whats-best.ai

Project Management & Collaboration

Airtable

Rest of world Report an error

Panel rating · 6 judges · How to read the stars

Category median

Sovereignty: 1 of 4 dimensions proven

0–5 in half steps. 5 means the rubric's top anchor is met on the evidence.

by Formagrid, Inc. · www.airtable.com

Report an error on this page Is this your product? →

Read this page as one judge. Each weighs the same scores by what they care about.

The panel's verdict

Airtable (Formagrid Inc, San Francisco) is strongest on pricing transparency and onboarding effort: Team and Business publish both billing periods ($20/$24 and $45/$54 per seat) with limits itemized, and the free plan allows up to 5 editors with 100 automation runs and 500 AI credits, though no templates, importers or guided setup are evidenced. It is weakest on support & documentation and reporting: no help center, docs or support channel appears beyond security contacts and paid professional services for 'qualified' Business/Enterprise Scale customers, and reporting is limited to record revision history and audit logs, with no dashboards, cross-project views or export paths evidenced. Collaboration depth scores 3–4: real permissions, but no evidenced assignments, comments, timelines or workload views. The bench genuinely splits on sovereignty (3–4, three judges at each score): lower scorers read a US vendor with services in Airtable's AWS environment and EU storage only 'the option'; higher scorers credited the self-service DPA and published subprocessor list as substance. Integrations spans 4–6 on whether the tiered API quota outweighs an absent native catalogue.

Report an error

Speaks for it

  • Self-serve pricing publishes both billing periods for Team ($20/$24 per seat) and Business ($45/$54 per seat) with record, attachment and AI-credit limits itemized.
  • Free plan offers up to 5 editors, 100 automation runs and 500 AI credits per editor each month.
  • Public API with published per-tier call quotas — 1,000/month Free, 100,000 Team, unlimited Business and Enterprise Scale — plus automations and Business-tier two-way sync.
  • Permission machinery is real: customizable collaborator permissions, fine-grained RBAC, and password-protected or domain-restricted share links.
  • Compliance artifacts are substantial on paper: self-service DPA, published subprocessor list, ISO/IEC 27001:2022, ISO/IEC 27701:2019 and SOC 2 Type 2.

Report an error

Held against it

  • No help center, documentation, changelog, SLA or support channel is evidenced beyond security contacts and professional services restricted to qualified Business/Enterprise Scale customers.
  • No dashboards, cross-project, time/budget/utilization reporting or report export is evidenced — only record-level revision history and audit logs.
  • No assignments, comments, dependencies, timelines or workload views appear anywhere in the evidence, leaving team planning unevidenced.
  • EU data residency is described only as 'the option', services run in Airtable's AWS environment, and deleted content may be retained in archived/backup copies with permanent deletion via privacy@airtable.com.
  • Guest access is a paid Portals add-on (from $120/month for 15 guests on Team), SAML SSO is Enterprise-Plan-only, and Enterprise Scale pricing is custom.

Report an error

Best for

  • You need a shared database with strong permissions more than a team-planning surface.
  • You can build your own integrations and reporting on the API, which publishes unlimited calls on Business and Enterprise Scale.
  • You want to compute seat costs yourself from published prices and limits before committing.
  • Your team can self-serve, since no support channel or guided onboarding is evidenced.

Report an error

Avoid if

  • You need out-of-the-box dashboards, time/budget/utilization reporting or exportable reports — reporting scores 1–2.
  • You depend on documented help, support channels or SLAs for a rollout — support & documentation sits at 2.
  • You need EU data residency as a contractual default rather than an option.
  • You expect evidenced assignments, comments, timelines or dependencies for team planning.

Report an error

The scores

Collaboration depth

Show reasoning
How this is scored

How much real multi-person project work the tool carries: shared planning, assignments, dependencies, workload, comments, guest access.

0 — A personal task list; a second user has no defined place in it.

3 — Tasks can be shared and assigned, but planning stays per-person — no shared timeline, workload or dependency view.

5 — Solid shared projects with assignments, comments and at least one team-level planning view; gaps appear in cross-project or capacity planning.

8 — Team-level planning is first-class: cross-project views, workload/capacity, dependencies, guest or client access with sane permissions.

10 — Handles the full span from two freelancers to a structured department in one model — planning, capacity, permissions and client involvement all without workarounds.

