Otter.ai

Name: Otter.ai
Brand: Otter.ai
Price: 08.33 USD

The original meeting transcription service, now a metered knowledge engine

Otter.ai is an AI meeting assistant that sends a bot into Zoom, Google Meet, and Microsoft Teams calls (or records locally from a desktop app), transcribes the conversation in real time with speaker identification, and produces summaries, action items, and an AI chat you can query across your whole meeting archive; it is metered in transcription minutes per user per month rather than sold as an unlimited seat.

Visit website

Overview

Otter.ai is the oldest continuously operating product in this category. It launched in 2016 as AISense, built its own speech recognition rather than reselling someone else's, and spent years as the default answer when anyone asked how to get a transcript of a call. That history shows up in the product as both strength and baggage: the transcription engine is mature and the real-time live transcript view is still the best in the category for following along during a call, but the pricing model is a minute meter inherited from an era when transcription was expensive, and it is the single most important thing to model before you buy.

The company has spent 2025 and 2026 repositioning from notetaker to what it calls a Conversational Knowledge Engine, wrapping the transcript archive in a set of role-specific agents: a Sales Agent that syncs to Salesforce and HubSpot, a Recruiting Agent that writes to Greenhouse, an SDR Agent that runs live demos, plus education and media variants. Otter reported crossing $100M ARR at the end of 2025, which makes it one of the very few profitable-scale independents in a category otherwise made of venture-funded startups.

For a small business the decision comes down to two questions. First, does the free tier's 300 monthly transcription minutes and 30-minute meeting cap fit your actual load, because for most people it does not. Second, are you comfortable with a bot named after the vendor appearing in your client calls, and with a vendor that trains models on de-identified customer recordings by default. Otter was hit with a proposed class action in August 2025 over recording consent for non-users, which is the clearest signal yet that this category has a legal edge to it.

Set against Granola, Otter is the opposite architecture. Granola captures your own machine's audio with nothing in the participant list and stores no recording; Otter's default path puts an Otter.ai participant in the room and keeps the audio. Otter has since shipped a desktop app that records without a bot, which narrows the gap, but the bot remains the default experience and the one your client will see.

Best for

Individuals and teams who want a mature, accurate transcript of every call in a searchable archive, who are comfortable with a visible bot, and who either fit inside the free 300 minutes or are ready to pay for the unlimited Business tier at roughly $20 per seat annually.

Not the right fit for

Anyone who cannot have a vendor-named participant visible in a client or candidate call; the bot is the default path and the desktop no-bot app is a secondary experience rather than the product's spine.
Teams working primarily in languages other than English, Spanish, French, German, Japanese, or Chinese; Otter's supported language list is short next to Fireflies, Circleback, or Sembly, and quality outside English is not the reason people buy it.
Buyers who need a hard guarantee that no customer audio is ever used for model improvement; Otter trains on de-identified data by default and the opt-out is not a headline setting.
Anyone in 20 meetings a week hoping to live on the free plan; 300 minutes covers roughly six or seven calls a month and then stops.
Sales managers who want scorecards, playbook adherence, and rep coaching analytics; the Sales Agent syncs fields and drafts follow-ups but is not a revenue intelligence platform in the way Grain or Avoma are.

How it works

1
You connect Google Calendar or Microsoft Outlook and choose which meetings Otter should attend. For scheduled calls, OtterPilot joins as a participant named after the vendor and begins transcribing; everyone in the room sees it in the attendee list.
2
During the call a live transcript streams in the web or mobile app, with speakers separated and labelled once you have trained Otter on voices. You can highlight, comment, and add photos to the transcript while it is running, which is the feature journalists and students have used Otter for since long before AI summaries existed.
3
When the meeting ends Otter produces an outline, a summary, and an action item list, and pushes them to whichever destinations you have connected. Paid tiers add automatic email follow-ups and CRM writeback into Salesforce and HubSpot.
4
Everything lands in a searchable archive. Otter AI Chat answers questions across meetings, so you can ask what a specific account said about renewal timing and get an answer drawn from the transcripts rather than from a single call. A Claude MCP connector exposes the same archive to external assistants. A separate desktop app for macOS and Windows records the local machine with no bot in the meeting, for the calls where a visible recorder is not acceptable.

Feature breakdown

30 features in 5 modules

Capture

Two paths, a bot and a local recorder, with the bot as the default.

