# Amberscript Financial Model AI + human hybrid transcription and subtitling platform targeting B2B customers in Europe - Canonical: https://finamodel.com/startups/amberscript - Excel download: https://finamodel.com/startup-models/amberscript.xlsx - Category: AI/ML - Model type: SaaS ARR / Valuation - Funding round: Series A - Funding: $10M - Founded: 2021 - Geography: Europe primary (European legislation cited as demand driver [DECK slide 5]); market size stated globally - Customer: B2B ## About the company Amberscript turns uploaded audio and video into transcripts and subtitles through a hybrid workflow: proprietary speech recognition produces a draft, then human editors correct and quality-check the final output. Its self-learning engine improves from those corrections. The European B2B platform addresses demand created by video consumption and subtitle legislation. Customers can use automatic output or a premium human-perfected tier, pay according to minutes processed, and access the product through self-serve or enterprise and API channels. The model is volume-led rather than seat-led: minutes by tier times price drive revenue. It then measures the automated-versus-human mix, editor labour, ASR compute, gross margin, scaling headcount, operating cash flow, and runway under adoption and pricing scenarios. ## What's included - 5-year monthly revenue build with stage-appropriate growth assumptions - Full P&L, headcount plan, and operating-expense schedule - Cash-flow statement, runway, and burn-rate tracking - Valuation via exit multiple with a DCF cross-check - Returns analysis with MOIC and IRR - Unit economics including CAC, LTV, payback, and cohort retention ## Product & value proposition - Hybrid "Assisted Intelligence" workflow: user uploads audio/video → proprietary ASR engine generates draft transcript → human editors correct and QA → final transcript or subtitle file delivered - Self-learning ASR engine improves with each job (data flywheel) - Claims 100% accuracy output (vs. ~75% for pure automatic transcription) at faster turnaround than manual-only - Software integrates with customer software suites - Manual transcription benchmark: 8 hours to transcribe 1 hour of audio; Amberscript frames itself as materially faster - Rated on Trustpilot: 377 reviews ## Market - TAM: €33 billion - global transcription market size - Demand drivers cited: video consumption boom; European legislation mandating subtitles for organisational video; untapped ASR opportunity in transcription market ## Revenue model - Usage-based pricing - Unit of consumption: audio/video minutes transcribed or subtitled - Channels: self-serve web platform (Free Trial visible in product screenshot); likely API/enterprise channel given B2B focus and slide 9 reference to API-led growth - Two quality tiers implied: automatic-only (lower price/speed) vs. human-perfected (premium) ## Traction & metrics - Trustpilot: 377 reviews - only quantified traction metric shown - Timeline: Product-market fit achieved 2018–2019; Scaling phase 2020–2023; Hyper-scaling from 2024+ ## Competition / moat - Competitive set named: pure automatic transcription (fast, 75% accuracy, B2C-acceptable) vs. traditional manual transcription (high accuracy, 8h per 1h audio, fragmented market) - Amberscript positioned top-right of speed × accuracy matrix - high on both axes - Moats: proprietary ASR engine; data flywheel (unique dataset from human corrections improves model over time); Trustpilot social proof; software integrations creating switching cost ## Team & funding ask / use of funds - Founder & CEO: Peter-Paul de Leeuw --- ## Recommended financial model - **Archetype + why:** Usage-based SaaS with human-fulfillment cost layer. The deck explicitly endorses usage-based monetization (slide 9), the product is priced per audio/video minute, and there is a variable COGS component (human editor time) that scales with volume. This is not a pure SaaS seat model - it is closer to a transactional/marketplace P&L with a gross margin wedge between ASR automation and human QA labor. - **Forecast horizon & granularity:** Monthly for Year 1–2; quarterly for Year 3–5. Five-year horizon appropriate for a growth-stage B2B platform targeting hyper-scaling from 2024+. - **Key drivers & assumptions:** | Driver | Value | | -- | -- | | Minutes processed (starting volume) | To be sourced from company | | Revenue per minute - automated tier | ~€0.10–0.15/min | | Revenue per minute - human-perfected tier | ~€1.00–2.00/min | | Mix: automated vs. human-perfected | 40% / 60% to start, shifting to 60% / 40% as ASR improves | | Volume growth rate (MoM) | 8–12% in scaling phase | | Human editor cost per minute of audio | €0.40–0.60/min | | ASR cost (cloud compute) per minute | €0.01–0.02/min | | Blended gross margin | 50–65% | | European legislation tailwind | Structural demand driver; modelled as accelerating adoption curve post-2024 | - **Scenarios (Base / Bull / Bear):** - **Base:** Volume grows 10% MoM; mix holds; human editor costs stable - **Bull:** ASR accuracy improvement reduces human QA requirement → gross margin expands to 70%+; volume growth 15% MoM driven by legislation mandates - **Bear:** Competition from commoditized ASR (OpenAI Whisper etc.) compresses pricing; gross margin falls to 40%; growth slows to 5% MoM - **Required sheets / outputs:** 1. Assumptions - all drivers in one place 2. Volume & Revenue - minutes by tier × price per minute → ARR/MRR bridge 3. COGS - human editor headcount/cost + ASR compute cost 4. Gross Profit & Gross Margin % 5. OpEx - S&M, R&D, G&A (headcount-driven) 6. P&L (Income Statement) - monthly → annual summary 7. Cash Flow & Runway 8. KPI Dashboard - minutes processed, revenue per minute, gross margin %, NRR, CAC payback 9. Scenario toggle (Base / Bull / Bear) ## Frequently asked questions ### Is the Amberscript financial model free? Yes. The Amberscript model is a free Excel download with live formulas. ### Can I change the assumptions? Yes. The workbook is editable and its live formulas recalculate when assumptions change.