Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Apple Removes 'Rate on Delete' for Apps in iPhone 4
- Thread starter MacRumors
- Start date
- Sort by reaction score
Exactly, often deleted apps, should be outed for what they are. There's a reason why they are deleted, and it should be reflected as it was by easy ratings upon deletion.
Plus as I pointed out, the system is not statistically accurate now that this option has been removed, and it wasn't statistically accurate when it was in place. To say that this is a statistically accurate survey now that the negative ratings are gone, is -no a offense everyone- a load of garbage. Ratings are an inaccurate indicator, always have, always will be.
What's more I would say it was more statistically accurate with the ratings on delete system in place as I am explaining in an above post.
so if i think the app is bad i cant easily rate down? eh whatever. ill just reenable it via jailbreak.
These are very good points, but I think you got carried away with your logic. Let me use your example:
One app, has a 1000 users, that is 99% satisfaction rate and 1% unsatisfied customers that give it 10 one star ratings.
Comparatively an app with 50% satisfaction rate would have say 500 one star ratings, hence the 50/1 ratio of bad ratings DOES reflect the actual difference between the apps.
Ratings are lowered overall, but comparatively between apps it's absolutely fair. Which is what the rating system is about right? It's not about absolute ratings, but comparative ratings that enable buying decision, it's not about keeping all the apps (as will happen now) to a 4+ rating and then not knowing which one to choose from. Even if a lower score is a by product, still the differences will be more prominent between the good and bad apps with the rate upon delete.
Because the alternative scenario, is what? The 50% and 99% satisfaction rate apps no longer have a 50/1 negative review ratio, but a 1/1 negative review ratio, and rely solely on the skewed vocal minority who actually bother to go to the app store and vote. And let me tell, the way it will work now after this decision (to appease developers) by apple, is that bad apps won't be penalized enough, and they will approach much better apps.
This is the real sampling bias, not the other way around, I and I am sure many, many other posters here, and what's more users who don't like to be vocal on such issues, have never bothered to submit a rating in the app store, ever, and I don't even know a person from my close friends and acquaintances who ever has to tell you the truth. BUT I do no know people, including myself, who managed to rate down a really bad, poorly updated or developed app upon deletion, with this simple, easy prompt, that wouldn't have done so otherwise, because let's face it, we don't have enough time to do so, and we are not some kind of 13 year old (no offence kids, just making a point) who has ample time to spent rating and commenting on the app store.
Now instead of offering the average user that easy option to rate down the bad, they are offering up the ratings system on a platter to every time waster that makes it a point to express his opinions on apps in the app store. Like I said, THIS is the real sampling bias, not the other way around. The negative rating enables EVERYONE to vote down bad apps, without it, only the vocal group, the group with a lot of time or (vested) interest will actually go to the app store and rate.
"Often" is a relative term in the arena of statistics. You are making a leap in logic in your usage here. You are taking a smaller group (those that rate the apps) and saying that they're really a larger group (those that download the apps).
The star rating system simply tells you how many were satisfied AMONG THOSE THAT OFFERED A RATING. Not among those that used the app. So you could potentially have 1000 customers who for the most part are happy with the app. But only a handful go into to rate the app (10 for example).
Its a basic fact that most who are deleting an app, are doing so because something went wrong. Could be a bug, could be user error, could just be it wasnt what they were looking for. But the bottom line is, the majority of those deleting the app are going to give a bad rating when asked at the time of deletion.
The customer who is deleting the app has more oportunity to rate an app, than the customer who is keeping the app. As a result, more ratings are coming from unsatisfied customers. So what the potential customer sees on the AppStore in the rating system is going to be 10 ratings (out of 1000 actual users). Most of those ratings are from customers deleting the app. Which means the rating is mostly going to be negative. Which means the rating system is skewed negatively away from the actual satisfaction rate of the 1000 actual users.
There are only two possible solutions to this issue:
- Eliminate the prompt on delete
- Add a prompt on X uses of the app
Certainly this will reduce the statistical bias that most reviews come from people who have just, for whatever reason, decided they not longer want your application. However, by having fewer total reviews this will amplify other unrelated faults in the review system.
