7 ms·
My perspective here is in no way Apple-specific, but I was reminded of something quite interesting in this story: if you pass a test and you still have problems
by kevinalexbrown 14y ago
My perspective here is in no way Apple-specific, but I was reminded of something quite interesting in this story: if you pass a test and you still have problems, it probably means the test isn't good enough.
Worse, passing a test doesn't magically change the state of the world and make these problems disappear. Turning to the customer and explaining, "there is no problem because you passed the test" does not alter the customer's experience at all: the ghosting still exists.
This has nothing to do with whether or not Apple was smart to behave this way, or what the tolerances should be - I don't have a dog in this fight. Instead, I just wanted to highlight something I have to be reminded of from time to time, in different contexts: no one cares whether you think their problem is real, they care whether the problem still exists for them.
Obvious point: This goes for more than just computer products.
- j-b 14y agoDon't forget the old adage that reality is 90% perception. If your customers "think" they have a problem, that in itself is an issue that needs to be dealt with. My customers don't care if their system issues are actually a problem in my software or not. "Not a problem" doesn't work if you want to keep paying customers around.
- cloverich 14y agoIn these types of situations, the cliche "The Customer is always right" is apt.
- personlurking 14y agoThis also works in personal relationships. If your SO feels hurt by something and you say "oh, grow up!", that doesn't help, even if you think some 'tough love' was needed/adequate. You have to deal with her as someone who is feeling hurt and until she perceives that you understand her, you'll just be sweeping the issue under the rug. I suppose this can be classified as listening but not really hearing.
- koide 14y agoOn the other hand, sometimes tough love is indeed the adequate response and makes the SO to notice (s)he's being childish. Especially if said SO is generally mature enough to notice.
- ak39 14y agoI disagree. The reason why we are often too eager to dismiss is because kindness takes effort. Empathy, and certainly sympathy, is hard work. We get lazy, we become exhausted and our reactions become snappy. Almost always the "dry your eyes" advice is inappropriate and insensitive. And it never really achieves its intended objective. I'm not saying we should all participate in crying orgies whenever someone shows signs of weakness (though, as I've aged, I think that is better than showing a cold shoulder) or that we readily invest in "idiot compassion". What I'm saying is: we are all trapped in this existential reality together and we all need to show kindness with all forms of suffering. Being cruel to be kind as a philanthropic philosophy has never worked (for me). Watch this: http://www.youtube.com/watch?v=u5um8QWWRvo http://www.youtube.com/watch?v=u5um8QWWRvo (RSA Animate on "Smile or Die" a talk by Barbara Ehrenreich)
- koide 14y agoNote that I didn't say you should always be tough. Of course that being kind and compassionate is the right answer most of the time. It's just that sometimes a figurative slap is the right thing the sufferer needs to wake up from the suffering and realize there's no actual need to suffer like that. It's all very personal and context dependent and, most importantly, it should never be done in anger.
- ripperdoc 14y agoThat's true about "customerS" but not for "a customer". There will always be a percentage of customers that expect more than the norm (for good or not so good reasons). The company cannot treat everyone's opinion the same, as they'd then need to design a product and customer support that makes the most picky person happy - which would be nice for all customers but very costly. In the end, the company has to decide where they put the bar for "good customer support", from "making at least 50% happy" up to "making 95% happy" but never "making 100% happy". And of course, the price paid should be in relation to this, which in Apple's case, it may not be (but to be fair - I hear a lot of good stories about Apple support as well, so this one event cannot determine whether Apple does support well).
- antoinevg 14y agoAll it takes to make 100% of the customers happy at low cost is to give the other 5% special consideration when they aren't happy with the quality control that satisfies the other 95%. Or to put it another way: The cost to Apple's reputation of an angry blog post about ghosting problems on their flagship notebook is considerable. On the other hand, the cost of simply swapping this fellows machine out would have been negligible.
