In my opinion, Fibery Automations have significant room for improvement. Iâll break my thoughts into two areas: Automation Workflows and Automation Rules.
Automation Workflows (Flows)
A Flow is a sequence of rules working together to accomplish one larger process. Fibery doesnât currently support this concept in a convenient way. To simulate a flow, you must create a separate rule for each event â which becomes cumbersome for complex processes like email nurture sequences.
Flows can occur instantly (e.g., on a state change) or over time (e.g., in a drip campaign). While Fibery can technically handle both with multiple rules, managing these rules quickly becomes difficult.
Ideal Feature:
A visual âFlow Builderâ where clicking a rule opens its corresponding Fibery automation would make this far more intuitive and maintainable. This builder might not be database specific. Multiple rules might reference different databases as one ruleâs output might be an entity from a different database. Also, placing flow functions like splits, drip sequencing, time delay, or A/B pathing would be sooo hot.
Note: if you go down this route, I would love to help design this.
Automation Rules:
Fiberyâs automation rules are generally solid, but I often hit limitations that slow development or force workarounds.
Key Limitations:
No âIf / Elseâ logic â only âIf.â
Flows would help address this gap.
No reusable inputs/outputs between stages.
Only the first stage has defined inputs from the trigger. Because of this, temporary entity updates and multiple chained rules are often needed. Flows would solve this too, because outputs for the rule would be referenceable.
Limited support for looping over multiple entities.
For example: âIf X occurs, do Y for each linked entity.â You can update related entities, but not easily create new ones per linked record.
This capability would be extremely helpful for:
Building project & task template automations
Supporting workflows such as ordering and fulfillment
Top Wishlist
If Fiberyâs new automation system were integrated with a native email and SMS marketing tool (tracking opens, clicks, unsubscribes, etc.), Iâd rarely need another CRM. Currently, Fiberyâs email integration is functional but limited.
So, as promised I would respond to this topic. Hereâs my thoughts after using automations a lot for the last weeks for generating PDFs.
Generating PDFs is cool but the use of it is limited because I canât shrink images and I end up with huge PDFs. That is the most painful thing still as I have shared and I wonât repeat more than I have (until now). On the bright side, I really like the ability to HTML + CSS + JavaScript to make PDFs. And the Fibery AI also makes this more accessible.
I am working on compensating my pain with n8n I think itâs great that I use that now with Fibery.
Good question. A hard one because I find better automations very important. But it is not critical because n8n. You probably expect me to answer âbetter mapsâ but that is not it Perhaps a Fibery managed n8n environment? So users who are non-technical can just go there if they want more scripting? Can be an addition to the price or people can get monthly credits.
What use cases you imagined with Fibery, but canât do them because of poor automations?
I would like to be able to use the old value of a field when a trigger happen.
I would like to show a prompt to the user if a condition is not met. For example, if a user close a task that contain active sub-tasks, I would like to ask the user if he want to close all sub-tasks or if he want to keep the task active.
Be able to have a Webhook to send messages to Google Chat.
Be able to filter on current connected User properties.
Do you compensate that with n8n, API, Make, Zapier or other external tools?
No.
What a single thing more important for you to have in Fibery instead of better automations? Please do not provide many, just one
Integration with Salesforce (Sync any object records, two way sync).
We use scripts in fibery to make API calls and trigger events in other software without the use of Zapier, but we have to use Zapier to catch webhooks from them. Would like to be able to catch hooks directly in fibery and use the information to automate things.
AI Interaction With Files
We currently have to send files out to Zapier to use AI to review them, then bring the responses back.
OCR On Files
We currently have to send files out to Zapier to use OCR to extract data from them, then bring the responses back.
Looping
When we have a form response with multiple items we use looping in zapier to create a fibery entity for each item. Having the concept of looping in fibery automations might be useful.
Branching
Zapier supports pathways which we find useful and might be worth you looking at including in Fibery instead of just the yes/no filter and single pathway you currently have.
Forms
We rely a lot on forms to run our business. At the moment we have to use Gravity or Cognito Forms and build Zapier connections to get that information into fibery. Our number 1 feature to work on ahead of automations would be to make fibery forms good so we can just use those!
I canât check the person who triggered the rule before it is triggered. âAdd Filter for Entityâ doesnât allow âWhere User is [function]â. A use case is âa field is updated, but what happens next depends on who updated itââsuper important for true workflows.
No. I will though if there is not a native solution.
Branching and conditionsâinternal logic (if then else) and button display rules. You said one thing, but these are so entwined itâs not funny.
I shared my feedback already. But just want to emphasize how annoying and limiting is the following thing - Iâd give up everything I wrote for you to fix this one:
Hi. There are suggestion about improving fibery. We want to know if what is the library/engine fibery use for its formula. Like are they using âHyperformulaâ? At least we wish to share what features to add and make less friction to the dev team in case they want to add some of our ideas in the product roadmap.
I wish SaaS apps did not invent their own formula/expression or workflow engines.
Itâs less friction in the short term, but becomes a massive time sink as the system and usage matures.
It feels like it should be possible to build on general frameworks like Jinja/Nunjucks for expressions or xstate/bpmn for workflows. It would still be a ton of work of course but at a higher level of abstraction hopefully.
Unfortunately it will never happen. Well, maybe in 20 years when it will not be needed at all and all work will done via API calls between AI Agents. Otherwise, all vendors will keep building own UI Views, own backend solutions, own DSLs
Itâs theoretically possible that we could add âexternal actionsâ that allow you to create automations which push information from Fibery to Google Calendar, but fundamentally, this is not the same as true two-way sync.
With two-way sync, there needs to be functionality to take care of conflict resolution, e.g. someone changes the calendar entry in Fibery and another person makes changes to the corresponding entry in GCal. If a Fibery automation pushed changes to GCal, it could end up clobbering the changes already made in GCal.
2-way sync is hard
I think itâs great that I use that now with Fibery.
Perhaps a Fibery managed n8n environment? So users who are non-technical can just go there if they want more scripting? Can be an addition to the price or people can get monthly credits.
