Have you looked at the rate limits? They changed it from 60 calls per user per client ID to 100 calls per minute per client ID. Nothing is getting built with limits that low
Something isn't well thought-out here. The pricing? The usage?
$49 per month for one million records, one million calls over the API and only 20(!) sessions. I don't think there's a consumer or business app that I could build that would stay under the session and call limits.
This kinda sucks BUT I would take it over any Intuit API any day. They have a simple REST API that works exactly like you would expect. I assume if I really need more than 1000 calls a day I could probably contact them and get it raised.
I'm hoping the data available at the "premium partner" level will come down to the basic users as time goes on, and it's just to make sure nothing is overwhelmed the first week or so.
For those curious, the API limits for basic users:
Thanks for the feedback. We do not have any immediate plans to change the call limit for the preview offering. We've seen from our data that 1K calls / month are sufficient for most developers who want to get started using the APIs. We we will certainly consider revising this in the future if there's strong demand. I encourage you to file this as a request here: https://cognitive.uservoice.com. We review user ideas and feedback there on a regular basis to improve our services. Thanks again!
The pricing model doesn't scale realistically and would require a subscription service for users. An app with 1M+ installs could do 1M+ calls per day making this service $24k / month.
I see you making this claim in multiple comments that he's making 600 API requests per user per hour to check for notifications, but this doesn't make sense to me. As you said, 100k users would result in 60 million API requests per hour, which would be about 43 billion API requests per month. I've seen pricing of $0.24 per 1k calls, so that would be $10 million per month, not $20 million per year. And that's just for this notification thing, which he could theoretically kill as a feature if that's 99% of the problem.
this immediately made me think, don't the amazon API limits prevent one from making more than a few thousand calls an hour. Would be very interested to learn how you are countering that. I simply avoid doing anything with Amazon's product api just because of that reason.
The problem is that it doesn't cost much to call someone. If the costs rise, then it lowers the rate at which people call. Not a bad idea, if only it can work for everyone!
In some rural areas there are still extremely high ratecenters that are costly to call. Looking at a ratedeck, 1240655 (Maryland), most Montana numbers and especially the Northwestern Territories (up in Canada) are the costliest to call, ranging from a few cents a minute to tens of cents a minute. Most flat rate VOIP operators won't deliver calls to those numbers, and if they do, they'll close your account if you call there too often.
The FCC has been cracking down on Traffic Pumping (which is what FreeConferenceCall, AOL, NetZero, etc were all created to do) to these prefixes, along with fake ringing that companies including T-Mobile do to make it seem like the call is being connected to these high cost areas, meanwhile the person your calling doesn't ever receive the call (or it gets through after attempting to call for multiple minutes).
Why aren't we charging $0.2 per 1k API calls plus $0.5 per privileged calls? Where are my revenue share programs? You guys are leaving so much on the table!
reply