June 10, 2008
Amazon’s Technical Difficulties: Part 2
When Amazon.com’s website went down on Friday, there were reports about the potential millions the online retailer lost as a result. While not happy over being closed for business, Amazon’s response to the Friday outage – that “on rare occasions” even it “may experience problems” – was intended to assure consumers that they could expect a quick return to the dependable performance they’ve come to expect from the site.
But what are consumers to think now that Amazon experienced outages on its site once again? Keynote Systems, a company that monitors websites, reported that Amazon began experiencing trouble about one o’clock eastern time yesterday and that consumers were unable to get on the site for an hour or more. Keynote’s probes found similar problems with Amazon’s U.K. site but others around the globe appeared unaffected.
Amazon had no comment on its latest troubles at the time of this report. The company had more than 58 million visitors to its U.S. website in April, according to comScore.
Discussion Questions: Do you think Amazon lost a large number of purchases with its recent outages or have consumers by-and-large waited for it to come back online to buy whatever they were looking for? Now that it has experience outages for a second time, do you think consumers will begin to question Amazon’s reliability?
BrainTrust
Recent Discussions
Discussion Questions
Poll
The biggest issue here is with Amazon’s response. Amazon is saying that it knows the cause of the crash and won’t release it.
Now that is going to make some people nervous. If it was isolated–or had been fully fixed on Friday–that would be a different situation. But with another crash on Monday, this is not advisable.
Yes, that “we know and won’t say” coupled with continuing incidents may indeed start to undermine confidence.
On the other hand, the site was still available much of the day. If this doesn’t repeat today (Tuesday), there may be no long-lasting damage. If it does repeat, then I’d argue that it’s critical for Amazon to announce what it knows to comfort nervous customers and suppliers.
Keynote has been on top of this and they’ve been pushing an interesting theory that the crashes are because Amazon’s legacy E-Commerce sophistication may have caught up with it, creating a brilliant system that now can collapse with the smallest glitch.
Now that’s an interesting theory.
I am sure the attention span for shopping online is much shorter than brick and mortar since there are so many choices but we have all experienced server issues, power failures and the computer freeze. While it is unfortunate for Amazon and must have caused lost sales, my impression they will still be first choice for their core customer and this will be forgotten quickly
As technology has become more and more a part of our daily lives, I believe consumers have become much more understanding and tolerant of outages. They have experienced problems with their own computers, cell phones, or automobiles that have been attributed to “glitches.” This technical term seems to apply to many of our failures to meet our own or others expectations.
To me, the classic demonstration of this is the growing number of instances where companies are willing to release “Beta” or even “Alpha” versions of new applications. Supposedly, a Beta version of a product has been tested by the QA department of the company releasing it. New bugs should be limited to “stress related issues” caused by higher volumes that the company was not able to emulate. An Alpha version has been tested by the development team, but not an independent QA department. The very fact that development companies are willing to announce “alpha versions” of a product implies that consumers are willing to complete the testing.
I don’t think Amazon will experience any quantifiable reaction from customers. The good thing is that it was not the first week of December. The more curious thing is “what were they doing?” I wonder what new shopping experiences Amazon has in store?
As one who generally has absolutely no patience whatsoever with malfunctioning websites, and is generally unforgiving, I am prepared to forgive Amazon, at least until it has happened a few more times or they come clean about what’s going on over there. Considering their reputation and the loyalty they have built with customers, I would be surprised if most people didn’t decide to give them room to manoeuvre for a little while longer.
Amazon’s lack of disclosure of why the site has been going down will possibly be a larger long term problem for it than the actual loss of business for a few hours. Why are they being so coy? Is this an internal problem? Has the site been successfully hacked from abroad? Has customer proprietary information and credit card data been compromised? Lots of people are looking for answers and may be reluctant to use Amazon until answers are forthcoming.
Do Amazon shoppers (who aren’t IT professionals) really care why the site goes down? And how many sites are trouble-free anyway? I find sites with broken links and other obvious defects all the time. Most fun of all: slow response time and other obvious defects on IT provider web sites.
Liatt is exactly right.
The problem is not that the site went down. The problem is that it went down again, and the company is not telling anyone why.
Evan knows I don’t buy the “web site is too complex” theory. If that’s true, you just roll back to an earlier state, and re-institute features once you figure out how it got screwed up. That’s just basic IT governance. You don’t take a chance on it happening again.
The fact that it happened again–well, it makes me think of my home burglar alarm. We’ve been having false alarms for over a year now, often in the middle of the night. Finally, my provider acknowledged it really doesn’t know WHY it’s happening. So I’ve decided to stop paying maintenance until they figure it out.
Amazon had better figure out why and let the public know soon. In a vacuum, people will think the worst. Now, there are worse times of years for this to happen, so even if it’s equipment failure, or bad programming (gasp!) I don’t see customers leaving in droves. It just needs to stop happening soon.
Online consumers will patiently wait for a few minutes to allow a web site to get back online. However so much of the equation depends on the type of item being purchased. If it’s a gift, specialty item, or an expensive item where loyalty to the web site induces free shipping, or discounts, etc, the consumer then will wait a day or two to make the purchase. However, if shipping charges are involved either way, and if the item is a commodity, or widely available elsewhere, then the consumer is likely to go elsewhere.
