do. Who cares if someone is black, if you can see and change their behavior individually with a computer algorithm learning what types of thumbnails they respond to? If someone disproportionately clicks on thumbnails with black characters, then give them more thumbnails like that. Is that person black, white, or purple ? Who knows? Who cares? You don't need race when you have behavior.
This person probably clicked on a lot of thumbnails with black characters and told the Netflix algorithm that's what they like, so the algorithm obliged with more thumbnails like that. The algorithm is a mirror; don't blame if if you don't like what the mirror shows you.
If person was white, and clicking on black characters in thumbnails would that have been a problem? Netflix has 137 million subscribers, so everything you can imagine is happening , so that's happening too. If anyone has a problem with it, stop clicking on thumbnails with black characters and the algorithm will learn.
5
@nerdrage yes this is how machine learning works and why it is exactly a scary proposition. Note that the targeting logic is sort of blind, but the artwork creation and design is not. There are human elements in this strategy, specifically designing a racially charged cover that goes through multiple layers of approval, legal, etc before released in the wild to he algorithms. Obviously, someone at Netflix knows that minorities who consume minority movies will react to this, it’s not like someone didn’t write the learning algorithm in the first place.
Intentional? This is probably a result of poorly designed strategies, probably junior data scientists, who lack common sense and are chasing a KPI.
There are deep implications on this sort of strategy. If each person’s world is personalized, our deepest biases and tendencies will be reinforced.
6
Micro-targeting is often irrational and downright awful. What it mostly is, is awfully wrong about what we are. If you are remotely complex as a human, micro-targeting will get you wrong.
I'm sure Netflix would not know what to make of me, as I watch a lot of "black" shows, feminist shows, cop shows, and everything in between. No freckles off my lily white nose if they can't tell what I am from what I watch.
If you see a range of images, they can't tell what the heck you are from what you watch. Keep watching a wide range of shows. If you see only certain targeted images... Netflix thinks they know what you are.
Shock them. Watch unexpected things.
3
@Dejah Businesses have been using targeting to market to customers for decades and we've all been exposed to it. It's far less accurate than what Netflix is doing.
For example, traditional marketing would think that you, being white, do not particularly want to see black shows. You're being lumped in with the majority and that's been true for all the marketing you've seen all your life, whether you know it or not.
But Netflix neither knows nor cares about your race. It just looks at your behavior, and behavior tells the truth. If your behavior is similar to black Netflix subscribers than white Netflix subscribers, then you will see thumbnails more similar to the black subscribers.
That is, assuming your self-report of your behavior is accurate. But Netflix won't care about that; the algorithm only cares about your actual behavior. It's a mirror. Whether anyone likes what they see in the mirror is up to them.
2
Hate it but have no way to stop it.
@TH Yes you do. The algorithm learns. If you are seeing too many black characters or too many white characters or any other quality that annoys you in the thumbnails, then click on the thumbnails you do like. The algorithm will learn your preferences and adapt to suit.
Or you could just not worry about it and let your actual preferences guide your choices, not your notions of what your preferences should be. Why bother to change your preferences to impress an algorithm who doesn't care about any of it? Nobody can see our thumbnails but us. Whatever they may reveal about us can stay hidden as long as we don't go on twitter and blab it to the world.
2
You can't have it both ways. You can't demand entertainment that is specific to your race and then be upset when someone categorizes those titles how they were originally intended.
This isn't Netflix having a file that says "this viewer is this race".
It's them having a file that says "this viewer really seems to watch a lot of movies that were specifically made with a particular race in mind."
---
As far as them promoting colored supporting characters just to draw a colored audience to an otherwise not-so-colorful movie... yeah... that's pretty lame... but I consider that to be simple marketing fraud that's prevalent throughout the practice even when race isn't a theme.
3
Are there any other racial groups in the U.S. besides black and white? Judging by how obsessively articles focus on those two groups, it would seem like 1965.
12
I've noticed this with female content. And I do critizise it because often the shows/films are not female-centered but the artwork gives that impression. It's a con.
2
@laura You're conning yourself. Netflix has learned you like thumbnails with females in them. Start clicking on thumbnails with men more and you can make it go away.
1
“If something is black, I take no offense in being catered to,” Aremu said. “I am black, give me black entertainment, give me more — but don’t take something that isn’t and try to present like it is.”
This viewpoint of insisting on separating our entertainment choices based on race - any race - will contribute to destroying this country.
I don't think the Netflix algorithm is explicitly design to target based on race, but rather implicitly. It is designed to choose the poster (out of range) that has the highest similarity score to other things you like (including other posters you clicked on, other movies you've liked etc). If a user, regardless of race, predominantly watches movies with African-American protagonists, the algorithm will choose the poster with the African-American characters, because it's programmed to find similarity, not necessarily race. If Netflix is to blame for anything is creating a poster that features minor characters in the first place. I agree that's false advertising because it's implied that characters on the poster are centre stage. Netflix is right that the algorithm doesn't target race (hence the phrasing of the statement), but that doesn't mean they don't have advertising practices that confuse the user.
