Clarification on hubspot.fetch() API Limits for Public Apps

[MKatewa]

Hi team,

I'm working on a public app using the new HubSpot UI Extensions framework, and I came across the documented limitation that each app is allowed up to 20 concurrent hubspot.fetch() requests per account. fetching-data 

I had a couple of questions I wanted to clarify:

1. Does the 20 concurrent request limit apply:

   • per individual user accessing a portal, or

   • is it shared across all users using our app in that portal?

2. Is there any plan to increase this limit in the future to better support high-frequency use cases or is there any way I can increase the limit?

Would appreciate any insight. Thanks!

[GiantFocal · Accepted solution]

Hi @MKatewa,

The 20 concurrent request limit is shared across all users within the same HubSpot portal or account. It's not per individual user but a portal-wide restriction. This means if multiple users are using your app at the same time in the same HubSpot account, they all draw from that same pool of 20 concurrent requests.

I don't believe there is a way to increase this limit. As a workaround, here are some ideas you can implement to avoid hitting that limit:

1. Request queuing: Set up a queue system to manage requests and prevent exceeding the concurrent limit.
2. Request batching: Combine multiple operations into a single API call when possible.
3. Caching: Store responses to reduce unnecessary repeat requests.

[MKatewa]

Hey,

Thanks for the detailed explanation — that helps a lot.

I just had one follow-up to better understand the limitation:

The 20 concurrent request limit — is that measured over a specific time window (e.g., per second, per minute), or does it refer strictly to simultaneous in-flight requests regardless of timing?

In other words:

• If I make 20 requests at once and wait for them to finish, can I make 20 more right after?

• Or is there a rolling window (e.g., X requests per second/minute) that also factors in?

Just trying to understand how aggressively we can retry or stagger requests without risk of hitting a 429.

Appreciate your insights!