Airtable API pricing

Airtable API Limits and Pricing: What It Costs as a Backend

Affiliate Disclosure: We review products independently. When you buy through our links, we may earn a commission or a compound recurring commission at zero extra cost to you. Read our editorial policy.

Airtable is often attractive as a lightweight backend because teams can model data, build interfaces, and connect workflows without maintaining a traditional database server. The cost question becomes more complicated when the same Airtable base powers a client portal, reporting dashboard, internal application, or automated data pipeline.

The subscription price is only one part of the decision. Airtable also sets limits on monthly API calls, request frequency, records per base, and attachment storage. The number and type of collaborators who need access can affect the subscription bill, too.

These constraints matter because a system designed around occasional spreadsheet updates behaves very differently from an application that retrieves records every time a user opens a page.

This guide explains Airtable’s published API limits, how to estimate request volume, which costs to include in a backend budget, and when to consider a different architecture.

Scope: This guide is based on Airtable’s published pricing and developer documentation. I did not test it hands-on. Pricing and limits were checked against official documentation on October 26, 2026.

Airtable API Limits and Pricing at a Glance

Airtable’s Web API does not use a simple pay-per-request price for ordinary plan usage. Instead, the subscription plan determines the monthly API allowance and other limits.

PlanPublished annual-billing priceMonthly API calls per workspaceRecords per baseAttachment storage per base
Free$01,0001,0001 GB
Team$20 per user/month100,00050,00020 GB
Business$45 per user/monthNo monthly API-call cap125,000100 GB
Enterprise ScaleCustomNo monthly API-call cap500,000+1,000 GB

Sources: Airtable pricing, Airtable workspace settings and usage limits.

These are published reference prices and limits, not a quote for a particular organization. Billing can differ for sales-led agreements and legacy plans. Airtable also bills eligible collaborators according to the permissions and plan involved, so the number of people with billable access may differ from the number of people who merely view a shared resource.

Two distinctions are especially important:

  • Monthly API-call allowance: Free and Team have monthly limits. Business and Enterprise Scale do not have a monthly API-call cap, but their requests remain subject to rate limits and other technical constraints.
  • API rate limit: Every plan is subject to the same published limit of five requests per second per base. Paying for a higher tier does not remove this ceiling.

1. How Airtable’s API Limits Work

Airtable’s Web API is a REST API that returns data in JSON. Applications can retrieve, create, update, and delete records, as well as inspect base schemas and table structures.

A request to retrieve a list of records is one API call. Creating or updating records through the API also consumes calls. Metadata requests count toward the same monthly allowance, rather than receiving a separate unrestricted allocation.

Airtable returns records in pages of up to 100 records. Write operations can batch up to 10 records in a request for supported endpoints. Batching can therefore reduce the number of API calls needed for bulk operations, although it does not eliminate the per-base request-rate limit.

The three limits to track

Monthly calls: Free is limited to 1,000 API calls per workspace per month, while Team allows 100,000. Business and Enterprise Scale have no monthly API-call cap under the published plan limits.

Requests per second: Airtable limits the Web API to five requests per second per base. There is also a separate limit of 50 requests per second for traffic using personal access tokens from a given user or service account.

Records and storage: Even when the API-call allowance is sufficient, a base can reach its record or attachment limit. The relevant thresholds depend on the plan.

These limits address different problems. Upgrading may increase the monthly API allowance or record capacity, but it does not raise the standard five-requests-per-second-per-base limit.

2. Estimate API Demand Before Choosing a Plan

The most useful starting point is not the number of users alone. Estimate how often the application calls Airtable and how many requests each user action generates.

A practical formula is:

Monthly API calls = requests per user action × actions per user per day × active users × days per month

This is a planning estimate. Actual demand also includes background jobs, retries, metadata requests, administrative activity, and scheduled synchronization.

Example: A small client portal

Suppose a portal has 20 active users. Each user opens a dashboard three times a day, and each dashboard opening makes two API calls. Assume 22 working days per month.

20 users × 3 opens × 2 calls × 22 days = 2,640 API calls per month.

That demand exceeds Airtable Free’s published allowance of 1,000 calls per month. It fits within Team’s 100,000-call monthly allowance, assuming other activity does not push the workspace above that limit.

Now consider an application that refreshes several records repeatedly in the background. Its API demand may be much higher than a portal where users open a dashboard occasionally.

This is why a small proof of concept can appear to work on Free while the same architecture becomes unsuitable when the number of users or frequency of requests increases.