9
I am having a hard time understanding what is wrong with this. Advertising targets specific demographics all the time. As long as it is done in a non stereotypical way this is good business. Should 30 something people be offended if youthful images are shown? Should people who live in southern states be offended if they are geographically targeted? Why does everyone look for hidden racial agendas so much in life? Advertising is a game of creating a return through targeting the message with the greatest chance of an action. If Netflix racially targets it is only because they think it will work and see value in attracting the African-American audience. That is a good thing.
8
@Leon Levitt I think people are unaware of how marketing and advertising has worked for decades. Segmenting populations by age, gender, race, household income and geographic area are standard. And it results in marketing that can be pretty off target because it is stereotyped. There's no other way to do it (at least not till recently). A black person who likes shows with white people will just get lumped in with the majority, who prefer shows with black people.
Netfix targets individually. That black person who likes shows with white people will now see marketing that is individual to him or her, and their race will no longer matter.
But if a black person does actually prefer black characters, they could end up getting what this person did: individually targeted marketing that hones in on their preferences, which may conflict badly with their self-image as someone without racial bias. The algorithms are just a mirror. Whether you like what you see in the mirror is your own business.
I'm African American and when I checked the shows/films listed I see a range of images. For Like Father I see Kelsey Grammer and Kristen Bell. For The Good Cop, just a fingerprint. For Love Actually I see Liam Neeson. And, for Set It Up I see Taye Diggs and Lucy Liu.
4
@Melanie That's because you click on thumbnails with white people in them. Stop doing that, click exclusively on black people and you'll start getting nothing but black people. The Netflix algorithm learns your preferences and feeds them back to you. That's what happened to that person: they were just clicking on black characters and the algorithm thought that's what they wanted. I mean how is a computer supposed to know otherwise?
1
Whatever your race, save yourself some time and avoid Set It Up. Silly story with hammy performances by whites, blacks, and Asians.
Have I left out anyone? Maybe. I left after 30 minutes.
5
the explanation by Netfix seems to make a lot of sense...doesn't really seem to be an issue here...
7
Wait! Many advertising for services and products are targeted by race, (other things like age, education, location.). As long as the advertising isn't using racist stereo types or racist images, why would targeting be a problem?
8
@Timothy
I think the objection is that it's false advertising based on race. They are promoting shows as being about African Americans when in fact the African American characters are secondary or marginal to the story.
13
@jim
In another time, and another era, we used to say, "Don't judge a book by its cover."
Similarly, maybe one shouldn't judge a TV show by a still promotional pick?
Disclaimer: I used to watch Seinfeld just to see Julia Louis-Dreyfus.
I am a white 34 year old women and Netflix has target marketed a fair amount of shows and specials with black stars to me. I continue to watch what they suggest because they have good taste, I don't want to scroll that far, and I like trying new things. I genuinely think it's a function of the algorithm because if they looked at me on paper, they would not guessed (correctly) that I am a huge Kat Williams fan
4
I don't care what image Netflix shows me to entice me to watch a show. What I do care about are the following:
-Stop those infernal auto-play previews! Do you really think anyone likes this? At least give us the option to turn it off.
-Stop sending me an email every time I log on telling me that someone has logged onto my Netflix account!
-Bring back the user reviews. Those were useful and helped you to avoid wasting your time.
31
I agree about the user reviews. They said they got rid of them because they were unused. It was my #1 go-to for info about a movie. I hope they bring them back.
7
@Pat
I couldn't agree more about the auto-play previews. It's getting harder and harder to avoid unasked for and unwelcome visual / auditory "stimulation" like this, even in one's own home. And I guarantee you that Netflix didn't axe the user reviews because they were "unused"--how would they even know? More likely, it's because the studios and production companies didn't like all those negative comments cutting into the sales potential of their product.
22
@Pat Agreed. The auto-play is VERY annoying.
5
So they're not targeting you based on the race you are, but rather by the race you watch the most.
Is that wrong?
8
@Justin What’s wrong is that they are falsely portraying their programming. People feel Netflix is baiting and switching:
Imagine that you have a child and therefore a lot of children’s shows are watched. Later Netflix shows you a poster for I Love Lucy with only Little Ricky shown. Yes he’s in the show, but he’s such a minor character and it’s not a kids show as implied.
2
As a Netflix user, I've noticed how they appear, in my opinion, to use race to garner new subscribers. I think that casting decisions in Netflix shows are made for similar demographic purposes (as is often done on network TV shows). In one instance re. Netflix, again in my opinion, that service paid African-American comedian Chris Rock the exorbitant sum of $40 million to produce only two comedy specials. Rock has fans of all stripes, but I think it was done to cater to viewers of "his" demographic. I'm Caucasian. It gets to the point where I see a program on Netflix, Amazon, etc. that seems "directed toward blacks," and I pass it by because I feel it will play on what I see as one-sided, PC "assumptions," stereotypes, and sappy "I can make it" story lines and not on things that are revealing and enlightening about the realities of American life. For example, one such disappointment was "The Chi" on Amazon. It started out good, then deteriorated into pretentious, cliched, PC plot lines.
1