In my experience the worst problem with the ratings system is not real user reviews, but false reviews by other application developers who are directly competing with you. The worst case of this (for me) was that I was recently hit by a deluge of (20 or so) false reviews and ratings from shill accounts of a competing developer. This problem will get worse when there are fewer real users submitting reviews for the application. It will also make it easier for developers to boost their own ratings using shill accounts.
So I'm not really sure if this is a good or bad thing.
In my experience the worst problem with the ratings system is not real user reviews, but false reviews by other application developers who are directly competing with you. The worst case of this (for me) was that I was recently hit by a deluge of (20 or so) false reviews and ratings from shill accounts of a competing developer. This problem will get worse when there are fewer real users submitting reviews for the application. It will also make it easier for developers to boost their own ratings using shill accounts.
So I'm not really sure if this is a good or bad thing.
alternatively...
When you delete an app, report back to the store how many times you started it over what period.
All opt-in, and anonymous.
So you could filter out in the app store all apps that people ran maybe once or twice before deleting.
No need to ask for their rating.
When you delete an app, report back to the store how many times you started it over what period.
All opt-in, and anonymous.
So you could filter out in the app store all apps that people ran maybe once or twice before deleting.
No need to ask for their rating.
I definitely see your point here. I've been a developer on other mobile platforms where I sold my apps through online systems similar to the AppStore. The big problem that I ran into was a sort of 'restaraunt service satisfaction' effect. Only those that had a bad experience left a review or rating.
What I found to be the best answer was a try before you buy system. This allowed a potential customer to try out my app without having to risk any money. If they didn't like it (for whatever reason) they were less aggrevated than if they had actually put money on the table. And they were as a result less likely to leave a bad review. The other side of the coin is they were as likely to leave a review as a satisfied customer.
The result, the success of my company is based on sales and not PR on those platforms. I've tried to reproduce that on iPhone with InApp purchase.
Another problem could possibly be some of these organizations that will post positive reviews for money from developers.
At the same time, by removing this potential bias there may be less demand for such services. But that will take time since Pandora's box has already been opened.
@cteel2004
You might not have read my post above when you wrote this as we might have been replying concurrently. Have a look at it. But let me also just reiterate that despite the former system lowering overall scores it did reflect the relative merits of each app when compared to another app more accurately than the current proposed one.
Because a real sampling bias is introduced here, which is that ratings reflect the most vocal groups of people, the ones with more time, the younger ones who care more, and are involved more and will bother to go to the app store and vote. Whilst the former system enabled everyone who deleted, and almost everyone has deleted and will delete an app, to vote down a bad app, with an easy, simple, hassle free prompt. It makes perfect sense for a frequently deleted app to get a bad rating.
Apple is just appeasing developers this way, and messing up a pretty good system to separate the good with the bad, even if it affects the overall scores. But, like I said it's not about absolute values, it's not about 4.5+ reviews across the board, it's about relative values that enable an informed buying decision.
I agree that ultimately a rating after X uses system, would be the most accurate by all accounts, but as you might understand it would be very hard to implement, as the user would find it bothersome to rate some maybe 100+ apps say at random intervals during use.
So, as it was, they gave the option to EVERY user - not just the vocal minority of the app store, for those apps that are bad enough to be deleted, to have a simple prompt for a rating.
You might not have read my post above when you wrote this as we might have been replying concurrently. Have a look at it. But let me also just reiterate that despite the former system lowering overall scores it did reflect the relative merits of each app when compared to another app more accurately than the current proposed one.
Because a real sampling bias is introduced here, which is that ratings reflect the most vocal groups of people, the ones with more time, the younger ones who care more, and are involved more and will bother to go to the app store and vote. Whilst the former system enabled everyone who deleted, and almost everyone has deleted and will delete an app, to vote down a bad app, with an easy, simple, hassle free prompt. It makes perfect sense for a frequently deleted app to get a bad rating.
Apple is just appeasing developers this way, and messing up a pretty good system to separate the good with the bad, even if it affects the overall scores. But, like I said it's not about absolute values, it's not about 4.5+ reviews across the board, it's about relative values that enable an informed buying decision.
I agree that ultimately a rating after X uses system, would be the most accurate by all accounts, but as you might understand it would be very hard to implement, as the user would find it bothersome to rate some maybe 100+ apps say at random intervals during use.
So, as it was, they gave the option to EVERY user - not just the vocal minority of the app store, for those apps that are bad enough to be deleted, to have a simple prompt for a rating.