- baddox 14y agoI don't think anyone at Apple would claim that the test magically changes the state of the world. They probably wouldn't even deny that there is a ghosting problem on your machine. It's clear that the test is meant to determine whether the particular machine's ghosting problem is within Apple's acceptable limits. It's just like the dead pixel guarantees you sometimes see on monitors, whereby you get a replacement if there are >4 dead pixels anywhere on the screen or >1 in the center of the screen. No one's denying that you have exactly 4 dead pixels, they're just not going to give you a replacement unless you have 5. Now, obviously it's fair to argue that Apple's test is too strict, and that OP does in fact deserve a replacement. The dead pixel guarantee is disclosed to potential buyers, and likewise this ghosting test probably should be as well (and I'm assuming it's not). But your implication that Apple is making a bizarre metaphysical claim about their tests is ridiculous.
- jrochkind1 14y agoBut your specs for what you repair have to be reasonable if you want to have customers think you deliver a high quality product (and if you want to actually do so). Two dead pixels near the edge of the screen, okay, most of us think, that's still usable, I'll soon stop noticing. (How many is 'most of us'? Enough that a monitor company can do that, and still be considered to deliver a high quality product). The ghosting OP describes and takes pictures of? I think would probably interfere with the use of most reasonable people. If those are Apple's specs, then Apple is willing to deliver a crappy product; either Apple made a mistake, their test isn't catching what they meant to catch, or Apple thinks it's going to have enough customers that keep buying Apple anyway even after getting a crappy product, I guess.
- baddox 14y agoI think I agree, but there's still a tone in your and OP's words that makes it sound like you're appealing to some objective measure of a "high quality product." Obviously, there is no such thing, and different problems will frustrate different people. It's quite possible that 99% of MBP buyers would never even notice that ghosting problem. Does that still make the MBP a "low quality product," and are Apple's policies still unreasonable, or does it just mean that the product simply isn't compatible with OP's needs? The same thing applies to my dead pixel analogy. Personally, I am extremely distracted by dead pixels (even a single one in a non-retina display drives me crazy). I have always purchased displays with very strict dead pixel tolerances. But it would be ridiculous for me to buy a low-end display then complain that the manufacturer makes crappy products and doesn't care about its customers because they refuse to replace my display that only has a single dead pixel.
- joe_the_user 14y agoI strongly suspect that the testing mania you get on all the technical support lines comes from a mania by companies to control/pressure their customer support people and those people more or less passing along this pressure. "You have to prove you have a problem here because I've got a lot of people I have to help". The last technical support call I had for a Windows network problem (which I had isolated to the OS before I called), I could strongly sense the guy was inching to cut me off with either "not our problem", "not a problem" or "I can't help you 'cause won't listen to me". The implicit threats were strong enough that the table turn from "I'm here to help you" to "you're here to prove there's a problem...". It seems like it's a symptom of making technical support a large, low-wage, heavy control work environment. Edit: It doesn't help that with large systems like Windows 7 (or modern OSX), there's no "here's the explanation of the problem", it is only "try this generic solution recipe, now try this other one..."
- chris_wot 14y agoIn Australia, the test wold be irrelevant. I'd just take them to Fair Trading, who would take one look at the test and dismiss it. Problem solved.
- jmcdonald-ut 14y agoI used to work phone support for a large auction site (your first thought is probably correct). I can confirm this notion or idea. We were given generic things that the customer could try to fix the problem. You know "Clear cache and cookies", etc. If the issue persisted and we could not reproduce the problems on our own machine we weren't allowed to push trouble tickets to our QA team. For a lot of the lower tech users, I suspect issues were on their end. However I often got users who obviously knew what they were doing, and it was frustrating that there wasn't more we could do.
- illuminate 14y ago"If the issue persisted and we could not reproduce the problems on our own machine we weren't allowed to push trouble tickets to our QA team." Yup. There's often an air gap between the grunts and the QA people, who mostly pore over trouble tickets well after the fact.
