The reason why Odoo is gaining mometum is because of thier pricing and its pyschological , clients will not even see its more expensive than ERPNext
Iâve found the costs make more sense once you account for server upkeep, backups, updates, security patches, and quick support. Self-hosting saved money but needed lots of my time.
Clearly learning operations is an effort.
If it can be shared between many entities, it naturally gets more efficient. Thatâs using the âcopy for almost nothingâ capability of modern computers for the benefit of more than one organizational entity (if âoneselfâ or otherwise), just as serving a community instead of just oneself naturally is the base of almost any economic activity and efficiency in society.
Running on a cloud, how can you even be sure that the code you interact with is the FOSS code you think it runs? Thatâs not a given, and you might not be able to ascertain it â depending on the type of access you have, of course.
And how can you assure the datacenter owners share the ethics of Frappe, or of yourself, or even of the country you are in?
Depending on your projects and the cultural/political/legal/etc. environment, these questions might matter.
Just looking at cost can only under very specific conditions be the ultimate and best criterion. In other conditions, it might be a trap, an invitation to become an inactive blob, a slippery path to unlearning self-reliance, to give up data, control, confidentiality and oversight, the ability to interact yourself with the system, to repair it and keep it running, etc.
In my case Oracle Cloud, I manage around 20 clients running small, custom ETL apps built with the Frappe framework. Paying $25 per site just to enable custom apps â or $500 per server â becomes hard to justify for these lightweight, internal tools.
I currently use Dockploy to manage and deploy these apps more efficiently across my own servers, which has been a great and affordable alternative.
I was hoping Frappe Cloud would eventually offer something similar â a more flexible, multi-cloud or per-server pricing model â so we could balance cost and performance without losing the convenience of managed Frappe instances.
I went through the same thoughts and ended up self-hosting to cut costsâit takes time to set up but saves a lot over managed hosting. I noticed some nextcloud providers offer better pricing on VPS and storage, so I started there and added ERPNext manually. Itâs not plug and play, but if youâre comfortable tweaking things, it works well and gives you more control.
We are solving this problem here: geniusdynamics/ns8-frappe-press: Full service cloud hosting for the Frappe stack - that powers Frappe Cloud
As well as Here: geniusdynamics/ns8-erpnext: ERPnext Module for Nethserver 8 Free and Open Source Enterprise Resource Planning (ERP)
Through Nethserver 8
For those who want to get Dirty, Managing their own servers, but still want the simplicity of doing it, without Significant Risks, during Updates, Upgrades, Management and Migrations.
I will still definitely suggest considering using frappeCloud as it offers both value for Money, as well as peace of Mind.
For those that want standard, single or multiple installations on a single server, Standard erpnext, via NEthserver 8 will solve that,
For those who want Simple to installa and manage, but clusters, frappeclud experience on their own server clusters and environment, for their onw Needs, then Frappe press, through NS8 will Offer that.
Otherwise, Just use frappecloud, WHat we have implemented, its the Simplest, it can be simplified by any standards. Challenge me on this one.
For those managing Dokploy, or coolify, you can also install coolify, or Dokploy, via NEthserver 8, and isntall ERPNEXT on a remote Dokploy managed server/instance.
Choose your Poison.
Frappe fills a gap at the lower end of the market. The problem is that frappe cloud does not sufficiently close this gap currently. There are many organizations who donât need a complete ERP system, but the platform and infrastructure provides a great and convenient way to integrate with other systems. The problem is that if you use Frappe for these use-cases it is often necessary to be able to make and add custom apps. The consideration is then between 25 dollar a month for Frappe Cloud or 5 for a Hetzner server. That is the gap that exists within the current pricing.
I see all the benefits of Frappe Cloud, but for some systems and use-cases there is no way to justify the cost difference.
I definitely agree.
For many of us, Frappe/ERPNext is the central nervous system of everything we do at work. Compromising that to save a few bucks seems incredibly short sighted.
But, like you say, Frappe is a great platform for rapid development of smaller applications, too. I recently helped my son build something to track match results for a school club heâs in, and it was a great experience. Frappe really shines with this kind of thing. But, youâre rightâŚ$25/month to support it would be tough to justify.
The challenge is technical. Arbitrary apps have arbitrary power, and thatâs not compatible with shared hosting. To host a custom app, Frappe Cloud presumably needs to give you your own bench. Thatâs a lot more resource intensive than a multi-tenant setup.
One possibility would be to extend site-level customizations (server scripts, client scripts, custom doctypes, etc.) There was some work a while back towards âbundlesâ of documents that could be packaged up and distributed. There could even be version control, git interfacing, etc. It wouldnât cover every need, but something like that might be a way to build an ecosystem for micro-apps.