Help

Can you scale up with Airtable? 5x/s limit

Topic Labels: API
1286 2
cancel
Showing results for 
Search instead for 
Did you mean: 
Matthew_Borton
4 - Data Explorer
4 - Data Explorer
Hi all!
Is it possible to scale with airtable?
 
I'm looking for recommendations on how to get around airtable's 5x / second limit API call limit.  Looking for a tool that can queue up the requests for me and send them through under the threshhold. I am using Zapier with airtable for a lot of automations.  I'm worried as I start adding my new clients then it will break.I've found two tools that an do this. Anyone have experience?
  1. https://nocodeapi.com/ - looks promising but customer support is atrocious.  How do I know i can trust them? Anyway their product signup doesn't even work.  Website has many problems.  Not trusting them with this important info.
  2. https://rapidapi.com/aircache/api/aircache - no experience with them. Again could work.
  3. Others?
2 Replies 2
Jack_Manuel
7 - App Architect
7 - App Architect

I use nocodeapi for Instagram post gets. They're decent for that relatively low risk function but I think your instincts are right, I wouldn't necessarily trust them with high-volume or sensitive data - there has been some down time.

For your purposes I can suggest https://sequin.io/ - I've not used them in production (just built some test cases) but it could be exactly what you're after.

Hi @Matthew_Borton ,

I generally use Make (cannot stand Zapier -slow to build, poor branching, need a step for formula, horrible debugging) or Pipedream (when you need to add some custom code or have higher volume of transactions and Make gets too expensive). In both cases don't remember ever rate liming being an issue.

Even if, both Make and Pipedream can run sequentially to avoid rate limiting, so that would be 5*60*60*24 = 400k+ read/writes per day at max - seems a lot.     

Going further not all reads actually need to go to Airtable. My own website is 100% on Airtable ( https://business-automated.com/ WIP... ) but it is statically generated on Vercel, and revalidation is set to at max 1 per hour - meaning it will not run into API rate limit.

If you built e.g. customer facing front on Softr.io I believe they also have data caching.