I despise their horrible advertising and marketing copy in general and I sometimes just have to tell myself, “it’s just oat milk no need to boycott it because some edgy marketing consultant forces you to read their nonsense”.
So no... I stand by what I said, I think they did a lot of things wrong in the marketing department, in spite of having a really solid product.
Just like random "coca-cola" billboards at the side of a road or a sports stadium
The market for oat milk consumers mostly consists of people that had actively made a choice about drinking oat milk already.
Millennial advertising doesn’t adhere to 20th century advertising norms.
Gen-X is getting older and needs fiber and to lose weight, so they respond to “oats”.
Millennials and Gen-Z want healthy-looking drinks that are trendy and fun, because they’ll never be able to pay off their college loans and grew up during a time where politics were divided and assume we may all die from global warming and WW3 so they just want to be healthy and happy.
This ad is trying to say, “We’re not going to say this is healthy, or will make you feel free, or is a fast meal substitute. It’s just oat milk.”
That pretty much appeals to all of them.
CPA pricing removes the burn
Not true. There is limited ad space which all advertisers compete for. No matter which model, CPC or CPA, the advertiser who pays most gets the ad placement.Its similar to SEO. Nobody says "Oh my god, advertising via SEO is free, what a blessing!". It's still a competition. It's still a zero sum game.
Manufacturers are in a fundamental conflict with Amazon precisely because they desire to fully control the retail channels and set their own promotions, online discounts etc. and capture most of the surplus themselves while still segmenting the market and, for example, selling at different prices through certain local distributors.
Amazon has the exact opposite incentives, they want distributors of the same brand to compete amongst themselves so they can offer the lowest global prices, and that it's Amazon and its users that capture most of the surplus.
This is the root of the forgery problem Amazon can't solve, manufacturers aren't willing to vouch for their products when sold in secondary channels they do not fully control. So this means they will not collaborate on the "global rating" scheme either.
Has chosen not to solve. They could trivially improve the situation by ending the practice of commingled inventory so a seller could be held accountable for counterfeit or stolen items but that would cost more so they don’t.
So if they can't be bothered to attach a zero cents sticker on the item, a 5cents rfid tag is out of the question.
If it’s both fulfilled and sold by Amazon, I have yet to have anyone actually provide an actual first person story that their item came from commingled or forged stock.
Yes, I’m sure it has happened. It’s not a widespread thing. Amazon does not sell Tide detergent sourced from some random third party that sent in stock matching the SKU being sold because it happens to be in a closer warehouse.
Random listings on Amazon? Sure I’ve received fake stuff. That’s the risk of using a third party storefront.
I have also not heard or seen of a well sourced story recently where someone’s esoteric custom product SKU was being sent in by third parties as the same ASIN and comingled with the legit companies listing.
The shady listings seem to be where this all comes from to begin with. Amazon could cull those nearly overnight but chooses not to. This is where the main problem lies. The times you click the legit listing for the FBA/SBA Amazon item, and then there are 12 “options” like multi-packs that are random third parties trying to take advantage of the unaware type of dark pattern.
It helps with inventory availability with the cost of risk of bad customer experience in case of fraud.
Personally I hate it, I would love to buy directly from the manufacturer instead of playing roulette with the distributor.
One big problem I see is if you go to the storefront of some seller there isn't a way to see all their items. I find this baffling, and a clear red flag. Just start clicking on the stores when you're poking around next time. It's very common for things like pet toy sellers or even electronic component retailers. I've seen even certain colors of products be listed on a different page and sold for much cheaper.
There's also the classic example where a page was originally for one product but now has another. You look at the reviews and they're talking about a different product, sometimes related sometimes drastically different. This is actually a great use case for some ML. Flag products if a classifier says the item is significantly different than the previous photos. Could use some basic NLP to check product titles and see if they match. It creates some issues but Amazon has such a big problem that it would be worth it for the customer experience.
But I think the problem is all of this doesn't "make more profits." If it's hard to verify a product a user will spend more time on Amazon. If the idea is that more time increases likelihood of buying more things then a little fraud is good for them. Worse for the customer but better for them. I'm not sure how you solve this problem
[0] https://www.reddit.com/r/fountainpens/comments/q14p5w/are_la...
Or perhaps, if it's infrequent, I just ignored it as obvious bullshit. I only did one order off them experimentally last year.
My gut suspects not?
Perhaps in the early and mid 80s you could still buy quality products, but now is seems 99% of things are just mass-produced where ever it is cheapest. People are conditioned on Amazon to find the same product from a jumble-of-letter manufacturer who is selling the exact same thing at the lowest price. I do not trust that if I buy a "known brand" for a product that it is going to be any different from a similar same no-name thing that is 20-30% of the price (...and very possibly built in the same factory). If it's all low quality crap (which a lot of the time it is) then you may as well get the cheapest one
Sadly you need to rely on things like YouTube videos to actually get any kind of idea on if the item is trash or not, and even then there is the risk of paid-reviews so you need to take multiple sources into account, who they are, trust levels etc. it's sad. Either that or - and I know this is madness - go to a physical store and inspect the goods before you buy it.
