One thing that I expected to be absolutely amazing in 2024 from online vendors was product recommendations.
That vendor, assuming you use a single, persistent account to do purchasing, has a full list of your purchase history. They may well also have browsing data.
And so, given all that data to mine and analyze, one of the few places where I actually have tried to see what a vendor can do in terms of analyzing my preferences…has been really unimpressive.
I’m mostly thinking of Amazon and Steam, since they’re the online vendors that I use the most; Steam in particular has a considerable amount of data it can gather, including video game playtime.
Yet even though Amazon grabs some eyeball space on every page to try to recommend products, I have rarely been recommended anything I actually want to buy on Amazon. Occasionally, sure, but virtually everything I get is via plain old searching. And the most-successful recommendation approach Amazon uses, by far, is just asking me whether I want to purchase more of something that I’ve purchased in the past. I’ll grant that maybe there’s subtlety there that I can’t appreciate from the outside, like computing frequency at which a given “repurchase” recommendation happens or taking into account past average purchase frequency, but it doesn’t seem like the most-sophisticated form of recommendation.
Granted, I normally make it a point to limit Amazon’s data-gathering. I browse logged out, make a list of what I want to buy, clear browser state, and log in only long enough to make a purchase. That probably makes it harder for Amazon to associate me with my browsing behavior. But it does know what I actually buy. And it has a pretty substantial history there.
And for Steam, Valve knows what games I play, how long I’ve played them for, and assuming that there’s any mining based on game achievements, even – at least as an abstract concept that would permit for correlating preference across video games – what I do in those games. Like, players who get “evil path” achievements in one game maybe prefer video games with “evil” routes, stuff like that. But I have browsed Steam’s discovery queue zillions of times, and while I’ve probably found a game or two on there, the success rate of its recommendations is abysmally low. Probably the most-useful recommendations system on Steam is the “similar games” section when viewing information about a game. But I’m pretty sure that most games I find on Steam that I actually like are just by using user ratings and searching for tags. While, Steam’s scoring is opaque, and it’s possible that they’re using some degree of input, I don’t think that it’s making use of information about me there. I wouldn’t be surprised if it’s nothing more than ranking games based on their player review score, which…isn’t much more than things like MetaCritic and similar have done. I’ve occasionally had luck looking for games that have very high hours played, with the idea that people wouldn’t play a game a lot if they didn’t like it. That makes some use of aggregate data about users, but not about me.
Most video games that I get on Steam that I like are games that I’ve discovered somewhere other than on Steam, often looking for human “roundup” articles comparing collections of similar video games and giving a brief blurb about pros and cons. That’s not new technology.
That comes as a very great surprise to me, when one considers the enormous amount of effort and resources that goes into harvesting and mining data about people. Now, okay, a lot of that is for ads. And advertising isn’t exactly the same thing as doing good product recommendation. An advertisement is trying to effectively get someone to buy a product regardless of whether they’ll ultimately like it or not, whereas a product recommendation – at least in the ideal, user-focused sense – is trying to find products that people will like. But there has to be a substantial amount of overlap between the two. Advertisers don’t want to waste money advertising to people who won’t buy their product, so trying to find people who are interested in their product is a major part of advertising.
I haven’t used any systems that log my music-playing and make recommendations; I’d rather keep my privacy there. Perhaps if I did, that area would be more-successful.
But by and large, it’s an area that I’m very surprised is not more successful than it is. It’s a “flying cars and jetpacks” thing, something that I’d always vaguely expected of the future, but which never seemed to really arrive. Product recommendation systems never really got to the point of anticipating my needs very effectively, even where they have what I’d consider a fair amount of data to work with.
What’s your experience? Does it differ from my own? Do you find that product recommendations from vendors are really useful, pretty much hit the nail on the head for what you want? How do you “find” products? Am I missing something, maybe like merchants on Amazon or publishers on Steam trying to game the recommendations system one way or another, and poisoning its inputs?
Something to keep in mind is that the top priority for the product that’s put in front of you isn’t what you want, it’s what the seller wants you to buy. It’s a high margin item, a vendor paid a premium for visibility, it needs to move so warehouse space can be cleared, etc. This goes back to the brick-and-mortar retailer days. If a product recommendation algorithm is a valuable service for you but ultimately isn’t more profitable for the retailer than putting their finger on the scale, it doesn’t make sense for them to play it straight. What they can do is determine you’re more interested in doodads than widgets, and show you more of the doodads. Which doodads get shown at the top isn’t 100% based on your preference.
Recommendations or reviews from writers/critics that have similar tastes and unpaid actors are how I find most products. This was one of the most valuable functions of Reddit, and it’s one of my primary motivations for helping to grow Lemmy.
It must work better for people who buy tons of disparate shit all the time. Otherwise why would they keep the feature at all?
I’m a pretty consistent buyer. There’s recurring essentials (maybe the brand of dish soap changes sometimes, but it’s essentially the same). There’s one-off larger ticket products. What do you recommend to someone like that?
I’m not exactly sure. I was hoping for something innovative and/or fun that fits the types of things I already own. But I don’t get that.
I get frequently asked by Amazon if I need another:
- large TV
- bidet
- xbox controller
- kitchen knife
I don’t know who goes through multiples of those items during their normal life. I’d expect to be shown something brand new or something consumable.
