Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Amazon Changed Search Algorithm in Ways That Boost Its Own Products

wsj.com

Post reply on HN
Print view View original

141–150 of 438 posts

Re: Amazon Changed Search Algorithm in Ways That Boost Its Own Products

#141

Post by noego »

The article is disappointingly misleading and buries one of the key details

"Amazon’s lawyers rejected the overt addition of contribution profit into the algorithm...

They turned to the metrics Amazon uses to test the algorithm’s success in reaching certain business objectives, said the people who worked on the project.

When engineers test new variables in the algorithm, Amazon gauges the results against a handful of metrics. Among these metrics: unit sales of listings and the dollar value of orders for listings. Positive results for the metrics correlated with high customer satisfaction and helped determine the ranking of listings a search presented to the customer.

Now, engineers would need to consider another metric—improving profitability—said the people who worked on the project. Variables added to the algorithm would essentially become what one of these people called “proxies” for profit: The variables would correlate with improved profitability for Amazon, but an outside observer might not be able to tell that. The variables could also inherently be good for the customer."

Tldr: some people in Amazon wanted to give a boost to Amazon-products. The search team fought them vociferously and refused to budge. The lawyers came out against it as well. Amazon's internal A/B testing framework measures a number of metrics, including both revenue and profit, when determining whether a specific feature/change should be deployed.

The fact that the A/B testing framework measures the profit impact of any change, is hardly earth shattering. This is one of the core features that any A/B testing framework attempts to accomplish.

This also doesn't tell you anything about what the search algorithm is actually doing. An A/B testing framework can only help you evaluate the relative effectiveness of different algorithms. It doesn't actually create/influence the algorithm in any way. As WSJ themselves reported, Amazon's search algorithm does not take profitability as an input.

WSJ has done a fantastic job in unearthing this very interesting internal-debate that's happening in Amazon. But they have reported it in a way that is very misleading and gives laypeople the impression that they have a smoking gun. In reality, the only thing they have produced is the fact that Amazon's A/B testing framework is profit-aware.

Re: Amazon Changed Search Algorithm in Ways That Boost Its Own Products

#142

Post by toasterlovin »

m_ke wrote: post #72
namdnay wrote: post #58

Earlier quoted context omitted.

Well yes and no. Most grocery chains have more than enough weight to make sure they can return whatever they can’t sell. And the pay their suppliers so late that 99% of the time they’ve already made the sale before even paying for the inventory

The product is still taking up their limited shelf space so it's in their best interest to see it get moved. Amazon made their business on the long tail of products that rarely get bought but in aggregate add up to a large portion of total sales because they don't have the same physical constraints.

Amazon also has an equivalent of shelf space: search results. They are just as incentivized to maximize revenue per customer search as super markets are incentivized to maximize revenue per shelf space.

Re: Amazon Changed Search Algorithm in Ways That Boost Its Own Products

#143

Post by toasterlovin »

learc83 wrote: post #8
monitron wrote: post #6

I would compare this to a supermarket allocating shelf space to maximize its profits, either by placing its own brands front and center or by featuring brands that have made a deal for prime shelf placement. I don't see this as scandalous or even new. Coincidentally, Amazon brand products are usually pretty good in my experience, so this might not be a terrible outcome for users who have already decided to come shop…

Grocery stores don't operate as a "marketplace" though. Grocery stores curate their offerings, vet vendors, deal with returns etc... Amazon is more like shopping mall that decides to start operating their own stores and competing with their tenants.

Malls curate the stores that they lease space to. That's why some malls are very high end and others are not.

Re: Amazon Changed Search Algorithm in Ways That Boost Its Own Products

#144

Post by scarface74 »

justinmchase wrote: post #102
monitron wrote: post #6

I would compare this to a supermarket allocating shelf space to maximize its profits, either by placing its own brands front and center or by featuring brands that have made a deal for prime shelf placement. I don't see this as scandalous or even new. Coincidentally, Amazon brand products are usually pretty good in my experience, so this might not be a terrible outcome for users who have already decided to come shop…

It should be scandalous in those cases too. Having a company be both the distributor and producer and have a monopoly is problematic. This is sort of the problem of cable TV also, where companies like comcast tries to be a producer and distriutor and have a monopoly. In the past we have had unions fight against this sort of thing but unions have been busted to hell in this country over the last few decades to the poi…

Your argument may have held weight in the past, but now, anyone can distribute video without going through the cable providers regular cable distribution. It’s quite simple technically to provide on demand video at scale, including creating apps for all the platforms. The only issue now is when/if cable companies as internet providers start deprioritizing competitors.