Report an error

The Data Protection Officer

Granular collaborator permissions, password-protected/domain-restricted share links, RBAC and paid guest Portals prove real multi-user work in a shared workspace — but the evidence evidences no assignments, comments, dependencies, timelines, workload or any team-level planning view. Airtable as documented here is a shared database with strong permissions, not a project planner; whatever planning exists, each team builds itself. 1 2 4

Report an error

The Bootstrapper

The evidence proves shared workspaces with customizable permissions, fine-grained RBAC, and guest access via Portals — but Portals is a paid add-on ($120-150/mo for 15 guests), and nowhere is there evidence of assignments, comments, shared timelines, dependencies, or workload views. For a 3-seat team doing actual project planning together, the planning layer is entirely unproven on this evidence. 1 2 4

Report an error

The Enterprise Architect

Multi-person work is evidenced as shared bases with real permission hygiene — fine-grained RBAC, collaborator permissions, domain-restricted share links, and guest access via Portals — but nothing in the evidence evidences assignments, comments, dependencies, timelines or any team-level planning view. That's shared data, not shared planning; I'm interpolating between rubric level 3 and 5. 1 2 4

Report an error

The UX Purist

Sharing is real — customizable collaborator permissions, password/domain-restricted share links, fine-grained RBAC, paid guest access via Portals — but not one fact evidences assignments, comments, dependencies or a workload view. It's a shared database, not a shared plan. 2 4 1

Report an error

The Integrator

Permissions are the strong suit — collaborator-level controls, fine-grained RBAC, IDP-synced groups, admin panels [E1/E5] — and guest access exists via paid Portals add-ons and password/domain-restricted share links [E2/E5]. But the evidence is silent on assignments, comments, dependencies, timelines and workload/capacity — none of the shared planning machinery the upper anchors require — so sharing works but multi-person planning is unevidenced. 1 2 4

Report an error

The Skeptic

Real multi-user machinery is evidenced — customizable collaborator permissions, password-protected share links, record revision history, up to 5 free editors — but not one team-planning surface (timeline, dependencies, workload, comments, assignments) appears anywhere in the evidence. Client/guest access exists only as a paid Portals add-on from $120/month, so even that is monetized. That lands between rubric level 3 and 5: shared work, zero evidenced planning. 1 2 4

Report an error

Reporting

Show reasoning
How this is scored

Whether the tool can answer questions about the work: progress, time, budgets, utilization — and export the answer.

0 — No reporting beyond looking at the board.

3 — Fixed per-project status views only; nothing aggregates across projects and nothing exports cleanly.

5 — Useful standard reports (progress, tracked time) with basic filters and CSV export, but limited customization and no budget/utilization depth.

8 — Configurable cross-project reporting incl. time, budgets and utilization, with dashboards and reliable exports.

10 — Reporting is a product in itself: custom report builder, scheduled delivery, API access to every metric the UI shows.

Report an error

The Data Protection Officer

Nothing in this sheet can answer a question about the work: no dashboards, progress, time, budget or utilization reporting, and no report export of any kind is evidenced. What exists is a per-record revision activity feed, audit logs and e-discovery as admin controls, and a plan-capped API — you could build your own reporting, but the product itself provides none that I can verify. 1 2 4

Report an error

The Bootstrapper

The only reporting-adjacent thing evidenced is record-level revision history as a per-record activity feed, tiered by plan (2-week to 3+ year). Nothing aggregates across projects, nothing covers time, budgets, or utilization, and no export path appears anywhere in the evidence — silence I have to score as absence. 2 4

Report an error

The Enterprise Architect

The only answer-this-question artifacts are change trails — audit logs and record-level revision history — which serve compliance review, not work insight; there is zero evidence of progress, time, budget or utilization reporting, dashboards, or any export path. rubric level 0 with a single point of credit for the audit trail I can hand to internal audit. 1 4

Report an error

The UX Purist

The only visibility evidenced is record-level revision history with an activity feed and admin audit logs/e-discovery; no dashboards, no cross-project aggregation, no time/budget/utilization, no clean export of an answer. The metered API is a raw pipe, not a report. 4 1 2

Report an error

The Integrator

Record-level revision history and audit logs/e-discovery give change traceability, but there is no evidence of any progress, time, budget or utilization report, no dashboard, and no aggregated export — the only egress path evidenced is raw API quota. 1 2 4

Report an error

The Skeptic

The only reporting-adjacent artifacts in the registry are per-record revision history and revision timeframes tiered by plan, plus enterprise audit logs/e-discovery. No dashboards, no cross-project views, no time/budget/utilization reporting, no export capability is evidenced at all. Missing evidence is information here — this is below even the 'fixed status views' anchor. 1 2 4

Report an error

Integrations

Show reasoning
How this is scored

How well the tool connects to the rest of the stack: native integrations, calendar/comms hooks, API and webhooks, import/export paths.