Maybe I’m doing consumerism wrong?
You just bought a nice new router. Here’s some more routers you might enjoy, in case you want to, you know…route some extra stuff…?
heyy, this guy right here just bought a pair of shoes, they’ve finally caved and decided to become a sneaker collector, you know what they’d definitely need? the same pair, they love these ones
I feel the same way about Steam. My discovery queue is useless. Similar games section is decent, but it tends to be the same few games I’m already aware of. Just randomly browsing has been the best way to find new things.
Product recommendations are skewed by paid placement and fake reviews. It’s similar to what’s happened with search engine results.
Am I missing something, maybe like merchants on Amazon or publishers on Steam trying to game the recommendations system one way or another, and poisoning its inputs?
It’s this. Monopoly and oligopoly players have little incentive to provide strong measures against data poisoning.
Amazon has been losing ground steadily against review data poisoning for many years, but (presumably) hasn’t seen a loss of profit, over it.
The elephant in the room is that “data” is not a solution to this in any meaningful way.
It’s easy to use data to recommend products that are similar to the ones you’ve enjoyed. But we don’t need algorithms for that. We don’t even need computers for that. Follow the genres/categories you know you like.
The subtle and mysterious things that are unique to each of us that cause us to like one game but hate another game that’s objectively similar to it, data won’t solve that because we don’t even understand it well.
Another reason recommendation algorithms don’t work well is because they’re designed to maximize profit, not satisfaction.
Removed by mod
Spotify and Pandora have done pretty good for finding me new music. It’s about the only thing that has worked in any useful way.
YouTube isn’t bad, but it feels like it updates the recommendations too fast. If I watch one or two videos of something random that got stuck in my head, I get disproportionate recommendations for the same thing, as opposed to things more consistent with my long term viewing.
Trade (coffee subscription) lists 3 flavor descriptors with every coffee, and I remove so maybe from my recommendations with the same terms, yet I still get them recommended. I end up just searching the new arrivals until something strikes my fancy. They’re tied with Amazon for most useless algo.
I’ve pretty much only used Shop to find niche furry stuff, so the recommendations are actually pretty spot on for that, since it’s all niche furry stuff lol
People talk hella shit about YouTube’s recommendations but I’ve never found it to be off what I generally like. It’s more of the same. Sometimes in eerie detail. Like how even the voice of the narrator can sound exactly like stuff I’ve subscribed to, but isn’t the same person.
Im glad that the youtube algo works for you lol. It keeps trying to get me to subscribe to manosphere channels and im just sick of it. Who knew that a little andrew shithead tate would be what cured my youtube addiction xd
Yeah, it’s such a shame. This should be the silver lining of ubiquitous data harvesting.
The problem is that they aren’t really serving individual consumers - the main customer for these big attention merchants is other businesses. Amazon and Steam et al could focus more on helping you surface great products you love, but they undercut that by helping sellers get visibility they want (almost regardless of product quality).
It’s certainly not simple still. They’re not incentivized fully to serve your interest, and even if smaller companies are, it’s hard.
Virtually the only data that Amazon has about you is when you’re browsing through Amazon, and its associated webstores, and there are a lot of those.
Possibly more if you’re a regular Twitch viewer, or watch video on Amazon Prime.
Also if you’re stupid enough to have an Alexa, which listens to everything inside your household says, Amazon doesn’t dominate the tracking cookie/consumer surveillance market.
Google has a much better ecosystem of consumer surveillance, by owning products and services that are indispensable or unavoidable parts of internet infrastructure.
It eavesdrops on your Android phone, tracks all your web browsing through its Doubleclick cookie network, all the videos that you watch on YouTube.
The world’s most popular browser? Chrome, also owned by Google.
I wonder if these platforms are doing that on purpose. So you don’t feel like in the tech dystopia you’re actually living in. If the algorithms are good, they could strike some balance. Mix up the recommendations so you don’t feel that vulnerable.
And I mean another thing is advertisements. Usually some of the recommendations are sponsored, or advertisements. And they’ll show stuff to you because someone paid them. Not because it maches well.
My experience is similar. But I try not to volunteer all my data. I opt out from targeted advertising, install browser addons to stop tracking mechanisms… I always thought it was just me who gets random recommendations in return.
Or maybe the algorithms just aren’t as good as we think. I’m not sure. Spotify for example can’t figure out at all what kind of music I like. I’ve been using it for quite a while but it recommends lots of pop songs anyways and I have to skip all the time. YouTube on the other hand does a half decent job in recommending videos that I click on and watch.
So you don’t feel like in the tech dystopia you’re actually living in.
What’s hilarious about that is when the algorithm is spot on to recommend someone something they’re thinking or talking about they usually freak. the. fuck. out.
They immediately go full conspiracy-brain that Bezos must have broken into their house last night and used a MRI machine to read their dreams, because how else could he know that they needed new dish towels.
I 100% wouldn’t be surprised if they skewed the accuracy if for no other reason than 100% accuracy would just cause a huge internet drama storm of freaked out people wondering how the hell Amazon could possibly know what they needed. (See: the Target coupon thing that outed a pregnant kid because they knew before her family did.)
