@zach We are approaching the 250 units imposed with last year’s plan, but we were happy to pay the $0.40 / mb. We recently switched to the tracker and began developing with it in earnest due to the advertised $6.99/ /mo plan with no need to migrate to the enterprise. I just spent 10 hours a day for 3 months programming a relatively complete control system and HMI that uses the particle tracker and a custom I/O daughterboard of my own design.
With ONE SINGLE ANNOUNCEMENT, ALL of that now goes in the trash.
You took the least complicated and least expensive plan and flipped into the most expensive overnight. The tracker product hadn’t even been out a single year yet and now my 600+ hours of development are toast.
The 230+ units we have sold to date operate in the gasoline distribution industry and are being used to monitor our devices in other countries. We have corporate and governmental data reporting requirements and contracts with multiple international companies. We are required to take 172,800 readings / month (once every 15 seconds) and we were able to accomplish this with 30-40 MB / mo, keeping the data charges we incurred to less than $20 / mo.
NOW… 720,000 divided by our 172,800 means that I can only service 4 devices with your new plan. I think I calculated that I will need 28 “blocks” of your new data that will cost me roughly DOUBLE what it was already costing me. If my data rate becomes $40 / mo (or higher – there are other ways to calculate this), then my project is completely out of business.
We love the Particle devices and I have championed the particle system in every technical conversation I have had, but this pricing is going to kill our endeavor.
Looks like sufficient info is published to conclude my $35/month bill (prototyping stage only) I spent a lot of time cutting in over half, will automatically become a $299/month bill.
Particle can’t really compete when there’s a superior Boston/New England area new competitor offering 10-year-500mb-built-in CAT-M and CAT-1 cards for $49 one-time, zero further cost.
In two ish years I could be making an actual commercial product with LTE and unfortunately the new pricing model seems like the opposite of what Particle would have needed to do to retain my business.
The Particle people though are so nice and helpful and I commend them for their gentility and great customer service.
To be honest the biggest mistake is the hardware design of the NA LTE Boron and not using the far superior BG95 modem. The product has been far less stable than first competitor I adopted, and I will be interested if Particle comes out with a totally overhauled LTE module hardware/data plan in the future that competes with its now two surpassing competitors.
Hey, not trying to be a pain, but offering another way of looking at this.
Is this superior Boston/New England area new competitor offering cloud infrastructure that allows you to manage your fleet of devices? If it is, then yeah, it's a good competitor. But if it's not, then the comparison is missing the features that Particle offers on top of the connectivity layer, OTA updates for your firmware, cloud/variable functions, etc.
@gusgonnet I think consumers will explore all your valid points judging against all three leading options for this sort of thing, comparing to their needs, and the beauty of the market will sort out a just equilibrium. But yes, good competitor. I wish Particle well and shouldn’t use their forum to evangelize against them.
But for me, $0.1mb/month LTE and the above blew away $0.4mb/month already, let alone now.
Hi @zach, I see you’ve answered this to several folks who argue their business model breaks completely with the new pricing.
I was wondering what exactly “find a good path forward for you” means, especially since I think there may be many more people out there who may not read these forums and be aware there is such an option, or who will not speak up about these changes.
We are working out different solutions for each person depending on their circumstances; we proactively sent emails to existing customers who will be on the Growth plan when we announced the new pricing model.
@zach Thanks zach. Why was the decision made to force people in my boat off the platform where my previous prototyping bill, see above, was already demonstrably 4x the competitor’s rate (which offers better and more stable hardware than Boron LTE as well), and now would become over 25+ times the competitor’s bill absent a special dispensation?
My $35ish of $0.4/mb data @ over 100k DOs but less than 720k now equals $299. Competitor #1 charges 1/4th of current (offers $0.1/mb instead of $0.4/mb). Competitor #2 offers $49 one time only for the LTE hardware module itself and 500mb of data, 10 years prepaid.
