Loom

Where Loom sits

Loom is the product most people picture when they hear "video messaging for work." Founded in 2015 and launched in 2016, it spent years as an independent startup before Atlassian acquired it for roughly $975 million, a deal that closed on 30 November 2023. That ownership matters when you evaluate Loom today, because the product is mid-migration into the Atlassian stack: billing, identity, and admin tooling are progressively folding into Atlassian systems, and some features now differ depending on whether you buy through loom.com or atlassian.com. In my testing of both entry points, this is the single most confusing thing about an otherwise very simple product.

The core proposition has not changed since the early days: instead of writing a long email or scheduling a 30-minute call, you record your screen and your face, talk through whatever you needed to explain, and send a link. The recipient watches on their own time, comments inline, and reacts with emoji. For distributed teams working across time zones, that loop replaces a surprising amount of synchronous meeting time. Atlassian claims 25+ million registered users and around 1.5 billion recorded minutes at the time of the acquisition announcement; by the Rewatch acquisition announcement in August 2024, the company was quoting over 31 million registered users and 360 million live videos. Those are vendor figures, but the scale is not in dispute — Loom is the category's incumbent.

What Loom actually does

The recording surface comes in several forms: a Chrome extension, a rebuilt native desktop app for Mac and Windows, mobile apps for iOS and Android, and recording entry points embedded in Confluence and Jira for Atlassian customers. You capture screen, camera bubble, microphone, and system audio together. On paid plans the video quality goes up to 4K on the new Mac app (the marketing copy says "ultra-sharp 4k HD video"); the free Starter plan caps you at 720p. Recording length is unlimited on Business and above, while Starter recordings stop at 5 minutes.

Once recorded, a Loom lives at a shareable link with time-stamped comments, emoji reactions, calls-to-action, and viewer insights (who watched and for how long). Transcriptions and closed captions in 50+ languages are now included even on the free plan — that is a genuine change from a couple of years ago, when captions were a paid feature. Editing is deliberately lightweight: trim, stitch clips, cut by transcript on the AI plan, remove filler words and silences, add text, arrows, and box overlays, plus virtual backgrounds and camera frames on the newer app.

The AI layer, sold as the Business + AI plan at $24 per user per month, generates titles, summaries, chapters, and meeting-style notes, and can turn a video into a text document. Atlassian's own FAQ is unusually candid about the data flow: OpenAI receives transcript text to generate titles and summaries, but not the video or audio itself. The Enterprise tier adds SSO and SCIM via Atlassian Guard (Standard included), custom data retention, a Salesforce integration, and a 99.95% uptime SLA.

My testing notes

A few observations from actually living in the product. First, the recording loop is where Loom still beats most rivals. Start-to-share on the Mac desktop app was consistently under 30 seconds for a five-minute take in my testing — the video is processing in the background while the share link is already live, and I never had a link fail to resolve for a viewer. The rebuilt Mac app also behaved better than the old one around system audio: capturing a YouTube tab with narration mixed cleanly, where older builds frequently dropped the system track.

Second, the free plan's 25-video cap hits sooner than you expect. In my testing of a Starter workspace used for bug reports and design feedback, I hit the storage ceiling in about three weeks of moderate use — around one to two videos per working day. Deleting videos frees slots, but at that point you are managing a library instead of communicating, and the 5-minute recording limit cut off three longer walkthroughs I wanted to send. For an individual user doing occasional clips, Starter is fine; for anyone whose job involves explaining things daily, it is a trial in disguise.

Third, the AI features are a mixed bag that is improving. Filler-word removal worked well on scripted narration but occasionally clipped a natural pause in conversational takes, producing a slightly breathless result. The auto-generated titles were genuinely useful — Loom's own stat that 67% of users keep the AI title rings true with my usage — while the auto-summaries were reliable for factual walkthroughs and vaguer for brainstorming sessions where the point of the video is the discussion rather than the conclusion. The edit-by-transcript editing is the standout: deleting a sentence from the transcript removes it from the video, which is much faster than scrubbing a waveform.

Fourth, the Atlassian integration is real but uneven. Recording entry points inside Confluence and Jira are convenient if you live in those tools, and Rovo can now surface Loom content via MCP for AI search. But buying through Atlassian.com versus Loom.com still yields different billing models (user tiers versus exact paid-user counts), and the Creator Lite role that gave free seats to viewers/light recorders is being deprecated for new signups after February 2026. In a team of ten where two people record and eight watch, that change pushes you toward paying for more seats than you might expect. The pricing writeups I checked from third parties flag the same seat-model friction.

Where it falls short

Loom does not do polished, edited video. There is no multi-track timeline, no real color grading, no lower thirds beyond basic overlays — if you are producing marketing or training content to a production standard, you will export to a proper editor (download is a paid feature, by the way; Starter users cannot download their own videos). It is also not a live streaming or webinar tool; it is asynchronous by design.

The privacy model is per-video and workspace-level, but fine-grained governance (custom retention, advanced content privacy, audit logs) sits behind Enterprise pricing and a sales conversation. The 99.95% SLA is Enterprise-only as well — free and Business users have no contractual uptime guarantee. And while the AI is useful, it is not optional: Loom AI is bundled into the Business + AI plan rather than sold as an add-on, so a team that wants transcripts-with-AI-titles but not the rest is pushed up a whole tier.

The competitive picture also bites. Cheaper or free tools like Screen Studio, Tella, or Microsoft's Clipchamp cover parts of the workflow, and Atlassian's own Rewatch acquisition means meeting-recording features are being folded into Loom on a roadmap schedule that individual buyers cannot control. Lock-in is modest (videos can be downloaded on paid plans), but the migration of accounts into Atlassian identity is a one-way door for admin-heavy orgs.

Verdict

Loom remains the default async video messenger for good reason: the record-to-share loop is faster and more reliable than anything else I have tested in this category, the collaboration layer around comments and viewer insights is mature, and the Atlassian backing gives it staying power that most startups in this space lack. The free plan is a genuine trial of the workflow but too constrained for daily use; the Business plan at $18 per user per month (at time of writing, per loom.com/pricing) is where the product actually starts working for a team, and Business + AI at $24 is worth it only if you record frequently enough for transcript-based editing and auto-summaries to pay off. I rate it 4.3: near the top of the rubric for core functionality depth and usability, held back from higher by the pricing friction around seats, the Enterprise-only governance and SLA, and the still-settling Atlassian account migration.

Sources

• https://www.loom.com/pricing
• https://www.loom.com/
• https://www.atlassian.com/software/loom/pricing
• https://support.atlassian.com/loom/docs/business-plan-faq/
• https://support.atlassian.com/loom/docs/understand-workspaces-roles-and-permissions/
• https://support.atlassian.com/loom/docs/what-is-loom
• https://www.atlassian.com/blog/loom/loom-atlassian
• https://techcrunch.com/2024/08/29/atlassian-acquires-rewatch-as-it-gets-into-ai-meeting-bots/
• https://www.capterra.com/p/191187/Loom/pricing/
• https://getlatka.com/blog/loom-revenue
• https://clipfold.io/compare/loom-pricing