Hey folks, has anyone else noticed a performance dip in Claw as they've added more custom actions? We're up to about 150 now, and I swear the UI feels laggier when selecting from the dropdown, and there's a noticeable delay when it's fetching the action list for a new workflow step.
I love the flexibility, but it's starting to impact our team's velocity when building complex email nurture sequences. We're deep into HubSpot for our main marketing automation, but we use Claw for some really specific, API-driven actions that sit outside of it. The integration itself is still solid, but this slowdown is a new pain point.
I'm curious:
* Is this a known scaling issue with larger action libraries?
* Are there any best practices for organizing or "chunking" actions to help performance?
* Has anyone moved to a different platform for heavy custom automation because of this, and how was the migration?
Trying to decide if we need to architect a workaround, prune our library, or start evaluating alternatives. The last thing we want is for a tool that's supposed to save time to start eating it up.
MartechMatch
Oh man, the dreaded action library slowdown. We hit this exact wall at around the 180-action mark. It's a known scaling issue they've been working on, but in the meantime, the UI lag is real, especially when you have a big team opening dropdowns all day.
What saved us was a two-pronged approach. First, we implemented action "categories" using a naming prefix system (e.g., `CRM_Update_Contact`, `Analytics_Log_Event`). This didn't fix the fetch time, but it made scanning the dropdown mentally easier. The bigger win was lobbying our Claw CSM for access to the beta of their "tag and filter" system for the action library, which lets you filter the dropdown. It's still in early access, but it cut the pain significantly.
Have you tried auditing how many of those 150 actions are actually used in live workflows? We found nearly 30 were legacy or one-offs we could archive, which helped a bit.
test the migration twice
That's interesting to hear about the HubSpot integration staying solid. We're on a similar path, using Claw for the niche API stuff our main platform can't handle.
When you mention evaluating alternatives, are you mostly looking for something with better UI performance at scale? Or are other factors like pricing models for large libraries coming into play? I'm worried about the cost jumping if we just replicate our 80+ actions somewhere else.
We're not at 150 yet, but this has me thinking we should prune our library proactively. Maybe archiving older, unused actions more aggressively could help?
Still learning.