Trello

Trello is a Kanban-style project management tool built around boards, lists, and cards. You create a board for a project, add lists to represent stages or categories, and move cards between lists as work progresses. Each card can hold descriptions, checklists, attachments, labels, due dates, and comments.

The core appeal is simplicity. Non-technical users can pick it up in minutes, and the visual board metaphor maps naturally to many workflows – from sprint planning to content calendars to personal to-do lists. Power-Ups (Trello’s plugin system) extend functionality with calendar views, custom fields, integrations, and reporting.

Butler, Trello’s built-in automation engine, handles rule-based and scheduled automations without code. Free accounts get 250 automation runs per month; Premium and Enterprise get unlimited.

Trello’s sweet spot is small teams and individuals managing straightforward workflows. It intentionally trades depth for accessibility – you won’t find Gantt charts, native time tracking, or advanced reporting here. For complex project management, Atlassian wants you using Jira.

Trello offers four tiers, all priced per user per month:

- Free: $0 – Up to 10 boards per workspace, 10 collaborators per workspace, 250 automation runs/month, unlimited cards, all Power-Ups enabled, 10MB attachment limit. - Standard: ~$5/user/month (annual) or ~$6/month (monthly) – Unlimited boards, 1,000 automation runs/month, 250MB attachment limit, advanced checklists, custom fields. - Premium: ~$10/user/month (annual) or ~$12.50/month (monthly) – Unlimited automations, dashboard and calendar views, CSV export, workspace-level reporting. - Enterprise: ~$17.50/user/month (annual only, 50-user minimum) – SSO, audit logs, organisation-wide permissions, attachment restrictions.

API access is included on all plans at no additional cost. Rate limits are the same across tiers. The free tier’s 10-collaborator limit per workspace is enforced strictly – workspaces exceeding this become view-only, which caught many existing users off guard when it was introduced in April 2024.

Lock-in risk is moderate-to-high, and it’s an unusual case because the risk isn’t that you can’t get your data out – it’s that the platform may not be worth staying on. Board data exports to JSON are available on all plans, and Premium adds CSV and bulk workspace exports. The raw data is accessible.

What doesn’t transfer: Butler automations, Power-Up configurations, custom field setups, and the implicit workflow context of how your boards and lists are structured. Any migration will involve rebuilding these from scratch. The lack of a native import mechanism means you can’t even round-trip your own exports back into Trello without writing custom API code.

The strategic concern compounds the lock-in question. If you build a deep integration with Trello and Atlassian continues deprioritising the product, you may find yourself maintaining an integration against a platform that’s receiving minimal updates. Consider whether the simplicity of the integration today is worth the potential migration cost tomorrow.

Trello was created at Fog Creek Software in 2011 by Joel Spolsky and Michael Pryor, launching publicly at TechCrunch Disrupt. It spun off as its own company in 2014 with $10.3 million in Series A funding from Spark Capital and Index Ventures. Atlassian acquired Trello in January 2017 for approximately $425 million.

Atlassian (NASDAQ: TEAM) is a publicly traded company with roughly 13,800 employees and a market capitalisation in the $19–22 billion range. Q2 FY2026 revenue hit $1.59 billion (up 23% year-over-year), and they achieved their first $1 billion cloud revenue quarter. The parent company is financially healthy.

Trello’s position within Atlassian is the concern. Multiple rounds of layoffs have reduced the engineering team to a skeleton maintenance crew, and Atlassian has explicitly stated they’re repositioning Trello as a personal productivity tool while steering teams toward Jira Work Management. A controversial May 2025 redesign drew widespread user backlash – The Register called it “possibly the worst redesign in tech history.” The product has over 90 million registered users and roughly 72–77 million monthly web visits, but community sentiment is at a low point.

REST API at https://api.trello.com/1/ – it’s been version 1 since launch and has never had a major version bump. The URL structure is clean and predictable: /1/{resource}/{id}/{nestedResource}. Nested resources are well-supported, so you can fetch all cards on a board in a single call rather than making individual requests.