Build in a margin for real-world activity

Do not select a plan based only on the average workload. Account for:

  • Scheduled synchronization jobs.
  • Retries after failed requests.
  • Multiple tables or linked records needed to render one screen.
  • Metadata and schema requests.
  • Bulk imports and updates.
  • Traffic spikes during reporting periods.

A forecast should also distinguish interactive requests from background work. A delayed report refresh may be acceptable; a request that blocks a customer-facing page may not be.

3. The Five-Requests-Per-Second Limit

Airtable’s published Web API rate limit is five requests per second per base across all plans. When the limit is exceeded, the API can return HTTP 429, and Airtable’s documentation says requests should wait 30 seconds before subsequent requests succeed.

The official JavaScript client includes retry and backoff behavior. Other implementations should handle throttling deliberately rather than repeatedly sending requests after an error.

Why this matters for client-facing applications

Imagine a page that needs to retrieve several sets of data from Airtable before it can render. If many users open that page around the same time, their requests can compete for the same per-base capacity.

That does not automatically mean every request will be slow or that Airtable cannot support a multi-user application. The outcome depends on request patterns, caching, batch design, concurrency, and how the application handles throttling.

But the limit is a real architectural constraint. A higher subscription tier does not turn Airtable into an unrestricted high-throughput database.

Practical ways to reduce request pressure

Batch supported writes. Airtable supports up to 10 records per request for applicable batch operations. Use batching when importing or updating multiple records.

Cache repeated reads. If a dashboard repeatedly requests the same data and that data does not need to be live on every page load, caching can reduce API traffic.

Avoid unnecessary calls. Request only the fields and records required for the task. Do not fetch a full table when a filtered query will do.

Use queues and backoff. Queue background work and retry throttled requests with an appropriate delay. Avoid an immediate retry loop that creates more traffic.

Separate interactive and background work. A scheduled report should not compete unnecessarily with requests needed to load a customer-facing page.

These are standard application-design approaches, not guarantees that any particular Airtable workload will meet a target response time.

4. The Subscription Bill Depends on Collaborators, Too

API capacity is only part of the cost. Airtable’s self-serve plans are billed per workspace, and collaborator permissions influence who counts as billable.

Airtable’s billing documentation distinguishes the main rules for Team and Business:

  • On Team, workspace or base collaborators with Commenter or higher-level permissions can be billable.
  • On Business, workspace or base collaborators with Editor or higher-level permissions can be billable.

Sales-led arrangements and legacy plans may follow different terms. Verify the billing rules for the workspace you intend to use.

This matters for agencies that build client portals or internal tools on Airtable. The application’s end users, Airtable collaborators, and people who need direct access to the underlying base are not necessarily the same group.

If clients only need to use an interface or view published information, assess whether that access can be provided without granting unnecessary permissions to the underlying workspace or base. Do not assume that every external user requires a paid seat, but do not assume that every collaborator is free either.

A simple subscription calculation

For a self-serve Team workspace at the published annual-billing price, the base subscription estimate is:

Annual base subscription = billable users × monthly price per user × 12

For example, five billable users at $20 per user per month would produce an annualized subscription cost of:

5 × $20 × 12 = $1,200 per year.

This is a simple illustration of the listed rate, not a quote. Confirm which collaborators are billable, the applicable plan, and the actual checkout or contract terms before using the figure in a budget.

5. Record Limits Can Become the Next Constraint

An application can have sufficient API capacity but still run into record limits.

Airtable’s published record limits are cumulative across tables within a base. Two tables with 25,000 records each count as 50,000 records toward the base’s total, rather than each receiving a separate allowance of 50,000.

The published limits are 1,000 records per base on Free, 50,000 on Team, 125,000 on Business, and 500,000 or more on Enterprise Scale.

Source: Airtable workspace settings and usage limits.

A base used to track a few hundred clients and projects may fit comfortably. A system storing every event, activity log, historical snapshot, and transaction can grow much faster.

Before using Airtable as the primary data store, identify which records need to remain operationally active, which can be archived, and which belong in a different storage system.

Do not assume that reaching a record limit deletes existing data. Airtable says existing records remain in place, but users cannot add more records or attachments when the relevant limits are reached until they move to a suitable plan or otherwise resolve the constraint.

6. Automations Are a Separate Limit From API Calls

Airtable automations have their own run limits. They should not be treated as interchangeable with API calls.