- joonix 14y agoMy advice is to use what the government has set out for people to use: the courts. Sue them in small claims court. Apple has sold something that is not suitable for its intended purpose, and has breached warranties required by law such as warranty of merchantability. Even a place like Texas has the Deceptive Trade Practices Act. Under this act a consumer can sue for breach of these warranties and receive damages, AND get court costs and attorneys fees! Take them to small claims, you will probably win, if they show up at all.
- jtheory 14y agoI can't imagine it's that clearcut. What's the "intended purpose" for a general-purpose laptop? Editing text, watching video, playing music, etc.., plus the long tail of more unusual things that may be possible depending on what software you install. Are there any relatively common tasks that will actually be prevented by this level of ghosting? I don't mean "it's a bit annoying", I mean "this task is no longer feasible". I've never used a computer with any ghosting that bothered me personally, so I may be missing something. I'm not a graphic designer though; but neither is the author of the post AFAICT (PhD business student). So my feel is that this is more in the category of a scratch on the bottom of a new laptop (doesn't actually affect the use, but annoys you each time you notice it and remember you paid full price) than, say, a space bar that only works 80% of the time (which would actively mar your ability to use the device). So sure, Apple may not be playing this right to give their customers the appropriate warm fuzzy feelings, but I highly doubt they're doing anything illegal.
- moostapha 14y ago>Are there any relatively common tasks that will actually be prevented by this level of ghosting? I don't mean "it's a bit annoying", I mean "this task is no longer feasible". Apple bills their computers as "the computers you do stuff with". They have a long history in the professional media/content creation business, and their marketing campaigns and advertizements for the last several years were based on bringing those capabilities to the mainstream. If nothing else, iPhoto, iMovie, and GarageBand come with every Mac. With those issues, two of the potential uses for the computer–which comes bundled with Apple software to do them–become infeasible for anyone who cares if they exist at all.
- andrewfelix 14y agoWell said. If the customer is unhappy with the product, there is a problem.
- kang 14y agoHaving worked in customer service I can tell you that sometimes you have to draw a line where boundaries are not very discrete. Customers can tell me that "their connection is too slow". The only proof I can give them is "it passes the speed test". This usually happens when the money paid for the service exceeds the service provided. Although, what I have said is a general reply to general comment above, but i think this does not apply to apple here because they certainly charge enough.
- lsc 14y ago>Turning to the customer and explaining, "there is no problem because you passed the test" does not alter the customer's experience at all: the ghosting still exists Yes, but you don't always want to do everything the customer wants. I mean, maybe this is a defect in the screen, in which case, you are right that apple should probably fix the test. but, maybe this is just how those displays work. Maybe this level of ghosting is simply the best you can do with the technology Apple used to have the screen made? In that case, it's really a problem of setting expectations. The problem is that even if you provide the exact same product and support to 100 customers, you will have 1 who will demand more support time than the rest. At some point, you need to decide how much effort you want to put in to that customer. I mean, the problem with those 1 out of 100 customers? they do complain to their friends, like this guy did. So yeah, you don't want those customers? but you also need to fire them in a reasonable way. "I'll give you your money back if you go away" sort of thing. If you are a company that really values your customer service reputation? maybe you don't want to fire the 1 out of 100 customer. Maybe you want to eat that cost. but you do want to fire the 1 in 1000 customer. There is a point at which you need to say 'no, enough' just because of how much money one user can cost you. (Again, I think the best way to handle it is to apologize sincerely (because, after all, it could very well have been an undetected defect in your product) give them a full refund, and point them at your competition. I think Apple may have a hard time doing that, though.) > Worse, passing a test doesn't magically change the state of the world and make these problems disappear. Turning to the customer and explaining, "there is no problem because you passed the test" does not alter the customer's experience at all: the ghosting still exists. The point of the test is to determine if your problem is with the customer, or with the product. Maybe there is a certain amount of ghosting on all of these screens, and most people just don't care? in that case, the test is a way of saying "this guy is so sensitive that we don't want him as a customer." What I want to know is what the hell OP is doing replacing the screen? I mean, clearly, this is within apple's tolerances, so why does he think a new screen would be better?