Generative media · USA (Palo Alto)
OpusClip
A single feature inside a failed live-streaming tool became a product whose watermarked clips advertise themselves; the user and view claims are large, revenue and retention are not official.
Verdict: Gaps remain
Updated: 2026-10-06
01
Profile
- What it does
- Turns long video (YouTube, podcasts, live streams) into short, captioned vertical clips with AI and picks them with a "virality score". OpusSearch (March 2025) monetizes video archives, Agent Opus produces short video end to end [1].
- Founded
- January 2022; first the "Opus" live-streaming tool, then OpusClip as a separate product in May 2023 (secondary account) [4]
- Founders
- Young Zhao (CEO) [4]; co-founder names do not appear in the sources opened
- Pricing model
- Free $0, Starter $15/month, Pro $29/month, Business custom. Caption templates on Free are "With watermark"; Starter and above get "Watermark-free video export". Pro and Business include "AI video B-Roll (powered by Agent Opus)" [2].
02
Growth timeline
2022-01
Opus live-streaming tool; about 200 users in three months (secondary account)
[4]2023-05
OpusClip beta; a five-person team built it beside a 24-person company (secondary account)
"Opus Clip hit $1M ARR in its first month after the beta launch"
[4]2025-03-11
SoftBank round; 10M creators and brands; OpusSearch launch
"172 million video clips created in the past year"; "57 billion" views
[1]2025
$20M revenue, 150% annual growth (Sacra estimate)
[3]2026-01-29
CEO interview: a real business, not a cool demo
"做一个真实的生意,而不是做一个很酷的演示"
[5]
03
Growth loop
- 1Long video uploaded
- 2Watermarked clips on social
- 3Other creators see it, sign up
- 4Pay to remove the watermark
- Leak: platform dependency: value sits with the algorithm
The output is the distribution: every watermarked clip is an ad. The red arrow is a reminder that the clips' value depends on the TikTok and YouTube algorithms.
The drawing shows the shape of the loop; each node is explained in the list below. The red arrow marks value leaking out of the loop.
04
Mechanism
- Acquisition channels
- The output itself is the channel: watermarked captions on the free plan [2], clips circulating on TikTok, Shorts and Reels; the company claimed 172 million clips in the past year and 57 billion total views [1]. (opinion) The watermark's contribution to acquisition was never measured or disclosed by the company. Early on, sales were semi-manual: the CEO describes AI-assisted edits cleaned up by hand and sent one email at a time, with about 60% of replies saying "I like it, I want to use it" [5]. Brand customers include Univision, Billboard, iHeartMedia, Visa and LinkedIn [1].
- Loop
- A free user uploads a long video, watermarked clips go to social media, creators among the viewers see the product and sign up, pay to remove the watermark, and produce more clips. A second loop runs through OpusSearch and Agent Opus, from archives to revenue and to business plans [1].
- Activation
- A set of clips within minutes of the first long video. According to the secondary account, pre-payment users asked to pay to skip the queue or get higher resolution [4]. The CEO: when users start complaining about these things, you are usually at the edge of product-market fit [5].
- Retention evidence
- No official retention figure. "Positive ROI within two months" is a marketing claim [1]. No retention evidence.
- Role of pricing
- A low entry price ($15) and watermark removal as the conversion trigger [2]. (opinion) Credits tied to minutes push users up a plan as usage grows.
- Distribution moves
- OpusSearch (March 2025) and the SoftBank signature [1]; Agent Opus B-roll on the pricing page [2]. Communities, X and Product Hunt as early channels (secondary account) [4].
- Team
- A 24-person company at the time of the pivot, with a core of 5: three engineers, a product manager and a designer (secondary account) [4]. No current headcount.
05
Real reading
The five questions of the Growth Claim Check, applied to this company's public claims.
- 01
What was measured, and by whom?
yes
Sign-ups, clip counts and views are company totals; ARR appears only in secondary sources.
- 02
Perception or outcome?
no
57 billion views is the clips' success, not the product's; the effect on creators' income was not measured.
- 03
Is there a stability or quality metric?
no
Churn, margin and paid conversion are undisclosed.
- 04
Is there a counterfactual or cohort?
no
No control group or cohort disclosure.
- 05
Nominal or real?
unclear
10 million creators is a sign-up count with no active or paid split. The ARR progression looks consistent but is secondary.
Verdict
Gaps remain
The loop, output as distribution, is convincing; measurement is limited to volume and the revenue figures are not official.
Test your own claim with the same five questions: Growth Claim Check →
06
Shadow side
- No official ARR disclosure; all revenue figures are relayed from a CEO blog post that could not be located [4] and from Sacra estimates [3].
- Platform dependency: the clips' value rides on the TikTok and YouTube algorithms; the "virality score" promise is not measurable. (opinion)
- A failed first product: 5 people built the right product while a 24-person team worked on the wrong one (secondary account) [4]; the cost of the pivot is not told.
07
What it teaches a founder in Türkiye
This section is opinion; a reading, not a source.
- 01
(opinion) If your output gets shared, the watermark is your marketing budget. Free plan plus a visible signature; the cheapest channel for creator tools in Turkey.
- 02
(opinion) Make the one feature users love into the product. In the CEO's words, nobody bought the live-streaming tool, but the auto-clipping feature inside it gave an early positive signal.
- 03
(opinion) Sell semi-manually, then automate. One email at a time and 60% "I want it" replies; demand first, infrastructure second.
- 04
(opinion) Do not confuse a volume metric with a revenue metric. 57 billion views look good to investors; the founder should watch paid conversion and month-three retention. Test question: if views went to zero, how much would revenue change?
- 05
(opinion) Low entry price and a clear trigger. The trigger is the same when pricing in lira: watermark, resolution, queue.
A question for you: remove this company's logo and drop the loop onto your own product. Which node stays empty?
08
Sources
Numbers are the references in the text. An 'opened' source was read by the researcher; 'secondary' ones are verified against their primary source before publication.
- [1]OpusClip raises a new round of funding from SoftBank and launches OpusSearch
OpusClip blog · 2025-03-11 · opened
“172 million video clips created in the past year”
- [2]
- [3]OpusClip profile (estimates)
Sacra · 2026 · opened
Note: 2025 revenue estimated at $20M with 150% growth; $50M total funding
- [4]Opus Clip: $1M ARR in Its First Month After Beta Launch
Startup Founder Stories (secondary account) · 2024 · opened
“Opus Clip hit $1M ARR in its first month after the beta launch · After 3 months, they had only 200 users · A 5-person squad (3 engineers, 1 PM, 1 designer) quietly rebuilt the clipping feature as a standalone product while the 24-person team continued working on the failing tool.”
- [5]Young Zhao interview
人人都是产品经理 (woshipm) · 2026-01-29 · opened
“我们一开始做的是直播工具,整体没人买单,但里面有一个自动剪辑片段的功能,出现了很早的正反馈”
- [6]SoftBank Is Betting On The Future Of AI Content Creation With OpusClip
Forbes (Ian Shepherd) · 2025-03-13 · secondary
Note: 403; headline only
- [7]Young Zhao (CEO of OpusClip) on Building a $215M AI Startup
Silicon Valley Girl podcast · 2026-01 · secondary
Note: 403; headline only
09
Open questions
- The URL of Zhao's "$1M ARR in the first month" post on opus.pro.
- Round size and the $215M valuation: no primary source (Forbes 403).
- The Agent Opus launch post and date (August 2025, secondary).
- Paid conversion, churn, current team size: none.
- The watermark's contribution to acquisition: not disclosed by the company.
- Podcast transcripts could not be opened; the listen-and-watch part is missing.