I would argue that the article is correct that quality is often secondary to speed of delivery and cheapness though. Amazon has totally won there.
Yes I agree that there are better products for more money, it's just that I have to filter by something else because I can't trust that the price corresponds to performance for any given listing.
I don't believe you. Link? That extra $900 gets a iot more lumens and portability and CPU and ports and such that the $100 ones I've seen just don't have.
The point is that if you go to a store or Amazon, you can find crap products at any price point, so much so that you cannot look at the price to get any information about quality.
> My gut suspects not?
You are correct that price is no longer an indicator of quality.
Look at this stick vacuum review, for example: https://vacuumwars.com/vacuum-wars-best-cordless-vacuums/#1
A $374 vacuum ranks at #4 while a $1049 vacuum ranks at #8. That is a HUGE price gap for items that are supposed to be of similar quality. The reason for the gap is irrelevant to this discussion too, because we are talking about the relation between price and quality.
And people try to blame the consumer for both sides of the issue.
People shrug and say it is the consumer who demands cheaper products and that is why they are of such low quality (despite what you say, which is that price and quality are much less connected than what people selling you things want you to believe), but then when those same consumers initiate returns, people cry foul like the consumer is somehow maliciously buying shit quality products and returning them to spite honest companies who just want to do right by their customers.
Look at how this article is written: https://www.nbcnews.com/news/us-news/retail-returns-890-bill...
Notice anything in that "Why returns are a big problem" section? A neat rhetorical trick which attempts to get inattentive readers to believe that the bulk of all returns are just people trying on clothes and returning the ones they don't like. Now, if it really were the bulk of all returns, they would have surely mentioned this, because it would buttress their point.
They did not, because they can not, because, I have a suspicion that the $890 billion worth of returns is primarily due to reverse/negative demand for shit products. I suspect those return numbers creeping up year over year are consumers getting annoyed with retailers for not curating products in a way that considers quality and longevity.
As the featured article mentions, Amazon does mark products as frequently returned, but I am not sure it really helps that much.
> Sadly you need to rely on things like YouTube videos to actually get any kind of idea on if the item is trash or not, and even then there is the risk of paid-reviews so you need to take multiple sources into account, who they are, trust levels etc. it's sad. Either that or - and I know this is madness - go to a physical store and inspect the goods before you buy it.
And even then, they miss important details, because most reviews only talk about the product in terms of the out-of-box experience. But how good is the product after two years? After four? Even the vacuum review I linked does not really comment on longevity.
> I would argue that the article is correct that quality is often secondary to speed of delivery and cheapness though. Amazon has totally won there.
I think it is starting to change where people are getting really tired of shit quality and that is going to continue to be reflected in the returns data. Returns do not make customers whole - they do not get a refund for the time they had to spend researching, procuring, and ultimately returning the product. I imagine companies are going to be in for a rude awakening when they have to reverse course and not only increase quality but also decrease the price because no one wants to spend hard-earned money on expensive junk during a recession.
If this was really the case then Amazon would not have a "frequently returned" concept as this would penalize sellers doing nothing wrong.
"frequently returned" warning makes only sense when the majority of returns are due to lack of quality control.
- use other information channels like review media (fashion sites)?
- trust secondary sites like brand retailers (IE. John Lewis in the UK?)
????
Not so sure now they are becoming just another drop-shipper but there are few alternatives in the UK.
Not that I see much ads any more anyway, what with Ublock Origin, Blokada 5 and Youtube ReVanced.
With a population of ~300 Million the US ad industry generates 400 Billion in revenue. Avg Bounty of >1000 bucks a year per person.
The only way to pull that off is to keep injecting more and more ads everywhere and make everything more tempting to watch.
The biggest threat to these companies is if time spent consuming media drops.
The same question, who is paying for all the data collection they do on customers?
You keep hearing about xxx billion dollar industries but where is that money coming from? The add and data collection industry in the US is probably a trillion dollar market or more, but ... does it really work? Sure, focused ads on some youtube channel about product that may be of interest, but all the rest is such a mess.
I do not buy more stuff because some ads showed up. Sure, they can be informative that there is a new car released, or something like that. But there is no desire to buy XXX cleaning product because they spammed it on TV.
Same argument with "what use is getting all my personal data, down to the things we do in bed". Sell it to specific companies? But what do they really gain from it. It does not feel like we are getting better products as a result of all that user data. O, people are not happy about Y product. You do not need invasive data collection for that, you see it in your revenue numbers.
It often feels like a lot of industries are just there to self sustain themselves. Nobody ever got fired for buying advertisement or personalized data. So companies keep spending on it, people keep pouring over the data, analyzing it, but its like nobody has common sense and its all a scam that inner feed itself.
