Hacker News Reader

Comment by BitwiseFool | Hacker News Reader

I invite people to checkout BestBuy's online shop. Just about every single item has a rating from 4.0 to 4.8 out of 5. I have yet to see a single listing under 4 stars. There are of course, unrated items, but can you honestly believe that virutally every product they carry is nearly perfect? Who actually writes a 5 star review for a no-name brand of tape and cable ties?

When everything is perfect, nothing is, and I don't trust BestBuy's review system at all. I can't help but think there is some kind of false advertising law against fake reviews that they're breaking.

View on HN

To answer your question about tape and cable ties, BestBuy will offer rewards points for writing reviews and sometimes offers bonus points for certain products.

Im sure they're also doing some review cleaning like Amazon does where if the review is negative they will moderate it more aggressively. After all, badly rated products hurt the entire storefront so it is in their best interest to keep "bad" products off.

But that said I still will trust BestBuy over Amazon. I know items from BestBuy are not being co-mingled with potential knockoffs and returns are pretty easy for items that are terrible.

In theory, they could actively remove bad items from the catalog, thus not having any lower than 4.0.

Yeah, I'm aware that is not what is happening there, but you seeing no item lower than 4.0 is not a certain and general sign of fraud.

I spend entirely too much money at Best Buy, and I’ve only written 5 star reviews for products I like. The bad products get immediately returned for a full refund - one of the advantages of buying from a big brick and mortar store. I never think about products I returned, I would prefer not to unless I’m given some incentive to write a bad review. Now, if I were stuck with the item, I’d probably trash it online in their review section.

I guess a useful metric would be if they posted return rates, warranty cases and so on.

This is one of the bonuses of a brick and mortar store if you can get a commissioned associate to tell you.

Ex: I’m not going to say phonefriend makes a bad phone but they have the highest return rate and people will buy a much more expensive Panasonic when they leave for a second time.

Same kind of experience here. On top of that, I'm usually a pretty picky purchaser who does a decent bit of market research ahead of time. So I'm probably more likely to only buy that good electronic item from a shop like Best Buy that I've got a high likelihood of giving a 4-5 star review.

I think the reality is not just that Best Buy is (maybe) creating fake reviews or removing bad ones, but that most reviewers only rate with 1 star or 5 stars. The majority of reviews I find are in one of those two buckets, and only a handful of nuanced reviews fall in 2-4, effectively making the rating of a non controversial product that does what it’s supposed to - but has lemons - from 4 (meh) to 5 (works as expected).

This is why I would support a law that says if you're selling a product and have reviews you must also report how many you've sold over the last year, and how many returns (or maybe a percentage or something).

That would be a much more useful metric.

It would be gamed anyway, you'd just get sellers saying "we'll give you twenty dollars for not returning the item."

Prosecute this as fraud.

Almost all online stores seem to be like this. Why would Dell or Lenovo show you products that had anything less than 3.5 stars? There’s a huge conflict of interest in showing reviews of their own products. BestBuy has to sell the inventory they carry, so sure, they will tell you everything is 4-5 stars. Amazon is actually slightly better because they profit no matter which headphones you buy, but then it’s the sellers that are gaming the system. In theory it should be easier to detect.

This is a problem with nearly every online rating place. 4.0 means TOTAL GARBAGE and 5.0 means it didn't burn by house down. Totally useless.

I gave an honest review of some $30 bluetooth headphones. I gave it a 3 star, and listed my complaints. The company reached out to me, offered to send me 2 more, for free, of their newer and better model, if I just adjusted my review to all stars. That was when I stopped believing them.

This is so common that I considered creating a browser extension to convert/normalize the the min and max values on a page into a 5-star rating. When the lowest rating of items on the page is 4.2, 4.2 would displayed as 1.

That is actually a really cool idea. Probably a lot more work than it sounds like.

and Doordash. how is everything >4? are we really to believe that McDonalds has a rating of 4.6 out of 5?

it makes the system completely useless.

If you order McDonalds, you probably know what McDonalds should taste like. If it meets 100% of your expectations, why would you rate it lower?

Similarly, if I get a meal from Michelin Star rated restaurant and it fails to meet my high expectations, should it get a better rating than McDonald's just because the restaurants meal was still likely much better than McDonald's even if it failed to meet expectations?

In practical terms it dilutes the value of 4 and 5 star reviews for other restaurants. There's some cognitive dissonance involved when McDonald's gets a 4.6/5 and a mom and pop haute-cuisine restaurant gets a 4.6/5 as well. These are two very different dining experiences but they are being judged using the same system and feels inaccurate.

