
Anthropic IPO Roadshow Presentation (November 2026), reconstructed by SlideDeck Co
Uses this deck’s cover as a style reference for slides made from your own content. Or pick any page below to use its layout as the reference.
Quick Navigation
Tags
Share the slides
Anthropic IPO Roadshow Presentation (November 2026), reconstructed by SlideDeck Co
Detailed view of each slide page, including layout, key content and visual elements.
Cover slide naming the Anthropic roadshow presentation, dated November 2026, with a reconstruction note.
Oversized left-aligned headline on warm cream background with date top-right and a full-width band of three flat color panels (espresso, sage, olive).
Condensed S-1 legal disclaimer covering the preliminary prospectus, market data, forward-looking statements and non-GAAP measures.
Two-column body text layout with bold lead-in phrases and a source note above the footer.
IPO term sheet: 825M shares, $70 to $78 price range, Nasdaq: ANT, $26B concurrent private placement, use of proceeds, staggered lock-up and syndicate list.
Full-width two-column table with bold labels left and right-aligned values, separated by hairline rules.
Mission statement: the responsible development and maintenance of advanced AI for the long-term benefit of humanity.
Centered statement column flanked by a sage panel on the left and a terracotta panel on the right.
$17B Q3'26 revenue curve since 2023 launch, set against a $40T+ knowledge-work opportunity spreading across employees, companies and industries.
Left exponential line chart with labelled endpoint; right column with a giant stat and three stacked colored pills.
Three-part agenda (core advantages, enterprise adoption, financial model) with the first section highlighted.
Blue grid-pattern side panel with serif agenda items; active item in black, others greyed out.
Section opener listing five advantages: mission and focus, talent and culture, frontier and safety research, approach to compute, enterprise strategy.
Terracotta side panel with large bold title and serif bullet list.
Three pillars: singular focus, safety and research, and talent density.
Three full-width rounded cards, each with a bold label on the left and description on the right.
Seven co-founders with roles, research pedigree (GPT-2/3, scaling laws, RLHF, interpretability), 7 of 7 founders retained and ~1% regrettable attrition.
Row of seven portrait placeholders with names, above a tinted panel combining bullets and two stats.
Excellence in research and safety cards beside a Terminal-Bench chart of Opus 5.5 score by effort level against cost.
Two stacked tinted cards on the left, line chart in a white card on the right, methodology footnote below.
Mission-side work (alignment research, misuse detection) maps to business outcomes (enterprise trust, protected IP and safety standards).
Two-column cause-and-effect layout with grey cards on the left, coral cards on the right and an arrow between them.
Task complexity versus time horizon chart (2024 to today) with a list of frontier applications from cyber defense to supply chain automation.
Left area chart with green shaded regions and task labels; right column with headline and stacked application list.
A fungible compute pool spanning AWS Trainium, Google TPUs, NVIDIA and AMD GPUs feeds research, model development, customer demand and internal use.
Three-tier stack diagram: supplier cards at the bottom, blue pool band in the middle, workload cards on top, joined by arrows.
Explains compute multipliers: algorithmic gains that turn a unit of compute into a much larger effective unit.
Definition card and visual equation (small grid x multipliers =) on the left, large blue square grid on the right.
Reframes tokens, benchmark scores and models into business outcomes, useful intelligence and virtual collaborators.
Two rows of three cards (grey problem cards above, white answer cards below) linked by down arrows.
Improvement loop (better models, more use cases, more feedback) and investment loop (higher growth, more investment, more capabilities) around Claude.
Central circle with coral nodes on the left and blue nodes on the right forming two loops.
Code contributed per engineer rose from 1.0x in 2024 to over 45x in Q3'26; more than 90% of code is written by Claude.
Four stacked statement cards on the left, a highlight stat card and a terracotta bar chart on the right.
Agenda slide with the second section, enterprise adoption of Claude, highlighted.
Blue grid-pattern side panel with serif agenda; active item in black.
Model family ladder: Haiku, Sonnet, Opus and Mythos (including Fable), arranged by cost and intelligence.
Four ascending staircase cards in different tints with serif model names, cost and intelligence axis labels.
Composable platform: API, Agent SDK and Claude Managed Agents build internal agents and external products using memory, tool use, skills, MCP and more.
Layered diagram with blue outcome cards, coral build-path row with control/autonomy axis, and a row of eight capability chips.
Frontier intelligence flows through the platform (API, SDK) and products (Claude Code, Cowork, apps) to diffusion across sectors.
Left-to-right flow of three stages joined by arrows.
Autonomy thresholds crossed in coding (Nov '25) and cyber (Apr '26), with broader knowledge work next.
Illustrative timeline curve with milestone annotations from 2025 to today.
Run-rate revenue in the days after research preview shows Claude Cowork outpacing Claude Code at the same point.
Two-line comparison chart over days since launch, with footnotes.
Go-to-market channels: intelligence-led growth, direct sales, applied AI, the three major clouds and ecosystem partners.
Channel cards arranged around Claude diffusion.
More productive employees, faster processes and embedded intelligence, each mapped to Claude products.
Three-column pillar layout with product mapping under each pillar.
Case studies: Mercado Libre (100%+ more code merged), Bristol Myers Squibb (5 minutes instead of 2 months for evidence questions), Walmart (~35% higher order value).
Customer result cards, each with a headline metric and supporting detail.