Amazon may have missed well over a million hits (based on the April numbers) during the hour or so when it was offline. Hard to know how many of these visits were browsers vs. buyers, but it doubtless cost some sales when it happened. Most regular shoppers at Amazon are loyal to the site based on its breadth of selection, ease of use and reliable customer service…so the long-term damage probably isn’t extreme. However, “ease of use” means (among other things) site reliability, so Amazon needs to push hard in order to ensure that this doesn’t happen again.
More Discussions
Has Same-Day Delivery Become an Expectation Across Retail?
Are expectations around retail delivery speed in flux?OpenAI Names the Infrastructure Standard for Agentic Commerce
Will OpenAI or Google (or another entity) set the gold standard for agentic commerce?Is Five Below Poised For Even Greater Growth?
Five Below is hitting serious growth numbers. Can this trend continue?
The biggest issue here is with Amazon’s response. Amazon is saying that it knows the cause of the crash and won’t release it.
Now that is going to make some people nervous. If it was isolated–or had been fully fixed on Friday–that would be a different situation. But with another crash on Monday, this is not advisable.
Yes, that “we know and won’t say” coupled with continuing incidents may indeed start to undermine confidence.
On the other hand, the site was still available much of the day. If this doesn’t repeat today (Tuesday), there may be no long-lasting damage. If it does repeat, then I’d argue that it’s critical for Amazon to announce what it knows to comfort nervous customers and suppliers.
Keynote has been on top of this and they’ve been pushing an interesting theory that the crashes are because Amazon’s legacy E-Commerce sophistication may have caught up with it, creating a brilliant system that now can collapse with the smallest glitch.
Now that’s an interesting theory.
I am sure the attention span for shopping online is much shorter than brick and mortar since there are so many choices but we have all experienced server issues, power failures and the computer freeze. While it is unfortunate for Amazon and must have caused lost sales, my impression they will still be first choice for their core customer and this will be forgotten quickly
As technology has become more and more a part of our daily lives, I believe consumers have become much more understanding and tolerant of outages. They have experienced problems with their own computers, cell phones, or automobiles that have been attributed to “glitches.” This technical term seems to apply to many of our failures to meet our own or others expectations.
To me, the classic demonstration of this is the growing number of instances where companies are willing to release “Beta” or even “Alpha” versions of new applications. Supposedly, a Beta version of a product has been tested by the QA department of the company releasing it. New bugs should be limited to “stress related issues” caused by higher volumes that the company was not able to emulate. An Alpha version has been tested by the development team, but not an independent QA department. The very fact that development companies are willing to announce “alpha versions” of a product implies that consumers are willing to complete the testing.
I don’t think Amazon will experience any quantifiable reaction from customers. The good thing is that it was not the first week of December. The more curious thing is “what were they doing?” I wonder what new shopping experiences Amazon has in store?
As one who generally has absolutely no patience whatsoever with malfunctioning websites, and is generally unforgiving, I am prepared to forgive Amazon, at least until it has happened a few more times or they come clean about what’s going on over there. Considering their reputation and the loyalty they have built with customers, I would be surprised if most people didn’t decide to give them room to manoeuvre for a little while longer.
Amazon’s lack of disclosure of why the site has been going down will possibly be a larger long term problem for it than the actual loss of business for a few hours. Why are they being so coy? Is this an internal problem? Has the site been successfully hacked from abroad? Has customer proprietary information and credit card data been compromised? Lots of people are looking for answers and may be reluctant to use Amazon until answers are forthcoming.
Do Amazon shoppers (who aren’t IT professionals) really care why the site goes down? And how many sites are trouble-free anyway? I find sites with broken links and other obvious defects all the time. Most fun of all: slow response time and other obvious defects on IT provider web sites.
Liatt is exactly right.
The problem is not that the site went down. The problem is that it went down again, and the company is not telling anyone why.
Evan knows I don’t buy the “web site is too complex” theory. If that’s true, you just roll back to an earlier state, and re-institute features once you figure out how it got screwed up. That’s just basic IT governance. You don’t take a chance on it happening again.
The fact that it happened again–well, it makes me think of my home burglar alarm. We’ve been having false alarms for over a year now, often in the middle of the night. Finally, my provider acknowledged it really doesn’t know WHY it’s happening. So I’ve decided to stop paying maintenance until they figure it out.
Amazon had better figure out why and let the public know soon. In a vacuum, people will think the worst. Now, there are worse times of years for this to happen, so even if it’s equipment failure, or bad programming (gasp!) I don’t see customers leaving in droves. It just needs to stop happening soon.
Online consumers will patiently wait for a few minutes to allow a web site to get back online. However so much of the equation depends on the type of item being purchased. If it’s a gift, specialty item, or an expensive item where loyalty to the web site induces free shipping, or discounts, etc, the consumer then will wait a day or two to make the purchase. However, if shipping charges are involved either way, and if the item is a commodity, or widely available elsewhere, then the consumer is likely to go elsewhere.
Amazon may have missed well over a million hits (based on the April numbers) during the hour or so when it was offline. Hard to know how many of these visits were browsers vs. buyers, but it doubtless cost some sales when it happened. Most regular shoppers at Amazon are loyal to the site based on its breadth of selection, ease of use and reliable customer service…so the long-term damage probably isn’t extreme. However, “ease of use” means (among other things) site reliability, so Amazon needs to push hard in order to ensure that this doesn’t happen again.