Could you explain why you charge now so much more based on the abstract notion of Data Operation instead of total bandwidth used, when the latter is what actually physically corresponds to your carrier costs and is how competitors charge?
It seems like I must be totally missing something in my analysis here. I doubt you could have taken this large step without evaluating the market landscape of competitors, so feel free to note whatever giant error is permeating my reasoning.
If I review your threads above, your concerns are driven by attempting to make a price comparison for alternatives based purely on a per-megabyte cost. That assumes that the only data that is being transmitted over the wire is your data - the information you intend to send from device to cloud or from cloud to device.
The problem with the practical application of an IoT product is that your expected data rates don't match up with actual data rates. That's because there is unanticipated networking overhead, data that you pay for because devices are attempting to establish connections, failing and retrying, encryption overhead, etc. This overhead can be quite substantial, in many cases consuming far more data than the messages themselves.
One source of value from our platform comes in the EtherLink services we provide (that Will presented at Spectra) that handle a lot of the networking layer automatically. As a result, we can deliver secure, scalable, and reliable networking without effort on your part to manage the device connection.
That said, since those services are automatic, they are outside of the user's control. That creates challenges for estimating data consumption, especially since devices in different networking environments might consume different amounts of data; a device with a weaker cellular connection will consume more data because it has to retry messages more frequently to ensure message delivery.
We believe it's our responsibility to ensure a reliable connection, but passing along that cost to customers as part of their variable bill doesn't seem right since it's outside of their control.
That's why we are switching to Data Operations - because it puts customers completely in control of their costs.
For me, I feel like Particle offers a great framework for rapid prototyping and especially in-situ development. Unfortunately my area doesn’t have reliable cell coverage. I have devices publishing on WiFi, Lora & LoraWAN.
On WiFi , to keep my Particle.publish count low I use UDP to send frequent data to an on-site server. A few times a day I also Publish to particle to keep my device status up to date. Unfortunately it means I lose most of the cloud functionality for high resolution data, but the great thing is that grafana loading data from a local machine is very very fast.
I also use statistics to reduce the amount of data stored. On well pumps I sample average current (Amps) at 12Hz, track the min & max values and only log them every 15 seconds. This way I still capture failed starts or drop outs. On Grafana I group time by max of max and min of min and it tells the story pretty well.
I have other sensors that only log data if there is a significant change or if a timeout is reached. For instance I have a tipping bucket rain sensor that logs every mm change or at most 24h after the last datapoint. I send the tip count rather than a tip event, so the data always increments. I can tell within 24 hours if the sensor is offline (hasn’t happened yet), easily identify errors/resets, I get high resolution, more or less realtime data and I’m not transmitting or logging a bunch of no-change events.
LoraWAN has been a mixed bag- it’s easy to setup and integrate with TTN but the LoraWAN header wastes way too much airtime in the US. It’s really only well suited for publishing data less than once an hour.
Lora works well and is compact but you have to build your own IP gateway to ingrate it.
I sympathize with those who are get bitten by Particle.publish counts- it was one of my hesitations when starting with Particle. I wish that Particle.publish could be directed at a personal server instance. I suppose the reason that the paid tier is such a jump from the free tier is that someone did the math and said a $30/year tier wouldn’t generate significant revenue.
In the mean time Particle is still a great way to do very quick development and I’m happy that OTA updates on WiFi and within Cellular data limits are free.
Hello Particle. As a customer, programmer, seller and developer with Particle, my business started with you many years from now. We have invested money, R&D and we have also invested a lot of time to help to develop the Particle platform that we use today. We were from the beginning beta tester and Particle product come a reliable stuff with the participation of business like me.
Your new announcement is in fact a jailbreak for all of us. 90 days to let us decide if we continue with Particle. 90 days to inform all of our customers that we have a contract for 2-3-4 years in advance that we have to double, triple monthly fees. 90 days to down a business.