OtterPilot meeting bot: Joins scheduled Zoom, Google Meet, and Microsoft Teams calls automatically from your calendar and appears in the participant list under the vendor's name.
Desktop app with no bot: A macOS 12.3 and Windows 10 desktop app records the local machine's audio directly, with no bot in the meeting, for calls where a visible recorder is not appropriate.
Concurrent meeting limits: Free covers 1 concurrent meeting, Pro 2, and Business 3, which matters if you send the bot to calls you are not personally attending.
Mobile and in-person recording: iOS and Android apps record face-to-face conversations, interviews, and lectures as well as calls.
Audio and video file import: Upload existing recordings for transcription. Free allows 3 lifetime imports, Pro 10 a month, Business unlimited, which is the cleanest illustration of how hard the free tier is capped.
Chrome extension: Starts and manages capture from the browser without switching to the Otter app.

Transcription and live view

The part Otter has been iterating on for a decade.

Real-time streaming transcript: Text appears as people speak rather than after the call, which is the reason Otter is still used for interviews, lectures, and accessibility rather than only for summaries.
Speaker identification: Separates speakers and labels them by name once voices are trained, so the transcript reads as a dialogue rather than a wall of text.
Six transcription languages: English (US and UK), Spanish, French, German, Japanese, and Chinese (Simplified). This is a genuinely short list compared with competitors claiming 50 to 100 languages.
Custom and team vocabulary: Teach Otter your product names, acronyms, and people's names so they stop being mangled; team vocabulary is shared across the workspace on paid tiers.
Inline photos and highlights: Capture slides or whiteboards into the transcript at the moment they appear, and highlight passages live.
Editable transcripts: Corrections you make persist and are reflected in downstream summaries and exports.

Summaries, agents, and search

The 2025 and 2026 repositioning from transcript tool to knowledge layer.

Automated outlines and summaries: Each meeting gets a structured outline plus a narrative summary rather than only raw text.
Action item extraction: Commitments are pulled out with owners where the conversation makes them clear.
Otter AI Chat across meetings: Query one meeting or the whole archive conversationally. This is the feature that turns years of transcripts into something you actually revisit.
AI workflows: Pro adds prebuilt AI workflows and Business adds custom ones, so a recurring meeting type can produce a consistent output shape and trigger downstream actions.
Role-specific agents: Sales, Recruiting, SDR, Education, and Media agents package the archive for a job function, with the Sales Agent handling follow-ups and CRM sync and the Recruiting Agent writing to Greenhouse.
Archive search: Full-text search across every transcript in the workspace, with sharing and folder organisation on top.

Integrations and export

Broad, and unusually strong on raw data export.

CRM sync: Salesforce and HubSpot integrations are available from the Pro tier, which is early compared with most competitors.
Claude MCP connector: Exposes transcripts and action items through the Model Context Protocol so you can search, recap, and draft follow-ups from inside an AI assistant.
Task tools: Asana and ClickUp receive summaries and action items so next steps land where work is tracked.
Storage and data destinations: Amazon S3, Dropbox, Egnyte, and Airtable receive transcripts, summaries, and metadata, which is a more serious export story than most notetakers offer.
Otter API: Programmatic access is an Enterprise-tier feature, not available on Business.
Slack: Summaries route into channels automatically after a call.

Administration and compliance

Mature controls, with a genuinely contested data-training posture.

SOC 2 Type II: Otter holds a SOC 2 Type II attestation report and offers a GDPR data processing agreement.
HIPAA compliance: Announced in 2025 and positioned as an Enterprise add-on with a BAA, which puts Otter in reach of healthcare-adjacent buyers.
SSO and SCIM: Single sign-on and directory provisioning are Enterprise-tier only, including Okta and Microsoft Entra.
Configurable retention: Administrators can set how long transcripts and audio persist, with role-based access and audit logging on Enterprise.
Model training on de-identified data: Otter states it de-identifies user data before using it to improve its models and that recordings are not manually reviewed by humans. This default is the single most argued-about thing about the product.
Video replay: Video capture and replay is an Enterprise-tier feature; lower tiers keep audio and text.

Use cases

4 documented

Solo consultant billing by the hour

Client calls need an accurate record for scope disputes, and the consultant would rather not retype notes after every session.

Otter Pro's 1,200 monthly minutes cover roughly 26 hour-long calls, transcripts are searchable years later, and the S3 or Dropbox export means the record survives even if the consultant changes tools.