Authentication is currently API key plus token. You create a Power-Up at trello.com/power-ups/admin to get your API key, then generate a user token via an authorisation URL. Tokens can be set to expire after 1 hour, 1 day, 30 days, or never. OAuth 1.0 is also available, and Atlassian is in the process of migrating to OAuth 2.0 (3LO) – timeline is in progress, so keep an eye on the changelog.

Rate limits are generous for most use cases: 300 requests per 10 seconds per API key, 100 requests per 10 seconds per token. Response headers tell you exactly where you stand with remaining request counts. The /1/members/ endpoint has a stricter limit of 100 requests per 15 minutes. If you generate more than 200 HTTP 429 errors in a 10-second window, all requests from that key get blocked for the remainder of the window.

The API covers 18 resource groups including boards, cards, lists, checklists, custom fields, labels, members, organisations, and search. There’s a batch endpoint that lets you combine up to 10 GET requests into one call, but it only supports GET – no batch writes.

The biggest pain point is pagination. Cards are ordered by their visual position on the board, not by creation time or ID. This means paginating through cards is unintuitive and moving a card changes its position in API results. The community workaround is sorting by -id and paginating backwards using the ‘before’ parameter. It’s workable but annoying.

Solid webhook implementation with 93+ event types. Payloads are signed with HMAC-SHA1 for verification. When creating a webhook, Trello sends a HEAD request to your callback URL – it must return HTTP 200 or creation fails. Failed deliveries retry 3 times with exponential backoff (30s, 60s, 120s). A webhook is automatically disabled after 30 consecutive days AND 1,000+ consecutive failures. Returning HTTP 410 Gone from your callback automatically deletes the webhook, which is a nice touch. All webhook traffic comes from 104.192.142.240/28 on port 443.

Data export is a mixed bag. Free and Standard plans can export individual boards to JSON via the UI or API. Premium and Enterprise plans unlock CSV export and workspace-wide bulk exports including attachments in a ZIP file. JSON exports include the 1,000 most recent actions (including comments); CSV exports exclude comments entirely.

The critical gap: there is no native import capability. You cannot upload a JSON or CSV file to recreate a board. If you need to restore from an export or migrate data in, you’ll need to use the API to programmatically create each board, list, and card individually – and since the batch endpoint only supports GET, every create operation is a separate API call.

Migration tooling exists through third parties (Relokia, Altosio) and native importers in competing tools (Asana has one, Jira has one since they’re both Atlassian). The Trello-to-Jira path is the best-supported migration route, unsurprisingly.

Vendor lock-in risk is moderate-to-high. Your raw data is exportable, but Power-Up configurations, Butler automations, and custom field setups don’t transfer. You’ll lose workflow context in any migration.

Documentation lives at developer.atlassian.com/cloud/trello/ and is decent but not great. The API reference is reasonably complete with request/response examples, rate limit documentation is clear, and webhook docs are thorough. Conceptual guides cover authorisation, nested resources, and Power-Up development.

The weak spots: pagination behaviour is poorly documented (developers reverse-engineer it from forum posts), some content was lost or became harder to find after migrating from the old developers.trello.com domain, and edge cases with nested resources aren’t well-covered.

There are no official SDKs. All client libraries are community-maintained. For JavaScript/TypeScript, trello.js on npm is the most comprehensive with claimed ~100% API coverage. For Python, py-trello on PyPI is the most popular. Trello provides a first-party browser client library at api.trello.com/1/client.js, but it requires jQuery.

There is no sandbox or testing environment. You test against your real Trello account using free boards. The recommendation is to create a dedicated development account and test boards, but there’s no isolated API mode.

The Power-Up development experience is iframe-based – your Power-Up is a web application you host yourself that communicates with Trello via a client library. It’s functional but requires you to manage your own hosting infrastructure.