For what McDonalds is they probably deliver a 5 star review the way their customers see it. You probably have to divide restaurants into different categories and compare within them. I beat a lot of people will rate MCDonalds higher than haute cuisine restaurants even in direct comparison. There is no absolute rating for restaurants.

If in direct comparison they rate McDonald's higher, then that's fine... McDonald's is a huge business, they obviously have people who like them.

But it should be a direct comparison, I certainly think KFC is far superior to most haute cuisine fried chicken.

There is no sensible way to compare fine dining to McDonald's.

The rating should be based on how well they deliver what they purport to deliver. If McDonald's delivers the absolute epitome of a fast-food burger and fries then it should get a 5/5, even if tastier food can be found at other restaurants.

Haute cuisine with a mediocre execution should get a 3/5 even if it's tastier than a McDonald's hamburger.

From a utility standpoint, I mean, yeah. You should definitely rate McDonalds lower than a restaurant with food you liked better, if people want to search for "how was the food" divided by "cost of the food" that's obviously a sliding scale that depends on the person. Maybe it should be something you can sort by, but I don't see how you could possibly reconcile ratings from different people if they all give five stars for any restaurant that meets their expectations.

Experience vs. expectations is the only reasonable way to interpret user-based scores even if they are not astroturfed. The problem is that you don't always know what the average user's expectations are and how they compare to your own and there is no basis for comparing scores between different objects.

This I think stems from a different issue. Doordash has a very easy, one second review they ask from everyone. And I think 99.999% of folks think “oh yeah that was fine, 4 stars” or “it was pretty good, five stars.” Do people get what they expected from that McDonald’s? Was it cheap, fast and nostalgic? 5 stars.

The people doing these reviews are not good critics, or critical with any sort of depth.

They are the sort of folks that get doordash to bring them McDonald’s.

This reminds of me leaving reviews on Uber or Lyft. A perfectly fine ride is a 5/5 because anything less than that hurts the driver. In reality I just need a binary "Satisfactory? Y/N" and then maybe an optional extended rating for leaving feedback if anything was out of the ordinary, good or bad.

good points. maybe that makes all the difference between Doordash compared to Google/Yelp, where McDonalds is more along the lines of 1 or 2 out of 5.

it means Doordash ratings are useless for discovery, but if you already know the restaurant, it might have utility in terms of delivery.

There is also unevenly distributed cultural knowledge: anything less than five stars starts causing hassle for the driver. I hate this inflation but I also wanna make sure the person who gave me competent service that I would realistically mark as three stars - not bad, not great, job done exactly as specified and I'm perfectly happy with that - is going to get five.

(hassle varies: offered fewer/worse-paying jobs is a good start)

When I moved to California for a while I once made the mistake of describing my food at a restaurant with "fine, thanks" when the server asked how it was.

Apparently this translates very differently in the culture of the Bay Area where it means something like "I am very upset" whereas in Boston I think this roughly would be read as "I am satisfied, but don't really feel like chatting."

"Good" is a California "fine", and "fine" is a California "pox be upon you".

Randomly reminded me of George Carlin's bit on that.

It’s like “fine.” Another weak word.

“How are you?”

“Fine.”

Bullshit! Nobody’s fine. Hair is fine.

“How’s your hair?”

“Fine.”

That makes a lot more sense to me.

There are clever words and/or phrases which do (or did) translate quite differently in British vs. American English. Starting with "clever" and "quite".

Key generally being that on a scale of 1--10, Brits tend to hover in the 5.00000003 to 5.000000031 range. Any deviation from that is an extreme reaction.

I've given up on leaving reviews. For 3 stars to be average, you need to actually give average reviews to some things, thus I rate almost everything as 3 stars which results in the "what went wrong", "Nothing, you were average, just like everyone else". I got sick of that exchange and gave up. Doing better than average is almost impossible in my book because almost everyone is doing their best. Sometimes someone will mess up and get a 1, but not often, and not bad enough that they make enough of a difference as to raise everyone else to 5.

Also, I just don't buy enough of most things to honestly know. Is Brand X thing better than Brand Y - how should I know? I don't have the time to spend a week trying each item to give a fair review. I happened to choose Brand X, and it is fine but it might really be far worse than Brand Y if only I had tried the other in stead.

They can both be 5/5 in terms of delivering the product you expect them to deliver.

Isn't this just the logical consequence of the "less than 5 stars means you want them punished" mentality popularised by Uber?