Journalist or researcher running interviews

Long recorded interviews need quotable, timestamped text quickly, and the live transcript matters because follow-up questions depend on what was just said.

The real-time streaming transcript with speaker labels and inline photos is still the strongest in the category for this job, and unlimited imports on Business handle a backlog of archive tape.

Five-person sales team on HubSpot

Reps forget to log calls, and the manager has no view of what was actually said on a deal beyond a one-line CRM note.

Business at roughly $20 a seat annually gives unlimited transcription, HubSpot sync, and AI Chat across the archive, so deal context is queryable without asking a rep to remember.

Operations lead building a searchable company memory

Decisions get made in calls and then lost, and nobody can reconstruct why a supplier was chosen eight months ago.

Every meeting lands in one searchable archive with AI Chat over it and an MCP connector into Claude, so the answer comes back from the transcripts instead of from someone's recollection.

Pricing

from $0 (Basic), then $8.33 per user per month (Pro, billed annually)

Freemium per-seat subscription with a transcription-minute meter on the lower tiers. Free and Pro are metered in minutes per user per month; Business removes the meter and switches to a flat per-seat price.

Plan	Price	Includes
Basic	$0 per user per month	300 transcription minutes per month 30 minutes maximum per meeting 1 concurrent meeting 3 lifetime audio or video file imports Live transcription, speaker ID, and AI Chat Roughly six or seven hour-long calls a month before the meter stops you. Treat it as a trial, not a plan.
Pro	$8.33 per user per month billed annually ($16.99 monthly)	1,200 in-app recording minutes per month Up to 90 minutes per meeting 2 concurrent meetings 10 file imports per month Salesforce and HubSpot integrations, team vocabulary, prebuilt AI workflows 1,200 minutes is about 26 hour-long meetings, which is under half of what a heavy calendar produces.
Business	$19.99 per user per month billed annually ($30 monthly)	Unlimited meetings and recordings, no minute meter Up to 4 hours per meeting 3 concurrent meetings Unlimited file imports Custom AI workflows, admin features, prioritised support The only tier a genuinely busy calendar can live on, and the one this profile prices against.
Enterprise	Custom quoted, demo required	SSO and SCIM Otter API access Video capture and replay HIPAA compliance with a BAA as an add-on Custom retention, audit logging, customer success programme The API and video replay both live here, which is a notable gap in the self-serve tiers.

Billing notes

The annual discount is unusually steep: Pro drops from $16.99 to $8.33 and Business from $30 to $19.99, so month-to-month buyers pay roughly 50 percent more for the same product.
The meter on Basic and Pro is transcription minutes per user per month and it does not roll over. There is no published overage rate; you stop transcribing until the month resets.
Per-meeting duration caps are a second, separate limit. A 100-minute board call is simply truncated on Pro at 90 minutes even if you have minutes left.
Downgrading does not delete your archive, but the minute meter reapplies immediately, so a team that drops from Business to Pro can hit the wall inside two weeks.
Video capture and replay plus the API are Enterprise-only, so any self-serve buyer should assume they are buying audio and text.

Value assessment: Model this in minutes before anything else. Someone in 20 meetings a week averaging 45 minutes generates about 3,900 transcription minutes a month. Basic covers 300 of them and Pro covers 1,200, so that person is on Business whether they like it or not: roughly $240 a year per seat on annual billing, or $360 month to month. At $20 a seat for unlimited transcription, mature speech recognition, CRM sync, and AI Chat over the archive, that is fair, and it is the same neighbourhood as Fireflies Business and Fathom's paid tiers. What Otter does not do is give you a workable cheap tier: the $8.33 Pro plan looks like the value option and is not, because 1,200 minutes is under half a real calendar. Budget for Business or use the free tier honestly as a trial.

Strengths & limitations

Strengths

The most mature transcription engine in the category, built in-house since 2016 rather than wrapped around a third-party model, with the best live streaming transcript view for following a call in real time.
Business removes the minute meter entirely for roughly $20 a seat annually, which is competitive with everything else at that price.
Unusually serious data export: Amazon S3, Dropbox, Egnyte, and Airtable destinations mean your transcripts do not become hostage to the vendor.
Salesforce and HubSpot sync arrive at the Pro tier rather than being reserved for the top plan, which is earlier than most competitors.
A real desktop app that records with no bot in the meeting, giving you an escape hatch for the calls where a visible recorder is unacceptable.
SOC 2 Type II, a GDPR DPA, and HIPAA compliance with a BAA on Enterprise, which is a broader compliance surface than most startups in this list can offer.
Financially independent at scale, with a reported $100M ARR milestone at the end of 2025, so the vendor risk is lower than for a seed-stage competitor.

