Airtable Limitations: When Your Spreadsheet-Database Hits the Wall
Airtable is a brilliant tool for organizing information. It's a terrible choice for a production database. Here's why.
Airtable occupies a unique space: more powerful than a spreadsheet, more accessible than a database. For organizing projects, tracking content, and managing small datasets, it's genuinely excellent.
The problems start when teams treat Airtable as something it was never designed to be — a production database powering real applications. This happens gradually, and by the time the limitations become painful, there's significant data and workflow investment to unwind.
The 100,000 Row Ceiling
Airtable's hard limit of 100,000 records per base is the most obvious constraint, but the practical limit is much lower. Performance starts degrading noticeably around 10,000-20,000 records, especially with linked records, rollups, and lookup fields.
For context, a modestly successful e-commerce store generates 100,000 orders in 1-2 years. A CRM with 100 salespeople hits the limit in months. This isn't a scale problem for enterprises — it's a limit that growing startups hit routinely.
100K Hard Limit
Airtable caps records per base at 100,000. No workaround, no enterprise exception.
Performance Drops at 10K
Views, filters, and linked records become noticeably slow well before the hard limit.
Formula Limits
Complex rollup and lookup chains create cascading performance issues.
API Rate Limiting Kills Automation
Airtable's API allows 5 requests per second per base. That sounds reasonable until you're building real integrations. A webhook that needs to update 500 records takes almost 2 minutes at 5 req/sec (batching in groups of 10).
Zapier, Make, and other automation tools frequently hit these limits during peak usage, causing failed automations and data inconsistencies. The rate limit doesn't scale with your plan — even Enterprise customers get the same 5 req/sec.
5 Requests Per Second
Hard limit regardless of plan. Enterprise pricing doesn't buy you more API capacity.
Batch Limits
Maximum 10 records per API call means bulk operations are painfully slow.
No Webhooks (Natively)
Airtable lacks real-time webhooks. You have to poll for changes or use third-party bridges.
No Transactions, No Data Integrity
Real databases have transactions — the ability to update multiple records atomically (all succeed or all fail). Airtable has no transaction support. If you need to update a customer record and their associated orders simultaneously, there's no guarantee both operations complete.
This leads to data inconsistencies that are difficult to detect and painful to fix. In a real database, this problem literally cannot happen. In Airtable, it happens whenever your automation hits a rate limit mid-operation or the API returns a transient error.
The Relational Data Illusion
Airtable has "linked records" that look like foreign keys. They're not. There are no referential integrity constraints, no cascading deletes, no real joins. You can link a record to a non-existent record. You can delete a record that other records depend on.
Rollup and lookup fields approximate SQL JOINs but with severe limitations: they can't aggregate across multiple levels, they have no WHERE clause equivalent, and they recalculate on access (not on write), which creates stale data in views.
Security and Compliance Gaps
Airtable stores your data on their infrastructure with limited control over encryption, access logging, and data residency. For companies in regulated industries or those handling sensitive customer data, this creates real compliance challenges.
Row-level security doesn't exist. If someone has access to a base, they can see all records. Audit logging is limited. Field-level encryption is not available. For anything beyond basic business data, these gaps are problematic.
No Row-Level Security
Can't restrict individual records by user or role. It's all-or-nothing per base.
No Data Residency
Can't choose where your data is stored. A problem for US companies with state privacy laws like CCPA and Washington's My Health My Data Act.
Limited Audit Logs
Basic revision history per record, but no comprehensive audit trail for compliance.
Pricing That Punishes Growth
Airtable's pricing jumps dramatically between tiers. Free gives you 1,000 records. Plus ($20/user/mo) gives you 50,000. Pro ($45/user/mo) gives you 100,000. For a 20-person team on Pro, you're paying $10,800/year — for a tool that still can't do transactions or handle more than 100K records.
Meanwhile, a PostgreSQL database on a $50/month server can handle millions of records with proper indexing, full ACID transactions, and no per-user licensing.
When Airtable Is Still the Right Choice
Airtable is excellent for what it is: a collaborative, visual database for small-to-medium datasets. Project management, content calendars, product roadmaps, CRM for teams under 10 people — these use cases rarely hit Airtable's limits.
The problem is using Airtable as the backend for a production application. That's the use case where every limitation above compounds into a serious problem.
Frequently Asked Questions
Can Airtable handle more than 100,000 records?
Is Airtable a real database?
Why does Airtable get slow with linked records?
What should I use instead of Airtable for my app?
Hitting Airtable's Limits?
Talk to our team about building a real backend that grows with your business — no row limits, no API throttling, no per-user fees.