* You have to charge from the very beginning. If you start a free service, and then try to establish a pay system afterward, your users will feel tricked and trapped and they will rebel loudly. Scribd seems to have been in a hurry to get adoption, so they made it free to host documents; as this guy said though, he much preferred hosting with Scribd over doing it manually on his university's web server. That could have been Scribd's value proposition, and a small yearly fee for that probably would have worked OK.
* SaaS could get itself in trouble if there are too many incidents like this. I already hear from clients that are concerned about using online services; the most common questions are, "What if they change their terms?", "What if they go away?", and those are legitimate concerns. Many of my clients aren't the most computer-interested people, so if they have concerns like that, then that means that stories like this have penetrated very deep into the consumer market.
> If you start a free service, and then try to establish a pay system afterward, your users will feel tricked and trapped and they will rebel loudly.
If you lock behind a paywall things previously available for free, sure. And they'll be right too.
An other option is to add features which are only behind a new paywall. Issue then is providing additional services of value and a way for users to discover them.
Just as a case study, Reddit did exactly that, and their users flipped out. Reddit ended up doing OK with it -- they've made at least enough from it to be able to afford some infrastructure upgrades and at least one new hire -- but there was a pretty vocal not-small group of users that were unhappy about it.
The initial response was to flip out, but once people realized reddit wasn't going to switch to a pay-only site the outrage died down. It might have only worked because reddit users/admins try so hard to foster a sense of community, but it seemed to me like reddit showed how to set up a pay service correctly, and people were just too jaded because of sites like Scribd to realize it.
> ...once people realized reddit wasn't going to switch to a pay-only site the outrage died down.
I'm not very active in Reddit anymore -- HN is my last distraction now -- but I did watch that situation carefully because it was interesting. I got the sense that the outrage died down only because Reddit promised to migrate its paid features into the "free" arena. Until then, a lot of people were angry that there was going to be some kind of "elite" class on the site.
Practically speaking, even if Reddit had managed to permanently piss off this chunk of its userbase, I doubt they would have left altogether, and even if they did, there's no way to gauge what fraction of Reddit's traffic was really represented by these guys. It could've been (and probably was) just a tempest in a teapot.
But either way, in that specific case, there would have been some vocal rebellion if the paid-for features stayed locked behind a paywall.
> I got the sense that the outrage died down only because Reddit promised to migrate its paid features into the "free" arena. Until then, a lot of people were angry that there was going to be some kind of "elite" class on the site.
As a Charter and gold reddit member... not really. It was well understood early on that most of the "gold" features would be stuff that was too expensive (computationally) to give to everybody (especially before the server upgrade), at least from the start.
The great fear gripping everybody was truly that Reddit would somehow become "for-pay", and that content previously free would now become non-free (a fear which didn't make much sense as the community is the one providing the content, and the admins might not be too media and ad-savvy, but they're not stupid)
> But either way, in that specific case, there would have been some vocal rebellion if the paid-for features stayed locked behind a paywall.
Doubtful. 1000 comments/thread is a nice feature for instance, but it didn't exist (at all) before gold and it's not exactly a deal-breaker.
Ahh, true. I read it as a reply to the entire comment, not a specific part (mostly because that's the way it was formatted), but you are correct. I'll edit my original comment. Thanks. =)
It seems like Scribd would solve more problems if they let uploaders pay to give their readers free access. I'm not the Scribd user; the guy publishing with them is.
Scribd would rather increase the amount of content on it's servers by making uploading easy and free. While the guy that publishes the article is really the Scribd user, he can publish elsewhere if the price makes it too much of a hassle. The viewer has less of a choice, if it's only published on Scribd it's either pay up or go without.
I don't agree with their choice (or methods) but it makes sense for them to do this.
I meant to suggest a strict superset of their current system- keep uploading easy and free, and the reader or publisher can pay for access after the "free" period. Scribd seems to have had a pretty large hook in this guy- I'm sure he's not the only one.
I think the takeaway is a lot broader: never assume that a company won't do evil things with your data if you don't have a service contract with them saying that they can't. See also: http://xkcd.com/743
charging from the beginning actually scratches two itches: the business can evolve steadily, the revenue benefits the company. And second, as a paying customer I have exactly those "rights" the OP would like to have in this scenario. Trying to implement the early-pay-plan is a whole other story, and I doubt that it would have worked for scribd, had they started charging money without any content. classic chicken-egg-problem
* You have to charge from the very beginning. If you start a free service, and then try to establish a pay system afterward, your users will feel tricked and trapped and they will rebel loudly. Scribd seems to have been in a hurry to get adoption, so they made it free to host documents; as this guy said though, he much preferred hosting with Scribd over doing it manually on his university's web server. That could have been Scribd's value proposition, and a small yearly fee for that probably would have worked OK.
* SaaS could get itself in trouble if there are too many incidents like this. I already hear from clients that are concerned about using online services; the most common questions are, "What if they change their terms?", "What if they go away?", and those are legitimate concerns. Many of my clients aren't the most computer-interested people, so if they have concerns like that, then that means that stories like this have penetrated very deep into the consumer market.