0 — A closed box — no API, no integrations, CSV out if anything.

3 — A handful of marquee integrations, no public API or an API that covers a fraction of the product.

5 — The common set (calendar, Slack/Teams, file storage) plus a documented public API; automation platforms partially covered.

8 — Broad native catalogue, well-documented API and webhooks, first-class automation-platform support and real import paths from competitors.

10 — Ecosystem-grade: everything in 8 plus the tool is itself extensible (apps/plugins, AI/MCP surfaces) and exit via API export is complete.

Report an error

The Data Protection Officer

There is a real API with published call quotas per plan plus an Enterprise API, native automations and two-way sync, and SAML SSO with IdP-synced groups. But the evidence is silent on the entire catalogue I'd expect — no calendar, Slack/Teams or file-storage integration, no webhooks, no automation-platform coverage, no import path from competitors — so I can only credit the connectivity that is actually documented. 2 4

Report an error

The Bootstrapper

A real public API is evidenced by per-tier call limits (1,000/mo free, 100,000 Team, unlimited Business+), plus two-way sync, IdP-synced groups, and an App Library for extensibility. But the 1,000-call free cap is stingy, and the evidence evidences none of the calendar/Slack/file-storage set, webhooks, or import paths from competitors, so I can't go higher than mid-scale. 1 2 4

Report an error

The Enterprise Architect

A real API is evidenced by tiered call quotas (1,000 free to unlimited Business+), an Enterprise API, automations with metered runs, two-way sync and an App Library, plus IDP-synced groups which matters for my 200-seat identity story. But no native Slack/Teams, calendar or file-storage connectors are evidenced, no webhooks, and no import path from competitors — the common-set half of rubric level 5 is missing. 2 1

Report an error

The UX Purist

A public API with real quotas (100k calls on Team, unlimited on Business+), an Enterprise API and two-way sync, plus SAML SSO and IDP-synced groups. But the evidence is silent on calendar, Slack/Teams, webhooks, automation platforms and import paths — absence is information. 2 4 1

Report an error

The Integrator

The tiered API quota table — 1,000 calls/month on Free up to unlimited on Business+ plus a dedicated Enterprise API — is genuine evidence of a public API with generous egress, and there's an automation engine, Business two-way sync, an Enterprise App Library, SAML SSO and programmatic provisioning [E2/E1/E5]. But no webhooks, no named native integrations (calendar, Slack/Teams, storage) and no import path from competitors appear anywhere in the evidence, which caps it well below ecosystem-grade. 1 2 4

Report an error

The Skeptic

A public API with tiered quotas, automations, Business-tier two-way sync, and SAML SSO/IdP groups are confirmed — a genuine API surface. But not a single named native integration (Slack, calendar, file storage) appears in the evidence, and the free tier's 1,000 API calls/month is a metering wall that makes connectivity effectively paid. Interpolated between 'handful of integrations' and the common-set anchor. 1 2 4

Report an error

Onboarding effort

Show reasoning
How this is scored

Time and friction from signup to first real value for a small team, without paid services.

0 — No self-serve path — onboarding starts with a sales call and an implementation project.

3 — Self-serve exists but the first project requires configuration a newcomer cannot judge yet (workflows, custom fields, permissions) before anything works.

5 — A team gets a first project running within a day; trial requires payment details or key steps live behind docs.

8 — Signup to working project inside an hour: guided setup, sensible defaults, templates, trial without payment details.

10 — Minutes to value even for non-tool-people, importers pull existing work in, and the tool teaches its own advanced features as the team grows into them.