Limitations

Six transcription languages is a short list. If a meaningful share of your calls are in Portuguese, Italian, Dutch, Hindi, or Korean, Otter is the wrong tool and Fireflies or Circleback are the obvious alternatives.
The free and Pro minute meters are tight enough to be misleading: Pro reads like the sensible plan and covers under half of a busy calendar.
Model training on de-identified customer data is the default, and the opt-out is not presented as prominently as Granola's is.
A proposed class action filed in August 2025 alleges Otter recorded people who were not Otter users without adequate consent. It is unresolved, but it is a real reason to tighten your own consent practice.
The API and video replay are locked behind an Enterprise quote, so a self-serve buyer cannot build on Otter or review recorded demos.
The agent proliferation (Sales, SDR, Recruiting, Education, Media) has outpaced depth in any one of them; none is a substitute for a purpose-built revenue intelligence tool.

Head-to-head comparisons

6 alternatives

Otter.ai vs Granola

from $0 (Basic), then $14 per user per month (Business)

Granola captures your own machine's audio with no participant in the room and keeps no desktop recording, for $14 a seat with no meter. Otter's default is a visible bot, it retains audio, and it meters minutes until you reach the $20 Business tier. Choose Granola if your meetings are confidential conversations and you want notes that read like you wrote them; choose Otter if you want an accurate verbatim archive, live transcript, and CRM sync, and the bot is not a problem.

Full Otter.ai vs Granola comparison

Otter.ai vs Fireflies.ai

from $0 (Free), then $10 per user per month billed annually (Pro; $18 monthly)

Both are bot-based archive platforms at a similar price. Fireflies supports far more languages, has a deeper integration catalogue, and pushes conversation intelligence harder. Otter has the better live transcript, the stronger raw export story, and a longer track record. Pick Fireflies for multilingual teams and automation breadth; pick Otter if English-language transcription quality and a searchable archive are the actual job.

Full Otter.ai vs Fireflies.ai comparison

Otter.ai vs Read AI

from $0 (Free), then $15 per user per month billed annually (Pro)

Read AI wraps meetings, email, and messages in one engagement layer with sentiment and speaker analytics, and its free tier allows only 5 transcripts a month. Otter's free tier is metered in minutes instead and its focus is verbatim accuracy rather than meeting behaviour scoring. Take Read AI if you want coaching signals across all your communication; take Otter if you want the transcript to be right.

Full Otter.ai vs Read AI comparison

Otter.ai vs Notta

from $0 (Free), then $8.17 per month billed annually (Pro)

Notta is the cheaper minute-metered transcription tool with a much longer language list and strong Japanese and Chinese support, at $8.17 a month for 1,800 minutes. Otter costs about the same at Pro for fewer minutes but a better meeting workflow, CRM sync, and agent layer. Notta wins on raw transcription value and language coverage; Otter wins as a team meeting system.

Full Otter.ai vs Notta comparison

Otter.ai vs Krisp

from $8 per user per month billed annually (Core), $16 month to month

Krisp captures bot-free from the local audio path, keeps recordings, adds noise cancellation, and charges $8 a seat annually with no minute meter, but caps storage at 10 GB and writes weaker notes. Otter charges about $20 for unlimited transcription with better recognition, a live transcript, and deep export, at the cost of a visible bot. Choose Krisp if a recorder in the participant list is unacceptable or the audio quality matters; choose Otter if the transcript itself is the deliverable.

Full Otter.ai vs Krisp comparison

Otter.ai vs MeetGeek

from $0 (Basic), then $9.99 per user per month (Pro)

MeetGeek meters in transcription hours rather than minutes and publishes an explicit overage rate of $0.50 per extra hour, which makes it more predictable at the edges, and it is cheaper at $9.99 for 20 hours. Otter has better speech recognition and a larger ecosystem. Choose MeetGeek if you want a transparent hourly meter and video storage on a small budget; choose Otter if accuracy and the archive matter more than the unit price.