I am not saying that advertisement does not work to grow a product, but especially unknow one's but the moment product reach a specific mass...
The big threat to the whole system is when the amount of time per day people spend consuming media starts to drop thanks to fatigue and finding better things to do.
All you need is two things:
1) a zero-sum game where the amount spent by companies on advertising is less than their profit margin
2) advertising channels that perform in a way that can be correlated with short-term outcomes
"We can afford to up our ad spend by 0.5% to try to gain an edge."
You juice your market share or total sales numbers for a quarter, everyone is happy.
Then your competitor does the same.
Assuming your advertising is equally effective, you may have netted out to about where you started. But you've both ratcheted up your spend and likely won't reduce that short of a macro change or parent company health issues.
And if certain advertising performs better than others, one company can very well make long-term gains here, if they have writers and producers and talent who are making more compelling pitches.
But it still ramps up the total advertising industry spend even with a fixed supply of ad capacity, if it gets a bidding war going.
As long as you don't have 100% market share with 0 competitors, advertising will look like a very compelling short-term ROI option.
Even if time spent watching media drops it may not squeeze a lot of parts of the advertising sector so hard - it'll just push demand into smaller buckets which will increase the prices for those buckets. Look at the crazy revenue sports television broadcast advertising has been doing as other forms of TV go away with commercial skipping or lower-ad-load streaming.
Of course, at SOME point the increases can't be sustained. Even the most profitable products have a limit to how much money could be possibly directed towards advertising. But we don't appear to be particularly close to demand being satisfied.
If any single company spends more, and succeeds in gaining advantage purely by spending more, all their competitors will generally just be forced to follow.
But, no one really wants to follow that to the logical conclusion of spending all possible dollars on marketing.
Sure, just like nobody wants to spend all possible dollars on rent.
But outside of "we just can't afford it anymore" external economic factors shocking the system, the natural incentives point to a slow and constant ratcheting up.
IMO this would be a good reason to have further regulations on advertising - locations, frequency, accuracy - since we're gonna be saturated with it otherwise.
I'm sure you've noticed everyone raising their prices together, or fast food restaurants all halting 24 hour service at around the same time.
People often operate based on perceived industry norms/trends, rather than some sort of rational economic analysis.
They pay the same, I will keep buying whatever my friends suggest me to buy. Win - Win
"When advertising, appealing to emotion is much more important than information or logic."
Last year I was skimming through IKEA furniture and majority of reviews are 4 or 5 stars. And that makes me think that people who don't like IKEA products are either so pissed that they do not want leave 1 star review or IKEA removes bad reviews.
The good ol' 7 out of 10 problem. Where the entire lower half of the rating system is unused.
The way I solve it is to have 0 be the middle, If the transaction exceeded expectations you can give it a +1 if it failed to meet them give it a -1 (thumbs up/down if you prefer) you can even throw in a +-2 if you want to allow for more subtlety of expression
Customer reviews are also useless. 10 people go 5/5 no comment. One guy goes 3/5 "good thing." ???? Another person didn't hold it right and rated it 1/5, another person got a defective product customer service would have replaced if they had reached out and rated it 1/5.
IMO the solution is doing actual due dilligence. What is the spec sheet? What is actually relevant to what I am doing? How was this thing made? What are the tradeoffs of using this technique in in this product vs other techniques? Where does this featureset stand against other offerings in this segment and pricepoint?
You can do all of this yourself from the primary materials, the product pages, etc. It also doesn't take longer than you'd spend agonizing over yet another half dozen 20 minute review videos. And you'd end up better informed than most content producers, most customer reviewers, even most redditors on the relevant subreddit.
You operate along these lines and all 5 friction points mentioned in the article become irrelevant. You don't give a shit about the brand, you care about the merits of the component. Prolific advertisement is no longer required as a signal for quality since you are actually evaluating the merits of the component directly and not using proxy signals. Return status also doesn't matter because you are evaluating the merits of the component directly. Ratings don't matter. Pricepoint also doesn't matter because you don't include price in your analysis save for comparing capabilities across a price point vs assuming costly product = better.
And of course maybe you are bad at estimating a components merit. Maybe you are suffering from Dunning Krueger effect. For most things in life this hardly also matters because the effective difference between a most optimal product and one that is merely good enough is basically zero especially when you lack the ability to determine the differences between perfect and good enough (probably means that precision is not required for you use case).
Unless you can't. What exact Ethernet card is embedded in your extensive brand port extension hub? No way to know unless someone knowledgeable disassembles it. Valuable information doesn't just float around one google search array. Product pages are too often the same kind of garbage like the reviews you've described, containing very little useful info
soared•6mo ago
neehao•6mo ago
neehao•6mo ago
gsf_emergency_2•6mo ago