Report an error

The Data Protection Officer

A genuine self-serve free tier — up to 5 editors, 100 automation runs, 500 AI credits, no sales call — means a small team can start without talking to anyone. But no templates, importers or guided setup appear anywhere in the evidence, so time-to-value rides entirely on building a base from scratch. 2

Report an error

The Bootstrapper

A genuine self-serve free tier (5 editors, 100 automation runs, 500 AI credits) with no payment details implied — a buyer can start tomorrow. But the evidence says nothing about guided setup, templates, or importers, so I can't credit the 'minutes to value with importers' part of the scale; first real value likely needs base-building a newcomer can't judge yet. 2

Report an error

The Enterprise Architect

A genuine self-serve free tier with 5 editors, 100 automation runs and AI credits means a team can start without a sales call or payment details. The evidence is silent on guided setup, templates and importers, so I can't credit rubric level 8 territory — a working day-one is plausible, minutes-to-value is unproven. 2

Report an error

The UX Purist

A genuine free tier — 5 editors, 500 AI credits, 100 automation runs, no payment details — means the stopwatch can start at signup. Docked because nothing evidences guided setup, templates or importers; first value depends on a newcomer inventing structure with no evidenced help. 2

Report an error

The Integrator

A genuinely free tier with five editors, 500 AI credits and 100 automation runs means self-serve without payment details, and paid professional services are gated to qualified Business+ customers, implying self-serve is the default path. The evidence says nothing about templates, guided setup or importers to pull existing work in, so I can't credit the hour-to-value or importer anchors. 2

Report an error

The Skeptic

A genuine self-serve path: free plan with 5 editors, AI credits, 100 automation runs and no payment details. But nothing evidences guided setup or templates at entry tiers — the only template-like artifact, 'App Library', is gated to custom-priced Enterprise Scale, and professional services are sold only to 'qualified' Business/Enterprise buyers. No evidence a newcomer reaches a working project fast; a blank database canvas is exactly the configure-before-value trap. 2

Report an error

European sovereignty panel opinion

Show reasoning
How this is scored

Where the company and the data actually sit: legal entity, ownership control, data residency, and subprocessor exposure to non-EU jurisdictions.

0 — Non-European vendor, data on US infrastructure under US law, no DPA worth the name.

3 — DPA and GDPR language present, but hosting defaults to the US or ownership sits outside Europe — compliance paperwork over substance.

5 — EU data residency is available and contractually fixed, but the vendor or its controlling owner is non-European, or core subprocessors put data within CLOUD Act reach.

8 — European company under European control with EU-resident data and a public DPA; remaining exposure is limited to identifiable subprocessors (e.g. a US hyperscaler's EU region) and is documented.

10 — European company, European control, EU-only data residency on European infrastructure, subprocessor list free of extraterritorial reach — verifiable, not asserted.

Report an error

The Data Protection Officer

Formagrid Inc sits in San Francisco; EU residency is an opt-in, not a default, and everything including the AI models runs in Airtable's AWS environment — a US hyperscaler with CLOUD Act reach even in its EU region. The self-service DPA and published subprocessor list are substance, not just paperwork, but subprocessor exposure is undocumented here and deletion is the weak point: deleted content persists in archived and backup copies, and permanent deletion means emailing privacy@airtable.com. 1 3 4

Report an error

The Bootstrapper

Formagrid Inc. sits in San Francisco, EU residency is only 'the option to have their data stored in the European Union' — US hosting unless you opt in — and the AI models run in Airtable's AWS environment. There is a self-service DPA, GDPR language, EU/UK/Swiss representatives, and a public subprocessor list, but that's compliance paperwork over a US-default, US-owned stack. 1 3 4

Report an error

The Enterprise Architect

Formagrid Inc. is a San Francisco entity with Article 27 reps in Cork, London and Zurich, EU residency is an opt-in not a contractual default, and AI models run in Airtable's AWS environment — textbook CLOUD Act reach. The paperwork is genuinely strong (self-service DPA, SCCs, ISO 27001/27701, SOC 2 Type 2, public subprocessor list), which lifts it above rubric level 3's compliance-theater, but a US owner on a US hyperscaler with uncontracted EU residency caps it between rubric level 3 and 5. 3 4 1

Report an error

The UX Purist

Formagrid Inc sits in San Francisco, EU storage is only 'the option' with a US default implied, and AI models run in Airtable's AWS environment. The GDPR paperwork is unusually thorough (self-serve DPA, EU representative, CAIQ) but the substance is a US owner with US-default hosting — anchor-3 territory with an opt-in EU nudge. 3 4 1

