Reasonable performance expectations for free Web 2.0 services
technology June 9th. 2008, 2:30amFor the 15 or so people who aren’t paying attention to Apple’s WWDC spectacle, I thought I’d offer something different for you to read. Actually, it’s more about a common topic that the blogosphere talks about, either directly or indirectly: the high expectations that we have for free Web services.
Mark Evans brings the point to the surface in his post The Wonderful World of Web 2.0 Whining where he writes about the way that people complain about Twitter’s now famous up-time and stability problems (which, it should be noted, appear to be improving) - remember, of course, that no one actually pays to use Twitter.
People, we’re talking about a free service that, for the most part, is all about entertainment, vanity and killing time. As much as I enjoy using Twitter, it’s far from being a mission critical or important online service.
It’s hard to rationally argue this point. Twitter isn’t in the same class as 911 (or 999) emergency services or public television (which, admittedly, is far more important in Canada and the UK than the US).
Mark cuts to the heart of user expectations in the following section:
The hand-wringing over Twitter’s technical troubles are really part and parcel of a more widespread phenomena: the enthusiastic groaning and griping that happens when a free online service fails to meet expectations - aka the Wonderful World of Web 2.0 Whining.
It’s bad enough no one wants to pay for anything, but the expectations placed on free services to deliver 99.99999% reliability are astounding.
If the expectations are really that high, then I continue to agree with Mark’s point. At my company, we have an internal standard of 98.0% system availability - which is 24/7/365 except for regularly scheduled outages for maintenance and such. So 100% uptime is a myth but less than 100% is acceptable when people have a reasonable expectation for when it won’t be available.
However, there’s an interesting point here to consider: many of us use free online services frequently these days, including:
- the vast majority of Google’s offerings, including G-Mail, Google Reader, Documents, etc.
- Yahoo and Microsoft’s websites, including search engines
- the Weather Network or its equivalents
- virtually every mass media website, including the broadcast and major cable networks, newspapers, magazines, etc.
- there’s still Facebook, MySpace, etc. to consider as well
Are any of these significant?
It really depends on your point of view, doesn’t it?
Would you say that search engines aren’t significant?
Consider this: do auto workers, nurses, teachers, construction workers, or the clergy depend on search engines for their livelihood? Possibly, but it’s likely that most don’t.
On the other hand, there are probably a lot of Internet marketers, probloggers, and small businesses that are extremely dependent on search engines to drive traffic and sales.
Which group is larger? Hard to say for certain, but I think there are more people out there who aren’t actually dependent on search engines than those who are. And, as far as free services go, it’s hard to find something more significant that search engines.
On the other hand, many of us are highly accustomed to using free services for communication and entertainment.
So here’s a few questions to consider:
- What’s a significant service that doesn’t cost the consumer anything in direct fees?
- What’s a reasonable expectation for uptime and performance?
- Should services like FriendFeed, Twitter, Plurk, Pownce, Facebook, or MySpace - presumably not mission-critical services - give any kind of performance or availability guarantee?
I would argue that Twitter users should be able to expect a high level of availability, considering that it’s not in beta (at least not officially). I can’t actually find any guarantee of system availability on their website, but IMO there’s an implicit assumption of high availability. However, since few systems ever truly hit availability greater than 98%, it should be reasonable to assume that systems may be down a few hours each week.
So, by the same token, wouldn’t it be reasonable to assume that Google’s Search Engine, GMail, Analytics, etc. wouldn’t be available 100% of the time? Or would it?
Is the real issue the unplanned, unexpected outage?
What do you think?
If you're new here, welcome! Please consider subscribing to my RSS feed to stay up to date with my latest posts and articles. Thanks for visiting!
Other posts that you might enjoy reading:
On FriendFeed, this post was liked by 6 people and commented on 0 times hide
June 9th, 2008 at 7:35 am
It’s true that Google is really never down. Many large businesses depend on it, too. And while the search and many other tools are free, Google never fails to make a profit and keep the shareholders happy.
When expectations are never addressed, you’re just asking for trouble. Twitter could’ve done something about this, but they didn’t. And if it would’ve seemed lame to suggest that we all expect it to have so much downtime, then they should’ve taken the real lesson in that and made it work right in the first place.
If something costs money, and it doesn’t work, I may or may not get my money back, but I certainly will leave. If something is free and it doesn’t work, my only choice is to leave. It’s not about entitlement or whining. What does Twitter want? Do they want to be popular? to be used by everyone? Well, then it had damn well better work! Otherwise they’re hurting themselves, and they’ll be over before they could ever really begin.
Now, if you’ll excuse me, I have to go plurk.