Re: Amazon Changed Search Algorithm in Ways That Boost Its Own Products

#145

Post by jhallenworld »

Pxtl wrote: post #34

Considering how shopping at Amazon now feels like shopping at an electronics bazaar in Singapore with giant bins of random knock-off products of suspicious quality, tweaking the algo to push name-brand options (even if it's their own name-brand) would be a welcome move to me as a buyer. Obviously it's grossly unfair to their vendors, but from a strictly user-centric view it's an improvement. Otherwise, Amazon feels l…

They do care about the quality of their own products. I posted a negative review on an Amazon product, and they called me(!) so that I could further elaborate. It was very odd because they called within an hour of the review. At some point I must have given them my phone number.

This was a review for their TV Cube. I was trying to buy something with good voice control to allow a paralyzed person to browse YouTube, and TV Cube could not do it (maybe it will be better in the future..).

I suspect the only reason I got the call is that TV Cube is in active development.

Re: Amazon Changed Search Algorithm in Ways That Boost Its Own Products

#146

Post by nokcha »

alistairSH wrote: post #134
jschwartzi wrote: post #111

Earlier quoted context omitted.

With a power-supply it's not a hard problem to avoid. You have to hire independent engineers to evaluate the product design and establish testing procedures to verify that the product isn't unsafe. And you have to do QA and design verification during receiving on all batches of product to confirm that the manufacturer followed the design. I've never had a power supply catch fire and I've done some really dumb stuff i…

How is a power bank different than a battery? Isn't the power bank just several batteries crammed in a casing, with a power-out port?

jschwartzi was talking about power supplies (i.e., devices that transform mains AC power to DC power), not power banks.

Re: Amazon Changed Search Algorithm in Ways That Boost Its Own Products

#147

Post by rdl »

alistairSH wrote: post #134
jschwartzi wrote: post #111

Earlier quoted context omitted.

With a power-supply it's not a hard problem to avoid. You have to hire independent engineers to evaluate the product design and establish testing procedures to verify that the product isn't unsafe. And you have to do QA and design verification during receiving on all batches of product to confirm that the manufacturer followed the design. I've never had a power supply catch fire and I've done some really dumb stuff i…

How is a power bank different than a battery? Isn't the power bank just several batteries crammed in a casing, with a power-out port?

Most modern rechargeable batteries have fairly sophisticated chemistry and thus charge controllers needed -- you don't just dump electricity into them blindly and then take it out. It's not quite rocket surgery, but there's a reasonable level of engineering involved, especially if you want optimal performance with the newer standards (which increase voltage on devices which support it, etc.)

Re: Amazon Changed Search Algorithm in Ways That Boost Its Own Products

#148

Post by toasterlovin »

pixelpoet wrote: post #140

ANTITRUST! If Microsoft can get hammered for packaging Internet Explorer with Windows (and probably rightly so, given the market conditions at the time), there is a direct analogy to be made here. Pick what you want to be: if you want to be the greatest goods index and shipment company in the world, fantastic. But you have to accept that you cannot sneak some of your own other-products to the top using that power. In…

No, there is not a direct analogy because Amazon is not a monopoly. They aren't even the largest retailer.

Re: Amazon Changed Search Algorithm in Ways That Boost Its Own Products

#149

Post by DangerousPie »

ikeboy wrote: post #70
jastanton wrote: post #56

Earlier quoted context omitted.

The one difference is that the Amazon Basics products seem to have a quality bar higher than the other random re-labeled products.

They recalled a quarter million power banks because they were catching fire. Everybody uses the same factories and Amazon doesn't have better quality control than anyone else.

At least they recalled them! I don't think random no-name brand #12345 would have done that.

Re: Amazon Changed Search Algorithm in Ways That Boost Its Own Products

#150

Post by momokoko »

VRay wrote: post #118
momokoko wrote: post #104

Earlier quoted context omitted.

At this point, anecdotally, it feels as though the top comment on almost every article I see on HN about Amazon is some variation of this complaint. Has anyone done any analysis on the HN posting dataset to see if that happens to be the case? And if that is the case, does anyone have any insight into why that might be?

Amazon used to have the best service, best quality products, and best prices, so practically every HN user loved them. Now they're wavering on all three of those fronts, and I guess that fact really bothers most of us

I'm not disagreeing with the premise that Amazon has issues. But I do not see a similar pattern with other companies and services. Of course, this is anecdotal, but it does seem significantly higher with Amazon itself.

It is possibly that most HN users purchase a lot via Amazon?

Is it a small but vocal group of HN users that have a grudge because they had an unsuccessful drop shipping operation that was harmed via Amazon?

Is there some competitor(ie. Walmart) that is trying to push an agenda?

It just seems high compared to other services, so I'm curious into additional insight into why Amazon, specifically, has this pattern when other services with issues do not.

I personally feel like it may just be that most HN users are, or were, heavy purchasers from Amazon. But I'm just looking to see if maybe there is a more insightful take on things.

Post reply on HN