Airtable’s published automation limits are:

PlanAutomation runs per monthUnique automations per base
Free10075
Team25,000100
Business100,000150
Enterprise Scale500,000200

An automation run is counted when the trigger is invoked, including failed and successful attempts. The unique-automation limit applies per base and includes automations that are turned off; disabling an automation does not free a slot, but deleting it does.

A workflow that relies on both the Web API and native automations therefore needs two separate usage estimates. Staying under the API-call limit does not prove that the automation allowance is sufficient, and vice versa.

7. When Airtable Is a Sensible Backend

Airtable can be a practical choice when the application benefits from a flexible data model, non-technical users need to maintain records, and expected request volume fits its limits.

It is worth evaluating for:

  • Internal tools with moderate traffic.
  • Agency operations dashboards.
  • Prototypes that need structured data and a usable interface.
  • Workflows where business users need to inspect and edit records directly.
  • Applications that can cache reads and process background updates asynchronously.

The decision deserves more scrutiny when the system requires sustained high request throughput, large volumes of historical data, strict latency targets, or a database designed for many concurrent application users.

Airtable’s API rate limit is the key consideration in these cases. Moving to Business or Enterprise Scale can remove the monthly API-call cap, but the standard per-base rate limit remains.

A different database may be a better fit if the application’s workload requires a higher request rate or more control over indexing, query execution, caching, and data access. That does not mean Airtable is inherently unsuitable as a backend. It means the architecture should match the workload instead of assuming a spreadsheet-style platform can scale indefinitely without design changes.

8. A Practical Decision Checklist

Before committing to Airtable as an application backend, answer these questions:

  1. How many API calls will the application make? Estimate interactive traffic, scheduled jobs, metadata requests, and retries.
  2. What is the peak request pattern? The five-requests-per-second-per-base limit applies even when monthly usage is below the allowance.
  3. How many records will the base hold? Include linked tables, logs, and historical data in the estimate.
  4. How much attachment storage is required? Check the plan’s allowance and expected growth.
  5. Who needs direct collaborator access? Map roles and permissions before estimating billable users.
  6. How much automation activity will run each month? Calculate native automation runs separately from API calls.
  7. Can the application cache or batch requests? Reducing unnecessary API traffic may improve reliability without requiring an upgrade.
  8. What happens if Airtable is unavailable or throttles requests? Define retry, queue, and user-facing error behavior.

If you cannot estimate these factors yet, start with a small pilot and measure the actual request pattern before using Airtable as the production backend for a larger user base.

Frequently Asked Questions

Does Airtable charge for every API request?

Airtable’s standard Web API pricing is based on the workspace plan rather than a simple fee for each request. Free and Team have monthly API-call allowances; Business and Enterprise Scale have no monthly API-call cap under the published plan limits. Rate limits and other plan restrictions still apply.

What is Airtable’s API rate limit?

The published Web API rate limit is five requests per second per base on all plans. There is also a separate limit of 50 requests per second for traffic using personal access tokens from a given user or service account.

Does upgrading Airtable remove the API rate limit?

No. Business and Enterprise Scale remove the monthly API-call cap under the published limits, but the five-requests-per-second-per-base limit still applies.

How many records can an Airtable base contain?

The published limits are 1,000 records on Free, 50,000 on Team, 125,000 on Business, and 500,000 or more on Enterprise Scale. Limits are cumulative across tables in a base.

Do Airtable API calls and automation runs share one limit?

No. API calls and native automation runs have separate allowances. Estimate each independently if your application uses both.

Is Airtable suitable for a customer-facing application?

It can be suitable for applications with moderate traffic and a workload that fits the platform’s constraints. For higher request rates or stricter performance requirements, evaluate caching, batching, a separate backend database, and the behavior of the application under concurrent traffic before committing.

The Stack Architects Verdict

Airtable’s API economics are best understood as a combination of subscription pricing, collaborator billing, monthly API allowances, record capacity, and rate limits.

Free can work for prototypes with low usage, but its 1,000 monthly API calls and 1,000-record base limit are restrictive for many production applications. Team raises the monthly API allowance to 100,000, while Business and Enterprise Scale remove the monthly API-call cap. None of those tiers removes the published five-requests-per-second-per-base limit.

For a lightweight internal tool, Airtable may be a sensible way to get a structured application running without managing a separate database. For a client-facing product with sustained traffic, estimate the workload, test the actual access pattern, and decide whether Airtable should be the primary data store or the operational interface connected to another database.

Before choosing a plan, verify the current Airtable pricing, API-call limits, and record and storage limits. The correct plan is the one that supports your real workload, not simply the one with the lowest advertised price.