The answer to your goldfish theory is very simple. The new node selection is a game changer. You can’t compare the product more than one year ago with the current version.
And the new node selection has somehow made it more expensive to select nodes? The fast nodes get chosen more, so…how does that relate to the end-user pricing?
I am reading the annoncement more as an proposal that still needs to get confirmed. So at the moment customers paying with STORJ token will also get the 5$ minimum fee.
For now we can only hope that the proposal gets approved.
Lets say there always have been price negotiations with the bigger customers. One year ago the performance was good but not mind blowing. Customers told us they switch over because it was cheaper. Now one year later the response is different. Not because everyone is a goldfish. It is because the product has envolved.
What you are saying is that one year ago the bigger clients didn’t choose us, but now they do. The same bigger clients that are choosing the select network (which is more expensive, twice as much for stored data) with a lower RS value (which is more profits for storj, since if you charge X to store 3 segments, but now you charge the same X to store 2 segments, that means you are getting the same money but spending less).
I’m curious on how much the account overhead costs change if the customer is paying in STORJ vs regular money.
This is not correct, accounts with zero usage will not incur the minimum fee for that month when there was no usage.
Can we get confirmation on the adjustment to the minimum fee being adjusted to $0.50 for users paying in STORJ token now too?
It sounds like exec team will meet in the next 24 hours and hopefully give us their decision.
My guess here is: someone stores on the order of 10M files/segments, most of them smaller than the threshold for putting them on storage nodes, hence stored on the satellite. Satellite is probably using Spanner, which for just storage of 100 GB is in the order of ~50 dollars.
Was there no prior discussion within StorjLabs of implications of such an abrupt change of billing structure, or was the idea floated Friday morning and implemented Friday afternoon?
Realistically the new pricing looks inline with other products, new pricing is unlikely to effect any businesses using product, I wouldn’t change my business use for an increase of $4/month.
I would guess that not many business accounts pay by Storj tokens, be to difficult for most accounts departments, especially the ones that are only a few $ per month.
It’s all about the “feeling” that StorjLabs “cares” about “ME” as a customer, by keeping me informed about upcoming changes, giving me time to investigate/implement alternatives if I desire. Trust that it’s not going to change again next month, or the month after. We have budgets, and other bills too.
Look at how Netflix handled the prolific abuse of account sharing as a model, rather than how VMware treated it’s customer base.
Never forget, every customer matters.
This is already supported since a while ago.
I exited a startup that thought so. 
The fact that an edge case like this can even exist points to a core business model failure. I suggested in the other thread that the S3 gateway service could be gated behind the minimum charge. Based on the overall data here, the S3 may be the root of all of your problems, kill it unless that service is paid for directly by usage, or charged separately.
The self-hosted gateway is fairly simple to setup, and for customers using that for internal data it should actually be preferred rather than your gateway: Less hops to their data, and likely higher throughput than proxying through you.
For customers using the gateway to publicly serve data, they’re using Storj as a CDN, which is fine, but the raw egress costs don’t pay for that kind of service.
This is what segment fee is there to address. Maybe it’s too low.
But either way, storj cannot have a cake and eat it too: either you charge for everything, like Amazon, so that every customers pays fairly for what they use or you try to charge everyone predicted average usage but then you can’t complain that some users are causing enormous load — because some others underutilize the system and it evens out.
As a customer I prefer Amazon approach. Why shall I underpay for what I use and have other users subsidize me, and in the same vein I don’t want to pay for others!
Storj chose the variation of the other approach - no api costs, but they do encourage large files though the segment fee. If this on average does not work (and we are not talking about couple of customers — in does not need to work on everyone) — turn the knobs and adjust strength of encouragement.
I agree, not every customer matters. It’s a fallacy to think otherwise. But the approach shall be consistent.
Of course storj can fire bad customer, the same way customers can fire their storage provider. But needs to be weighted against publicity damage: firing customers that stay within the terms but are inconvenient does not send a good message
This is discouraging. My personal website (just my contact information) is hosted on Storj, and I got on my account to check the status:
Stored data: 60kB (yes, kB, not MB or GB…)
Download (last month): 2.04MB
$5 is way too much for me. Even 50 cents is too much. Paying 50 cents for 60kB is costing $8.53 per MB.
Sadly, I will offboard…
And that’s fine. Why would you pick storj for such a tiny web side? You are not taking advantage of any benefits storj can offer. Seems like you picked an ill-fitting product for what you need, even before cost is considered.
Use cloudflare pages to host your web site. It’s free for personal use and it’s an actual CDN.
It kept me involved with the platform. I wrote some PoC programs that needed refining but worked: backup scripts, a photo export app, CCTV backup… Having my site on Storj kept me working with Storj: when you have a hammer…
Now I don’t have much free time to thinker with those projects as much, and offboarding Storj means I probably won’t resume working on them: when I do have more free time or ideas, thinking about reopening the account, buying tokens or funding the account, create API keys again, download the SDK and everything else will make the idea of resuming the projects feel a lot less enticing…
You never know which customer will become your next cash cow. Could well be the guy(s) costing $160/month - testing before he makes a deal.