It was popularized at least as far back as eBay (albeit as a positive/neutral/negative scale). "Seller shipped me an empty box but at least there wasn't dynamite in it. A+++++++ seller."

I do think restaurant and product ratings are a bit different. With Uber, eBay, etc., a lot of people recognize that any less than perfect rating can lead to a personal impact on someone even if a small number of people leave a negative.

That said, many will leave good marks on satisfaction surveys because a middling mark may lead to someone reaching out to find out what the problem was and that can just be more hassle than it's worth.

Arguably the punishment side existed before the web then in US tipping etiquette.

But at least the eBay 'approve' asks you whether the vendor was adequate or not, not to give them the maximum on an entirely artificial multiplier scale unless you think they should face consequences.

>Arguably the punishment side existed before the web then in US tipping etiquette.

True. In general, even if service is bad for whatever reason, most people would say you should leave a tip even if a less generous one than you normally would.

I haven't left a rating on eBay for a long time and don't use it much. But for many situations there's something to be said for thumbs up/down or an up/neutral/down rating when you're not really trying to fine tune.

And I suspect the use of grades in US Highschools where anything that isnt an A is considered a failure is an element why every ones gives a 4 or 5

> are we really to believe that McDonalds has a rating of 4.6 out of 5?

Why shouldn’t it? You know what you are getting with McDonalds. For the price point they sell at, it is pretty good food. They are consistent. And they do a really good job with food safety, probably better than a lot of more highly rated restaurants (when was the last time you heard about an outbreak of a food borne illness at McDonalds)?

I very rarely eat at McDonalds because I don't like them for the most part. But, if I did, and I was promptly served what I ordered, why would I ding them because their burger isn't up to the standards of the (costlier) burger joint down the street or even Shake Shack.

Even if you didn't feel the reviews were gamed--why would BestBuy leave bad reviews on its site--it would basically be admitting they sell bad products. Ideally maybe they are removing products that get continuous bad reviews before you see them, but generally reviews can't come from the same place you buy items and be trustworthy.

> why would BestBuy leave bad reviews on its site

Because any website that doesn't do this I immediately assume is untrustworthy and I will choose to buy nothing from them at all.

How would someone know this when visiting them for an occasional purchase? You might easily assume that the lower rated items rapidly stop being carried, that the items that are carried are curated to have a real bias towards higher ratings, or merely that the lower rated items are on a different page of results.

Which isn't to say I trust Best Buy's customer reviews, I definitely don't. I probably trust Amazon's more simply by sheer quantity combined with fakespot.com to sanity check the reviews.

But I don't think a website having weird average scores is very suspicious, especially when they have small numbers of reviews. If many items are totally unreviewed and others have only a few dozen reviews, the results are essentially noise right? That's why I would trust Best Buy reviews less.

I can understand if a company uses truly bad reviews as an indicator that they should stop carrying the product. That makes sense, and you'd have survivorship bias from that. However, the problem I see with BestBuy's approach is that even average reviews are being diluted by a deluge of 5 star reviews. Decent products should be getting 3 out of 5 stars and only the truly exceptional ones should be getting the perfect 5/5.

I sense the problem is partly due to the shortcomings of the X/5 rating system.

Lots of products have <4 stars on Best Buy's website. If you're not seeing products with < 4 stars, you're just looking at their featured items.

I also agree its hard to trust online reviews, but we don't need to throw out falsehoods about how every product is 4+ stars.

https://www.bestbuy.com/site/ge-7-8-cu-ft-10-cycle-electric-...

https://www.bestbuy.com/site/whirlpool-7-4-cu-ft-36-cycle-el...

https://www.bestbuy.com/site/combo/washer-dryer-bundles/d40c...

https://www.bestbuy.com/site/hisense-32-class-h55-series-led...

https://www.bestbuy.com/site/furrion-49-class-led-outdoor-fu...

https://www.bestbuy.com/site/jam-been-there-wireless-on-ear-...

https://www.bestbuy.com/site/tcl-mtro-series-mtro200btbl-wir...

https://www.bestbuy.com/site/sennheiser-ie-80-s-bt-wireless-...

https://www.bestbuy.com/site/supersonic-7-tablet-8gb-black/6...

https://www.bestbuy.com/site/azpen-10-touch-screen-tablet-an...

To be honest though, if you were a retailer trying to push the idea that almost any product you sell if cool tech, wouldn't you probably remove most SKUs that are genuinely 2-star products?

Reply on news.ycombinator.com