Full Otter.ai vs MeetGeek comparison

Implementation & onboarding

Setup time: Under fifteen minutes. Sign up, connect Google or Microsoft calendar, choose whether OtterPilot auto-joins all meetings or only ones you approve, and the next call is captured. Training speaker voices takes a few meetings before names attach reliably.
Learning curve: Low. The interface has accumulated a decade of features and the agent menu is busier than it needs to be, but the core loop of join, transcribe, summarise, search needs no explanation.
Onboarding: Fully self-serve through Business. Enterprise adds a customer success programme. The only configuration worth doing deliberately is the auto-join policy and the vocabulary list.
Migration notes: Otter imports audio and video files for transcription, so you can backfill an archive from another tool if you still have the media, subject to the import caps (3 lifetime on free, 10 a month on Pro, unlimited on Business). Leaving is easier than most: transcripts and summaries export to S3, Dropbox, Egnyte, and Airtable, which is the best exit story in this category.

Platform, API & security

Platforms: WebmacOS desktop app (12.3 or later)Windows desktop app (10 or later)iOSAndroidChrome extension
API: The Otter API is an Enterprise-tier feature and is not available on self-serve plans. A Claude MCP connector exposes transcripts and action items to AI assistants.
Compliance: SOC 2 Type II attestationGDPR with a DPAHIPAA compliance with a BAA available as an Enterprise add-on
Data residency: US-hosted. Otter does not advertise regional data residency options for self-serve customers.
SSO: SSO and SCIM provisioning, including Okta and Microsoft Entra, are Enterprise-tier only.
Security notes: Encryption in transit and at rest, configurable retention periods, role-based access, and audit logging on Enterprise. Otter states it de-identifies customer data before using it to improve its models and that recordings are not manually reviewed by humans; a proposed class action filed in August 2025 challenges the consent basis for recording non-users, and remains unresolved.

Support & resources

Channels: Email supportHelp centre and in-app supportPrioritised support on BusinessCustomer success programme on Enterprise
Documentation: Help centre at help.otter.ai covering setup, supported languages, OtterPilot behaviour, AI Chat, integrations, and HIPAA.
Community: Large installed base and heavy third-party coverage; no substantial official user forum.

Company

Founded: 2016
Headquarters: Mountain View, California, United States
Ownership: Venture-backed (operating company AISense, Inc.)
Founders: Sam Liang, Yun Fu
Employees: Roughly 200 to 280 depending on the tracker; the company has described itself as deliberately lean relative to revenue
Funding: Approximately $63M to $70M raised across four rounds, including a $50M Series B in 2021 led by Spectrum Equity. Otter announced crossing $100M ARR at the end of 2025.

Funding history

Round	Amount	Year	Notes
Seed and Series A	Undisclosed early rounds	2017	Backed by Horizons Ventures, GGV Capital, and Draper Associates among others.
Series B	$50M	2021	Led by Spectrum Equity; the last large publicly reported round.
Milestone	$100M ARR	2025	Announced December 2025 alongside the AI Meeting Agents launch.

Timeline

2016Founded as AISense by Sam Liang and Yun Fu in Mountain View, building proprietary speech recognition rather than licensing it.
2018Otter Voice Notes launches publicly and becomes the default free transcription tool for journalists, students, and researchers.
2021Raises a $50M Series B led by Spectrum Equity and launches OtterPilot, the meeting bot that automates joining calendar calls.
2024Expands transcription beyond English to Spanish and French, later adding German, Japanese, and Chinese (Simplified).
2025Achieves HIPAA compliance, faces a proposed class action over recording consent for non-users, and closes the year announcing $100M ARR and a suite of role-specific AI Meeting Agents.
2026Positions the archive as a Conversational Knowledge Engine, adds a Claude MCP connector, and ships a desktop app that records without a bot.

Integrations

Zoom, Google Meet, and Microsoft Teams
Google Calendar and Microsoft Outlook Calendar
Salesforce and HubSpot (Pro and above)
Slack
Asana and ClickUp
Greenhouse (via the Recruiting Agent)
Amazon S3, Dropbox, and Egnyte
Airtable
Claude via MCP
Otter API (Enterprise only)

Frequently asked questions

10 questions

What is Otter.ai?

Otter.ai is an AI meeting assistant that transcribes calls in real time, identifies speakers, and produces summaries, action items, and a searchable archive you can query with AI chat. It joins Zoom, Google Meet, and Microsoft Teams as a bot from your calendar, or records locally through a desktop app with no bot in the meeting. It has been operating since 2016 and is the oldest product in this category.

How much does Otter.ai cost?