Customers start with one use case and expand to core infrastructure; ~900% net revenue retention.
Staged expansion graphic with a large NRR stat.
Quarterly revenue for a quant trading firm (69x YoY), a semiconductor maker (168x) and a pharma company (115x).
Three small-multiple bar charts by sector.
10 of 10 of the largest software, internet and banking companies use Claude, plus 9 of 10 in pharma and semiconductors.
Grid of industry stat tiles.
Over 30% of the Global 2000 and over 50% of the Fortune 500 spend $100K+ with Claude, which is still under 1% of their expenses.
Headline stats with a horizontal spend-share bar.
Expanding into cybersecurity (129K+ verified vulnerabilities) and life sciences, from today to the long-term vision.
Matrix of business lines across today, building and long-term vision columns.
Agenda slide with the third section, the financial model, highlighted.
Grid-pattern side panel with serif agenda.
~$81B run-rate revenue, $17B Q3'26 revenue (12x+ YoY), 11.9K+ $100K customers, 43% compute margin, 85%+ gross margin and 12% adjusted operating margin.
KPI dashboard grid split into growth and efficiency groups.
Revenue curve from March 2023 to $17B in Q3'26 with model and product releases along it; 250 releases YTD.
Annotated growth curve with release markers.
Customers spending $100K+ grew from ~20 (2023) to over 11,900 (Sep 2026); 79% of revenue comes from outside the top-5 customers.
Bar progression with supporting stats.
Fable 5.1, Opus 5.5 and Haiku 5.5 lower customer cost by about 25%, 40% and 75% respectively, while margins hold.
Model comparison rows with cost-reduction callouts.
Monthly spend per Enterprise user is under $75 without Claude Code and under $1,500 for the top 5% of users (~20x).
Three-bar comparison chart.
Revenue grew >45x while compute cost grew ~10x; revenue per compute dollar rose from 0.40x to 1.72x+.
Quarterly paired bar chart with a ratio line.
Non-compute cost fell from 70% to 27% of revenue, helped by internal Claude use.
Stacked bar chart by cost category with side stats.
Adjusted operating margin went from (221)% in Q1'25 to 15% in Q3'26, up 236 points.
Quarterly margin bar chart crossing zero.
Compute tiers by cost: Anthropic-direct chips, multi-year cloud commitments and short-term peak capacity.
Tiered layout along a cost-per-unit axis.
$141B raised, $132B cash, $216B pre-IPO liquidity and $85B expected net IPO proceeds.
Grouped KPI tiles.
Q1 to Q3 2026 P&L ratios versus a steady-state target of 35%+ adjusted operating margin.
Financial table comparing quarters with steady state.
Appendix divider.
Minimal section divider.
Contractual obligations of $320B, staged by year from Q4'26 onward.
Multi-year obligations table.
Definitions of adjusted non-compute cost, adjusted operating income and margin, and other non-GAAP terms.
Dense definition text.
Net revenue retention and run-rate revenue methodology.
Dense definition text.
How the $40T+ knowledge-work value is estimated from BLS and ILO data.
Methodology text block.
FY2023 to FY2025 income statement: revenue from $41M to $4.585B.
Financial statement table.
Q1'25 to Q3'26 quarterly income statement: revenue from $376M to $16.956B.
Wide quarterly financial table.
Bridge from operating loss to adjusted operating income for FY2023 to FY2025.
Reconciliation table.
Quarterly bridge to adjusted operating income, turning positive in Q2'26.
Reconciliation table.
Quarterly reconciliation of total operating expenses to adjusted non-compute cost.
Reconciliation table.
Australian retail offer disclaimer.
Dense legal text.
European Economic Area retail offer disclaimer.
Dense legal text.
Spanish-language EEA disclaimer.
Dense legal text.
Swiss FinSA retail disclaimer.
Dense legal text.
Common questions about this slide and the underlying presentation content.
No. It is a reconstruction by SlideDeck Co based on the retail roadshow deck, and every slide is marked 'not an Anthropic file'. Use it as a design and structure reference, and check figures against Anthropic's SEC filings.
Offering summary and mission, five core advantages, enterprise adoption of Claude, the financial model (growth, margins, compute efficiency, long-term framework) and an appendix of non-GAAP reconciliations and regional disclaimers. 57 slides in total.
Yes. An editable PPTX is included alongside the PDF, so you can change text, tables and colors in PowerPoint, Keynote or Google Slides.
Yes. The arc of mission, moat, traction, unit economics and long-term targets maps directly onto late-stage fundraising and IPO decks. Replace the data with your own and keep the layouts.
A minimal editorial look: warm cream backgrounds, bold sans-serif headlines with serif accents, flat sage, terracotta and slate-blue panels, and thin-rule tables.
Exponential line curves, quarterly and stacked bar charts, a benchmark line chart, small-multiple sector charts, KPI tiles, flywheel and platform diagrams, plus full financial statement and reconciliation tables.
Yes. Use 2Slides to create an investor deck in this style from your own notes or documents, then refine the result.
Create Your Own Slides
Turn your ideas into professional presentations in seconds with 2slides AI.
Reference professional designs, choose your style, and generate slides with perfect text rendering. Powered by Nano Banana—start creating your presentation now.
Your AI Agent for slides. Save time, shine faster with intelligent presentation creation.
All services online© 2026 2slides. All rights reserved.