For comparison, we are customers of darksky.net weather API. DS was bought by Apple last year. DS sent a message about the transaction and informed us that service will be shut down on 31 december 2021. Not 90 days, maybe 1 year and half. That’s crazy you jailbreak us in 90 days.
I don’t need to enumerate frustration and questions, many of us have already commented here. I can’t believe again this new announcement and I hope that you will consider it again to apply some change or give us more time to think about it.
Looking at the new announcement vs the qty of devices we use, Particle stuff can’t be reliable for our business. I don’t know also how a company like Jacuzzy and Scooter will be able to stay competitive.
I could accept that, and really going from 30 cents/device to $100 per 100 for WiFi and $3 each to $300 per hundred for cellular isn't worth getting upset over.
The issue is their ridiculous limits. 720,000 data operations for 100 devices is 7,200 operations per device per month, or 240 per day, one every 6 minutes. And using a webhook to request data where you want a response is at least 2 operations. Any semblence of near real time monitoring is no longer possible with particle. Years ago, Particle talked about vineyard monitoring; just asking for a weather update every 30 minutes (using a request/response) will be 96 of your 240 daily operations. And it is, frankly, quite insulting that they pretend the $300 plan is for 100 devices -- 10 might be more accurate, but I guess it's bad PR to publicly state they're charging $30/month for a cellular IoT module. 100 is such a ridiculous number, they might as well make it unlimited and then only charge for the data operations -- but of course that will also just emphasize how expensive the data operations are.
The plans should offer at least 10 times the limits, 1 million for free, 7.2 million for paid, and 20 times would be more reasonable. And 450kB of cellular data per device is similarly just plain bad, that should also be 10 times higher on the paid tier.
I think you're right, under the new scheme, Particle is a great option for quick development using the free tier to develop a product to show investors/VC, but then the funding will have to factor in dev costs to get the live product off Particle.
Why must the “blocks” have so steep increments, both in device count and cost?
A sample case similar to my situation:
With 199 active cellular devices, I’d be paying $600/mo (2 blocks), which is actually pretty much the same vs. the old (current) $2.99 per device, at ~$595.
The problem is when I need to activate 2 more devices, going to 201 total. Just because of that one additional device, I would need an entire new block at a whopping $300/mo extra just to support that one device, when with the current model that’s just $2.99 more.
The Growth plan in my opinion promotes extreme overspending and underutilization, and that is just considering device count, not even mentioning data usage/operations… I just can’t see how this kind of pricing means “designed for small businesses and startups” both in terms of cost and scale-up.
I’m using few argon for a special project. My issue is that I use that for specific events during the year.
Events duration are between 1 to 5 days, and have 1 to 3 events like this during the year.
Using max 10 argon it would take max 500 000 D.O for an event.
My question is, is it possible to take a block on for a specific month (99$) and then go back to the free plan? I mean i need more than 100 000 D.O only during specific months per year. I’m not willing to take one block for one year (99$ x 12) as i’m not making money of this project and this would kill the project.
If i have a event during July and another in september, i’'m fine to take a 99$ block for july and another
99$ block september, but not for an entire year.
@zach I have 12 wifi devices and they publish one message ever 2 mins unless there is an issue then its like 4 or so extra messages in the short time frame. By my math that’s 260k a month or more. That went from free to very expensive.
Thanks for the question. Yes, you will be able to use Growth blocks for individual months as there is no requirement to make any commitment in terms of duration when purchasing a Growth block.
I don’t really see anyone talking also about the data limits on the cellular plans. It was stated multiple times that things like OTA updates, network reconnections, etc. were not data operations, and thus free. If you read the fine print, there is a ridiculously small data cap on the cellular plans, and if you go over (at least on the free plan) your entire fleet is shut down for the month.
If you dig far enough in to read the fine print, it states that OTA updates and all the other things that are supposedly now “free” count against your data cap. There is no option to pay overage fees for data. You just get shut off.