Report an error

The Integrator

Formagrid Inc sits in San Francisco, services run in Airtable's AWS environment, and EU storage is merely 'the option to have their data stored in the European Union' — an opt-in, not a default or evidenced contractual fix. GDPR paperwork is thorough (self-service DPA, EU/UK/Swiss representatives, SCCs) [E5/E4], but that's compliance paperwork over US-hosted substance, exactly the anchor-3 pattern. 1 3 4

Report an error

The Skeptic

Formagrid Inc, San Francisco — a US vendor under US control, and EU storage is explicitly 'the option', not the default or a contractual commitment. SCCs as the transfer mechanism confirm data leaves the EEA, and the Cork/Zurich/London 'representatives' are a contracted DataRep mailbox, compliance paperwork over substance. Self-service DPA exists, which keeps this at rubric level 3 rather than lower. 3 4

Report an error

Support & documentation

Show reasoning
How this is scored

Whether a stuck user gets unstuck: docs quality, support channels and their availability per plan, and the health of the learning surface.

0 — No docs beyond marketing pages; support is a contact form that may answer.

3 — Docs exist but are thin or stale; support on entry plans is effectively email-only with slow turnaround.

5 — Searchable help center covering the core product; chat or email support on all plans with reasonable response times.

8 — Deep, current docs incl. onboarding guides and templates; fast chat support not gated to top tiers; visible changelog.

10 — Docs, academy, community and support so good they are a reason to buy; SLAs published, support quality independent of plan.

Report an error

The Data Protection Officer

The evidence evidences no help center, no documentation, no chat and no SLAs whatsoever — what is visible is security@airtable.com plus a HackerOne bounty, SOC 2 reports handed out by sales on request, and paid professional services restricted to 'qualified' Business and Enterprise customers. A stuck user on Free or Team has no evidenced path to getting unstuck. 2 4

Report an error

The Bootstrapper

No help center, docs, chat, email support, or response-time evidence appears anywhere in the registry — missing evidence is information. The only support artifacts evidenced are a security contact/HackerOne channel and professional services packages restricted to 'qualified customers' on Business and Enterprise Scale, i.e. paid and gated to top tiers. 2 4

Report an error

The Enterprise Architect

The only support surfaces evidenced are paid professional services for qualified Business/Enterprise buyers and security-issue channels (HackerOne, security@) — no help center, no docs, no support channels per plan, no SLAs, no changelog. For a 200-seat rollout I cannot evidence that a stuck user gets unstuck at all; rubric level 0 territory with a point for the security reporting path. 2 4

Report an error

The UX Purist

No help center, docs surface or chat is evidenced anywhere — just a privacy email, a security@/HackerOne channel, and professional services sold only to qualified Business and Enterprise Scale customers. A stuck free-tier user has nowhere evidenced to go. 4 2

Report an error

The Integrator

The only support surfaces evidenced are a HackerOne bug bounty and security@ email, a privacy contact, and paid professional services restricted to qualified Business+ buyers — no help center, no docs, no support channel per plan, no response times, no changelog. For a stuck user on a Free or Team plan, the evidence evidences effectively nothing. 2 3 4

Report an error

The Skeptic

The registry contains zero evidence of a help center, docs, changelog, or any support channel for paying customers — the only captured support surfaces are a HackerOne bounty and security@ email, a privacy mailbox, and professional services restricted to 'qualified customers' on Business/Enterprise. I hunt dead changelogs for a living; here I can't even find a live one to suspect. 2 3 4

Report an error

Pricing transparency

Show reasoning
How this is scored

Can a buyer determine the real, payable cost of the tool for their team before talking to anyone?

0 — No public pricing at all — every number sits behind "contact sales".

3 — Some tiers are public, but a deciding factor (seat minimums, mandatory add-ons, billing-period surcharge, tax treatment) only surfaces at checkout or in fine print.

5 — Self-serve tiers are publicly priced and the billing-period difference is stated, but the headline number still needs qualifying — e.g. annual-only headline, taxes excluded without saying so, or an opaque enterprise tier doing real work.

8 — Every self-serve tier is fully priced with billing periods and tax treatment plain on the page; only genuinely custom enterprise work lacks a number.

10 — Every tier including enterprise has a public price or published price formula, and the headline number is the number on the invoice.

