Issue with scale plan

Hi all,

I just wanted to submit this complaint about the lowest-tier payed plan, which, from my understanding, is the Scale-Fixed plan, capped at Tier-1.

The main reason I wanted to upgrade was to increase limits from 50 requests per minute, and to remove the free-plan banner on the UI-builder web app.

The story of how I got here:

In January, I was burned pretty badly by subscribing to the scale variable plan. I didn’t realise that high levels of requests, that would happen while development, would push me into higher tiers. While I expected a bill of $15 per month, I ended up with a bill of $70. Thankfully the person I’m building the app for was happy to cover those costs, but from my side it felt very unfair.

I then switched to the Scale Fixed plan. The next day I realised I had messed up again, since I had an invoice of $22 come in, which meant I was paying more for one day than I would expect for an entire month. I quickly found out that I needed to cap the requests at “Tier 1” to limit unexpected price hikes. The app is not especially a high-usage app, and we can optimise further, so I left it like that.

Now, my client is telling me that the app is not working at all. Indeed, I have realized that the “Tier 1” has only 5 API requests per minute, which is not enough to load a web page if it includes icons, scripts, and custom components. Thus I’m paying for a completely broken service.

I’m saying this as my only resort now is to switch back to the free plan, which has 50 requests per minute, or to self-host a backendless pro instance, which I’m also setting up in parallel. This seems completely counter-intuitive as I’d expect an easier time and a better service if I was paying, even for an entry-level pricing.

Please consider this as feedback for your services and pricing. If you have any questions I will be happy to answer.

The page not loading due to 429 errors:

In order to match the free plan, I would need to pay $55 per month, which is a lot for a small cloud application. Let alone the fact that I would like a higher API call limit per minute.


Thank you for the detailed feedback - I can see how frustrating this experience has been, and I appreciate you walking through the full timeline. I want to make sure the full picture is clear, because the comparison between the free plan and Scale Fixed Tier 1 is more nuanced than it might appear.

You’re right that Tier 1 on Scale Fixed caps at 5 requests per minute, which is lower than the free plan’s 50 req/min burst limit. However, the free plan also has a hard limit of 1,000 API requests per day, along with other significant restrictions - 10 data tables, 15,000 database records, 1GB file storage, limited event handlers, and more. So while the free plan allows short bursts of higher throughput, it’s designed for development and prototyping, not for running a production app serving real users throughout the day. All of this information is published on our pricing page and in our documentation - we don’t hide anything about how the plans and tiers work.

I understand that $55/month (Tier 4, up to 50 req/min) may feel like a lot for a small cloud application. But if you consider what you’re getting - a managed database, file storage, business logic hosting, API infrastructure, UI Builder - and compare that to the cost of provisioning and maintaining your own server with equivalent capabilities, you might be surprised how quickly those costs add up and surpass that amount.

You mentioned setting up a self-hosted Backendless Pro instance as an alternative. You’re absolutely welcome to run the Backendless Community Edition for free, but I’d encourage you to go in with realistic expectations about hosting costs. Between compute, storage, networking, backups, and ongoing maintenance, things get expensive quickly in server-land. Unfortunately, we also have bills to pay to keep the platform running, and our pricing reflects the real cost of the infrastructure and services we provide.

Happy to help you analyze your usage and find the right tier if you’d like.

Regards,
Mark

Thanks Mark for the understanding and the thorough reply. Just so you understand, on our side this is an application designed to be used in a working environment (a local shop) with only a few interactions per day. The interactions are complex since they require uploading photos, OCR, AI analysis and user-correction, but it is designed to be handled in the span of 2-3 minutes (cutting their IT work down from 30 minutes and up to multiple hours). This means typically less than 1000 requests per day, but sometimes more than 50 per minute.

Since the benefit of a paid plan will be minimal compared to our allocated budget, we decided it’s better to hit those limits and catch up on work in following days since the load is rarely that high. In the long term, since we have close partners that actually do VPS hosting, we will have the bare-metal to host an instance of the BL Pro community edition, even if that would mean we will migrate flow runner work to API services or other.

Totally understand. Thank you for the additional context. Quick heads-up on a change that is coming so you can start making any architectural changes if needed - FlowRunner will be splitting from Backendless soon into a separate dedicated service. An announcement with more details is coming soon.

1 Like

Thank you Mark for the head’s up. I wish you luck with the FlowRunner project.