June 9th, 2008 at 8:43 am
@Michael - to Twitter’s defense, they don’t have the ongoing revenue streams that Google has in order to fund world-class development and support. So, it’s a bit of a catch-22 with getting funding vs. needing funding when all they seem to be able to do is to get the occasional dose of venture cap to pay for their cash burning. Eventually, though, the service will have to bring in some regular revenue of some kind.
My thought on Twitter was that those guys built a neat little application for their own purposes, put it out on the Web on a whim, and got caught when people adopted the service. Not the best position to be in, but not the worst either.
Don’t Plurk too hard, you might break it…
June 9th, 2008 at 9:04 am
My experience with college-aged and younger people is those who are looking for information go to Google by default. The same people have Facebook accounts but not Twitter or FriendFeed, nor do I see them needing either service. Especially FriendFeed - because they don’t have multiple online accounts to aggregate - they are highly focused on their already established friendships and like to meet people in person first….that’s my experience and obviously quite a small subsample from the reality of true user behavior
Check out Google, We Need Your Help To Lower Gas Prices from Julian Baldwin
June 9th, 2008 at 9:39 am
@Julian - nonetheless, would you say that we should be able to expect high availability from these “free” services?
June 9th, 2008 at 9:59 am
I think part of the problem is that we’re in that kind of boom-mentality. Many of these free services are talking (or being talked about) in terms of their prospects for monetisation and overall value. Therefore while the service is free people are starting to think about these groups as economically benefitting from providing the service.
The fact that the relationship isn’t as direct as “I pay you, you give me a service” doesn’t seem to matter. I think the idea is that “you’re getting paid…from somewhere…so I expect things to work”.
Check out Be Realistic, Be Like Internet Explorer from Robin Cannon
June 9th, 2008 at 10:41 am
@Robin - Ev Williams (one of the founders of Twitter) has said they have plenty of money operate. It’s interesting that you see a boom mentality as part of the problem. I find that it makes more things possible because people are more willing to take risks.
They are indeed getting paid from somewhere, and they didn’t build it to break, so I feel that it’s reasonable that we expect it to work.
Check out Should You Squidoo? from Michael Martine | Remarkablogger
June 9th, 2008 at 11:22 am
@Michael - I think the boom mentality is a wider issue for online business as a whole. It seems to me to be a mirror of the original dot com boom, lots of excitement and lots of big numbers, and not enough focus on monetisation and sustainability. The end result will also be the same, a bust and a consolidation.
In terms of performance expectations, you kind of prove the point. Regardless of whether you specifically are paying, the fact that the providers of the service have said they are doing fine financially has created an expectation of performance. That’s a legitimate expectation given the PR of the free services and their monetisation.
Check out Be Realistic, Be Like Internet Explorer from Robin Cannon
June 9th, 2008 at 11:58 am
@Robin - I see much more caution this time around. For many of the new breed of services we’re now used to, their business model was in place from the beginning and is solid (Twitter knows more than they’re telling). I guess I just don’t see it as a boom, but as normal tech growth, which should be expected to grow faster than other economic sectors. My bet is that there is no bubble, and that there will be no pop. But time will tell.
Check out Should You Squidoo? from Michael Martine | Remarkablogger
June 9th, 2008 at 12:09 pm
@Michael, @Robin - this is so cool, you guys are actually debating here! I almost didn’t leave this comment for fear of breaking the train of conversation. Woot!
June 9th, 2008 at 12:32 pm
@Mark - more like discussing, really. Debate sounds too much to me like argument.
Check out Should You Squidoo? from Michael Martine | Remarkablogger
June 9th, 2008 at 4:53 pm
@Mark
I think we should expect what the service claims they can do for us. If they guarantee 97% uptime, because it’s free we accept it. If they undershoot their estimate by a significant % then they need to readily admit what their new guarantee is and why the last one fell through. Better communication and predicted availability would help, users assume 100%.
Check out Google, We Need Your Help To Lower Gas Prices from Julian Baldwin
June 10th, 2008 at 6:54 am
I have to just stand up and applaud here. I went to read Mark Evan’s post as well.(Excellent!) I am so completely sick of the web 2.0 whining. It’s free, for Pete’s sake, they never promised a rose garden or anything. If I am being quite honest, the sarky tech snobbery and whining are two of the major things that have kinda put me off Twitter. I still post there but, the sark factor gets monotonous after awhile.
June 12th, 2008 at 7:31 am
[...] no practical way for most of these sites and services to charge. They’ve created an expectation that they will be provided for free. The quickest way to kill their user base will be to charge, with the exception of some niche sites [...]