IIRC in the US, if you miss the payoff date, they apply interest all the way back to when you purchased the item. And it ain’t cheap.
Is this a better deal? (I have not been able to figure it out yet.)
$4000 per customer is an extraordinary amount to pay. No matter what the market, how is it possible to recoup that value from a single person’s data and future business?
Afterpay is the big dog in the game here in Oz. They pretty much built the market and are the default provider. So much so that “afterpay” is becoming a verb.
Being a straight stock transaction makes the value a bit arbitrary. It is perhaps better understood as a relative valuation against the purchasing company. That stock is probably also vastly overvalued.
Well, Wikipedia says, “As of 28 February 2020, Afterpay was recorded to have 3.6 million active customers in the US, 3.1 million active customers in Australia and New Zealand, and 600,000 active customers in the UK.” I don’t know how many customers Square has right now but I’m guessing it’s a whole lot more. So they can start by offering Afterpay to all of their existing customers.
Yes, you have to qualify for it just as you would any bank loan. You fill in an online application form in the store, they run a credit check on you and approve or not pretty much on the spot.
In Australia Afterpay has no access to credit reference facilities and does no credit checking. They check your identity to ensure that you do not create multiple accounts but it is up to consumers to ensure that they can afford the service. If you are over 18, have a valid email address, a valid mobile phone number and a valid debit or credit card you are automatically approved.
On their website they say:
Afterpay does not affect your credit score or credit rating. Your credit score can be impacted when somebody does a credit check on you or if you are reported as paying debts late; at Afterpay, we never do credit checks or report late payments.
It’s installment payments for smaller $ purchases. Looks like the default new user limit is only $500 & the default is four payments over six weeks, with the first being made at time of purchase. No cost if you pay on time but up to 25% late payment fee if your late, & they also cut you off from making further purchases until you’re caught up so it looks like if you’re late with a payment for a purchase to Store A, you can’t buy something with them from Store B. They also reserve the right to reduce your limit if you were late.
In the UK at least, Credit Cards are in the same market. Both Mastercard and Visa regularly offer me x months (often 12 or more) of free credit.
My daughter bought her last car on a 15-month deal and avoided the steep interest charge by setting up a standing order to pay it off in 14 months. Missing a payment is a real ball-breaker as I believe that this makes you liable for all the accumulated interest from day one, not just the outstanding debt.
Since the debt gets added to your regular debt on the card, there used to be a catch that any payments went towards regular debt first and not the delayed debt, but I think that has been ruled unfair these days. DD left the card in a drawer for the whole period, just to be safe.
Then why are my YouTube ads at least 50% stuff I have no conceivable interest in like business insurance or production software, or automobiles. I do not own a business and the last new automobile I bought was a VW Beetle in 1969 fir $1,800.
Sometimes it comes up with stuff you have no interest in buying because you’ve clicked on something at some point that triggered that ad. Even, for example, if a few months ago you were writing something here and needed to know the price of a new Ford Escape. If you google it, Google now thinks you may be in the market for a new car.
Same with facebook. Even accidentally clicking on something can make FB think you want to buy it.
And the algorithms aren’t perfect. For example, why does Amazon think that just because I bought a garbage can I need to have ads for garbage cans on my front page for the next month. I bought it, I don’t need another one.
And, lets not forget that if you have an Android phone (not sure if this applies to iPhones) and your GPS (or in some cases wifi) is up and running, Google knows where you are. Maybe you spent 20 minutes within a close enough distance to a car dealership that Google considers that you may have been car shopping.
Also, the more things you do to keep various big tech companies out of the loop by preventing them from knowing what you’re looking at/buying, the more random your ads will be. So that’s always a possibility as well. VPN and duckduckgo and private browsing and disabling tracking cookies and anything else along those lines and they have less info about you, but that doesn’t mean you’re not going to get ads, just not ones that are targeted to you.
Because what is being optimized for is not specifically ads that are relevant to the user. What’s going on under the hood is that different companies bid a certain amount of money to run on different categories of people, and Google goes with the one that pays the most money. So it’s very possible that a company that is less relevant is willing to pay more. You have a bunch of competing systems trying to figure out the best way to advertise, rather than one overarching algorithm that can narrow it down more directly.
If you’re like most people, you also don’t tend to give a lot of feedback. You don’t click on the vast majority of ads, and the whole “mark ads you don’t like” concept just tends to lead to people marking all ads. So there’s no real direct training.
This means that what data they have is imperfect. However, it’s still better than random chance at picking something you would like, and that data is worth money. If you can go from 5% of people you advertise to click on your ad to 20%, that’s still a huge difference, even though it means most people who see them still don’t click on your ads.
And, yes, there’s the issue of putting AI on the problem, and how much random noise those often wind up with. The same reason watching one video can cause all your YouTube recommended videos to be related to that video can result in them linking related ads. AI is basically a drunk man stumbling around in the dark. The difference is that it can do it faster and thus winds up often being more accurate when it is accurate. But it’s also often more wrong when it’s inaccurate.
Heck, AI is usually designed to intentionally try things that it doesn’t think will work just to make sure it doesn’t settle on a system that is good enough but could be better. You don’t want a local maximum.
Plus, what actually gets the most money can be counterintuitive. Maybe having some ads that are bad actually causes people to be more likely to click on the better ones.
Well, I don’t do the social media thing – except here – so there’s not much feedback from that direction but when one of the pre-play questionaires – Which of the following would you consider buying? – come up I answer them.
Has had no effect that I see.
Sidenote: If you want to watch a video and don’t want your related videos to be flooded with suggestions based on it for the next two weeks, watch it in a private window (and check the volume, it’ll be at 100%).
I find I more often don’t really think of it ahead of time, so I just go to my YouTube History and delete it. On the main site, it’s in the sidebar literally under the word Library. (You may have to click the button in the upper left to see the sidebar.) In the app, it’s just listed in the Library tab.
This is also useful if you start getting bad recommendations, and know which video caused it.