Report an error

The Data Protection Officer

Every self-serve tier is priced with annual and monthly billing side by side, and the operative limits — records, attachments, AI credits, API calls — are published on the same page. Tax treatment is never stated, Enterprise Scale is 'Custom pricing', and the Portals add-on adds a second $120–150/month line for guests, so the headline number is close to but not quite the invoice. 2

Report an error

The Bootstrapper

The price page answers: Free/Team/Business are fully priced with both billing periods ($20/$24, $45/$54), no seat minimums, and record/AI-credit limits stated — I can compute a 3-seat and a 30-seat invoice myself. Docked points for Enterprise Scale being pure 'custom pricing' while holding real features hostage (SAML SSO is Enterprise-only), unstated tax treatment, and 'starting from' Portals add-on pricing. 2 4

Report an error

The Enterprise Architect

Self-serve tiers are fully priced with both billing periods stated side by side ($20/$24 Team, $45/$54 Business) and even add-ons are priced (Portals from $120–150), which is honest. But the Enterprise Scale tier is 'custom pricing' while doing decisive work — SAML SSO is Enterprise-only, alongside EKM, HyperDB and the Enterprise API — and tax treatment is never stated, so the real number for my deployment is undiscoverable short of a sales call. 2 4

Report an error

The UX Purist

Credit where due: both billing periods on the page ($20/$24, $45/$54) with record, attachment, automation and AI-credit limits spelled out, and even the Portals add-on priced. Tax treatment is never stated and real capability — SSO, HyperDB, Enterprise API — hides behind 'Custom pricing'. 2 4

Report an error

The Integrator

Every self-serve tier is fully priced with both billing periods on the page — Team $20/$24, Business $45/$54 per seat — plus itemized record/attachment limits, AI credits, Portals add-on pricing down to the annual-vs-monthly footnote, and Enterprise Scale honestly marked 'custom'. The only gap against a clean page is tax treatment, which the evidence never states. 2

Report an error

The Skeptic

Credit where due: self-serve tiers publish both billing periods ($20/$24 Team, $45/$54 Business) with record and attachment limits plain. But the headline number isn't the invoice: guest access is a separate Portals meter from $120/month, AI credits are metered with no overage price, tax treatment is unmentioned, and SAML SSO plus HyperDB sit behind an opaque custom-priced Enterprise tier doing real work. 2 4

Report an error

European sovereignty — proven facts

1 of 4 dimensions proven

Built only from facts shown on the vendor's own pages. A dimension we could not prove is left open, not scored as zero.

Ownership Not determined — uncited Report an error
Data residency Not determined ⚠ unverified — uncited Report an error
Subprocessors US CLOUD Act reach ⚠ unverified 0/2 pts 1 Report an error

Where this could be wrong

What we left out

A claim that does not survive our checks costs us the claim, not the page. This is what was taken off this one.

Sources (14)

The pages every claim on this page was read from — each one checked, dated, and kept verifiable.

  1. 1 Vendor homepage www.airtable.com Checked 5 Oct 2026 +2 earlier captures: 15 Sep 2026, 20 Aug 2026 Details →
  2. 2 Vendor pricing page www.airtable.com Checked 5 Oct 2026 Details →
  3. 3 Privacy policy www.airtable.com Checked 5 Oct 2026 Details →
  4. 4 Security page www.airtable.com Checked 5 Oct 2026 Details →
  5. 5 Collaboration depth — found from sitemap support.airtable.com Checked 5 Oct 2026 Details →
  6. 6 Collaboration depth — found from sitemap support.airtable.com Checked 5 Oct 2026 Details →
  7. 7 Reporting — found from sitemap www.airtable.com Checked 5 Oct 2026 Details →
  8. 8 Reporting — found from sitemap www.airtable.com Checked 5 Oct 2026 Details →
  9. 9 Integrations — found from sitemap www.airtable.com Checked 5 Oct 2026 Details →
  10. 10 Integrations — found from sitemap support.airtable.com Checked 5 Oct 2026 Details →
  11. 11 Onboarding effort — found from sitemap support.airtable.com Checked 5 Oct 2026 Details →
  12. 12 Onboarding effort — found from sitemap support.airtable.com Checked 5 Oct 2026 Details →
  13. 13 Support & documentation — found from sitemap www.airtable.com Checked 5 Oct 2026 Details →
  14. 14 Support & documentation — found from sitemap support.airtable.com Checked 5 Oct 2026 +1 earlier capture: 1 Oct 2026 Details →