Basic is free with 300 transcription minutes a month and a 30-minute cap per meeting. Pro is $8.33 per user per month billed annually, or $16.99 monthly, for 1,200 minutes and 90-minute meetings. Business is $19.99 annually or $30 monthly per user and removes the minute meter entirely with a 4-hour meeting cap. Enterprise is quoted and adds SSO, the API, video replay, and HIPAA.

How is Otter metered, and what would 20 meetings a week actually cost?

Free and Pro are metered in transcription minutes per user per month; Business is a flat seat price with no meter. Twenty meetings a week at 45 minutes each is roughly 3,900 minutes a month, which blows through both the 300-minute free allowance and the 1,200-minute Pro allowance well before the month ends. That person needs Business: about $240 a year per seat on annual billing, or $360 if they pay monthly. Minutes do not roll over and there is no published overage rate, so you simply stop transcribing.

Does Otter put a bot in my client calls?

By default, yes. OtterPilot joins as a participant named after the vendor and is visible to everyone in the room, which some clients find intrusive and some regulated buyers will not allow. Otter now ships a macOS and Windows desktop app that records your machine's audio with no bot in the meeting, which is the path to use for sensitive calls, but the bot remains the default experience.

Does Otter keep my audio and video?

Audio is retained alongside the transcript so you can play back what was said, and administrators can configure retention periods. Video capture and replay is an Enterprise-tier feature, so self-serve Basic, Pro, and Business customers are getting audio and text only. If you want recorded video of demos for coaching, Otter's paid self-serve tiers will not give it to you.

Does Otter train its AI on my recordings, and can I opt out?

Otter states that it de-identifies customer data before using it to improve its speech models and that audio and transcripts are not manually reviewed by humans. That means training on customer data is the default posture rather than something you opt into. Opt-out and stricter handling are available, but they are negotiated at the Enterprise level rather than surfaced as a one-click setting the way Granola's is. If a no-training guarantee is a hard requirement, get it in writing before you buy.

What languages does Otter support?

English (US and UK), Spanish, French, German, Japanese, and Chinese (Simplified). This is a materially shorter list than Fireflies, Circleback, Notta, or Sembly, all of which claim 40 to 100 or more. Non-English quality is credible in the supported set but the safe assumption is that Otter is an English-first product and that anything outside those six languages is not covered at all.

How does Otter handle consent in two-party recording states?

Otter provides in-meeting notification through the bot's presence in the participant list, and the bot being visible is itself the main consent mechanism. That is not sufficient on its own in two-party consent jurisdictions such as California, Illinois, Pennsylvania, and Washington, or under GDPR in the EU. You are responsible for announcing recording and getting agreement before you start. A proposed class action filed in August 2025 alleges Otter recorded non-users without adequate consent, which is a strong argument for adding a verbal announcement to the top of every recorded call rather than relying on the bot being noticed.

Can Otter write back to my CRM?

Yes, on Pro and above. The Salesforce and HubSpot integrations attach call summaries, transcripts, and action items to the relevant records, and the Sales Agent handles follow-up drafting alongside the sync. The field-level mapping is shallower than a dedicated revenue intelligence platform, so treat it as attaching context to the record rather than populating a structured deal-review schema.

Is Otter a good choice for a sales team?

It is a reasonable choice for capture and CRM hygiene and a poor one for coaching. You get every call transcribed, summarised, and synced to Salesforce or HubSpot, plus AI Chat across the archive so deal context is queryable. What you do not get on self-serve tiers is video playback, scorecards, playbook adherence tracking, or rep-level talk analytics. A manager who wants to coach should look at a conversation intelligence tool; a manager who wants the calls logged should be fine with Otter Business.

Editorial verdict

Otter is the safe, boring, mature choice in a category full of two-year-old startups, and it is priced accordingly once you understand the meter. Ignore the free and Pro tiers as anything more than an evaluation: a real calendar produces four times the minutes Pro allows, so the honest price is Business at about $20 a seat annually, where the meter disappears and you get unlimited transcription, CRM sync, AI Chat over the archive, and the best raw export story anyone offers. Buy it if you want an accurate, durable, searchable record of everything said, in English or one of five other languages, and a visible bot in the room is acceptable. Do not buy it if your calls are multilingual, if a vendor-named participant would embarrass you in front of a client, or if you need a contractual guarantee that your recordings never touch a training pipeline; Granola and Circleback answer those objections directly and Otter does not.

Written by the SaaSTracker editorial team. Awards, when shown, are judged against the published criteria in our methodology.