For sure, very good points. Btw, I feel for you... I suspected this kind of behaviour, boy is this a cut throat environment...
Anyway, people with vested interests, like I said, will find more motives to go into the app store and use shill accounts to rate negatively. This I think is taking away the users ability for a simple, upon deletion, rating system.
Now someone can organize a nice up voting for his or her app in the app store, get only a few very motivated, very vocal negative comments by people who actually bother to rate and review, instead of just going about their business, and they will present a perfectly respectable image at the app store. Because there won't be that simple easy prompt upon deletion for the average guy/gal to vote in click and forget about it, to bring the app down to where it belongs.
A shame really, we ll be seeing better overall scores, and apple want that to reflect well on the app store, but this is a poor marketing decision and nothing with the developer or user really in mind. Just putting a nice shine on the surface of the app store with better overall scores...
That's a very wise strategy. How's the inapp purchase method working for the iphone? The only equivalent to try before you buy is the lite/ad supported version of an app in the iphone, which is not really the same, but close. Btw, give me the heads up for your apps in a pm, if you get the time.
InApp purchases can be used to 'unlock' features in an app. You can't do time based trial versions (which is what I typically have done on other platforms), but you can give the user a good preview of your app's functionality. If they like it, they do the InApp purchase, and the app then unlocks the more usable features.
Essentially you have a lite version and a full version all in one download. The full version differences get unlocked by the InApp purchase.
I wrote a routine to do that within the app. I felt bad doing it, and every so often someone will write a bad review just because of that, but it did boost the written reviews tenfold (90% positive).
I will be more likely to give away free apps with this method. Too many folks download a free app that has nothing to do with them (such as ringtones on an iPod touch) and rate it 1 star when they delete it, because it wasn't for them.
In my experience, only a small percentage of app purchasers rate the app in the first place, and only those who have strong opinions one way or the other. Doubt this will make a substantive difference.
It did make a noticable difference when the system was introduced a while ago. I'm glad it's gone now.
I agree, if I delete an app, it is because it doesn't work as advertised, or crashes. If it is a good app, and it works correctly, there should be no problem. Maybe just add a rating screen in the app so we could rate it high if we are pleased with it.
If it actually worked that way, I'd agree with you. But there's no way to see that ratio. All you know is that app A has a 1 star rating over 10 ratings, and app B has a 1 star rating over 500 ratings. You don't get to see the ratio. It could be that app A was downloaded 10 times and app B has been downloaded 200,000 times. But they both look bad because they have a 1 star rating when you browse the app store. If you try to be "intelligent" about it, you might even think B is worse because it has way more bad ratings than A does.
Does taking away the rate on delete suddenly make it a fair system? No. As it hers have said, it's just human nature that only the extremes will take the time to rate so you still wind up with a skewed rating. Does this make it less skewed? Absolutely. At least now both extremes have to go to the same amount of effort to leave a rating.
And with iPhone OS 4.0, the only reason to jailbreak is to use dodgy, often pirated software from illegitimate sources. All the other rationalizations (eg: multi-tasking) will have gone away.
For anyone who thinks they're "leet" enough to want to jailbreak their phone: fine, but be honest about why. And don't be a numb-nuts and jailbreak, like, your mother's phone for them. Someone who doesn't even know what it means, doesn't know the risks, doesn't know why you'd want to in the first place, and won't miss whatever random esoteric thing you use to justify it on your phone. Particularly since they won't know or understand that they can no longer go to Apple for help with it. (I hope you've got your headset on, cos you become Support.)
On the "Rate on Delete" thing, the other thing Apple should consider doing is trying to teach the morons who don't understand 1 star versus 5 stars, by giving a text description when they choose a star level. eg: 1 star displays "Poor" in some manner, while 5 stars displays "Excellent". This isn't solely directed at "Rate on Delete" - the App Store needs to do the same thing. I don't know how many times I've looked at the reviews in the App Store, and seen people rate an app 1 star, with a comment of something like "BEST APP EVA!!1!". I think these idiots think 1 star means it's "#1 app". A little prompting might gradually train the monkeys into responding correctly.
Actually, jailbreaking still offers some solutions to two remaining issues - crappy notifications and the inability to share files between apps.