Rogo

1.1 Identity and operating model

Rogo was founded in New York City in 2021 by Gabriel Stengel and John Willett; a separate account identifies Tumas Rackaitis as the third member of the Princeton-origin founding group. The company is a private, Series D-stage vertical AI vendor for investment banks, private-equity firms, asset managers, and related financial institutions. Its product combines firms' internal information with licensed and public financial data, then automates research, company profiles, investment memos, slide decks, and financial models. This is an enterprise workflow product rather than a general-purpose chatbot: deployment depends on data integrations, auditability, access controls, and fit with Excel, PowerPoint, Word, and institution-specific templates. The business model is described by Sacra as multi-year enterprise subscriptions with seat-based expansion and paid data or feature add-ons, but contract pricing has not been disclosed by Rogo in the reviewed evidence.[CO001, CO002, CO003, CO004]

1.2 Leadership, founder-market fit, and governance

The founding team has direct finance-workflow exposure: an independent account says Stengel, Willett, and Rackaitis met at Princeton, pursued finance careers including work at J.P. Morgan and Lazard, and expanded an econometrics-chatbot project after developer access to GPT-3. Gabriel Stengel is the named chief executive and public spokesperson, while product and engineering evidence also names functional leaders below the founder level. That combination supports founder-market fit, but it also leaves visible key-person dependence around Stengel's customer narrative and the founders' domain knowledge. Only Keith Rabois is identified as a board member in the assigned evidence; the complete board, observer rights, ownership, protective provisions, and succession depth are not publicly established here. The leadership table is therefore a partial enumeration, not a complete management or governance roster.[CO002, CO003, CO015, CO016]

1.3 Capitalization and reported scale

Rogo announced a $160 million Series D led by Kleiner Perkins on April 29, 2026 and said the financing lifted cumulative funding above $300 million. The round included Sequoia, Thrive Capital, Khosla Ventures, J.P. Morgan Growth Equity Partners, BoxGroup, Mantis VC, Jack Altman, Evantic, and Positive Sum. The company also reported more than 35,000 financial professionals using the platform at over 250 institutions. These adoption figures are repeated by independent outlets but originate with Rogo and should not be treated as audited usage. Sacra separately estimates $53 million of ARR in August 2026, up from $15 million at year-end 2025; that is an analyst estimate, not company-confirmed revenue. Publicly assigned evidence does not establish current headcount, gross margin, net revenue retention, cash balance, burn, debt, or the Series D post-money valuation, so those metrics remain null rather than being inferred from funding.[CO005, CO006, CO007, CO008, CO009, CO010, CO011, CO012]

1.4 Milestones and diligence view

The financing sequence shows unusually rapid capital formation: seed and Series A records in 2024, a Series B in April 2025, a Series C in January 2026, and the Series D three months later. The exact seed chronology is not clean: Sacra places seed funding in February 2024, while Lex reports a $7 million first seed in late 2023; Lex also reports $18.5 million for the Series A versus Sacra's $18 million. Product scope expanded in March and April 2026 through the Offset acquisition and Felix launch, which moved the narrative toward autonomous, multi-step work. Investor and company sources present strong adoption, while Lex explicitly notes unknown licensing costs and grounds its revenue discussion in assumptions. The central diligence conclusion is therefore asymmetric: customer reach and funding are visible, but unit economics, retention, governance rights, and company-confirmed revenue remain unavailable. Those omissions prevent a reliable assessment of capital efficiency or operating leverage from this chapter alone.[CO005, CO007, CO008, CO009, CO010, CO011, CO012, CO013, CO014]

1.5 Exhibits

2.1 Market boundary and substitutes

Rogo's defensible core market is institutional-finance knowledge work: recurring research, financial analysis, and production of models, reports, presentations, and transaction materials for investment banks and investment firms. The boundary excludes the underlying market-data terminals and licensed datasets, core banking systems, general-purpose productivity software, and unrelated finance functions such as close, audit, and broad FP&A. Private-markets and public-markets teams are credible adjacencies, but their workflows and willingness to pay should not be assumed identical to investment banking. Status-quo substitutes include analyst labor coordinated through spreadsheets, documents, email, and data terminals; internal AI builds; horizontal copilots; and specialist research or modeling products. Partner data access expands utility while also exposing Rogo to licensing and entitlement constraints.[CM001, CM002, CM003, CM014]

2.2 Sizing lenses without false precision

The strongest public top-down proxy is Sacra's report of a $117 billion global investment-banking fee pool in 2024, with 45% generated outside the Americas and Asia representing about 35% of historical deal value. Those figures describe banking economics and geographic activity, not software spend or Rogo's addressable revenue. Bottom-up observables are also incomplete: Sacra estimates $53 million of ARR in August 2026 and roughly $3,300 per seat annually, while Rogo's current homepage claims 60,000 users, 450 institutions, and 200,000 daily queries. April company-issued coverage cited 35,000 professionals at 250 institutions. The later figures may indicate expansion, but no common measurement date, paid-seat definition, institution definition, retention rate, or independently verified contract value supports a growth calculation. The evidence therefore supports several sizing lenses, not a single TAM.[CM004, CM005, CM006, CM007, CM008, CM010, CM011, CM012, CM013, CM016]

2.3 Buyer, user, payer, and adoption path

The economic buyer is likely an institution rather than an individual analyst because the product is sold through negotiated enterprise deployments, requires firm data and licensed-content entitlements, and has no self-serve tier. Bankers and investors are the users; operating, technology, data, security, and business-unit leaders are likely participants in purchase approval and budget ownership. Adoption should begin with a bounded, repetitive workflow whose source set and final output can be reviewed, then expand only after the institution validates accuracy, permission propagation, review burden, and realized time savings. This makes implementation and governance part of the product, not merely post-sale services. Smaller boutiques face a higher hurdle because opaque enterprise pricing and integration costs must be spread across fewer users.[CM002, CM008, CM009, CM014, CM015]

2.4 Growth drivers, constraints, and valuation relevance

Adoption is driven by expensive recurring analyst work, demand for auditable outputs, the ability to combine internal records with external financial data, and agents that can produce complete work products rather than isolated answers. The same design creates constraints: enterprise procurement is slow, private data permissions must survive into generated artifacts, third-party content licenses may limit access or redistribution, and human reviewers must preserve period, currency, unit, and source meaning. Pricing opacity also makes ROI difficult to benchmark before a pilot. For valuation, reported adoption and Sacra's ARR estimate are useful scale indicators, but the $117 billion fee pool is not a revenue multiple denominator. Underwriting requires contract-level ARR, retention, seat utilization, gross margin, data costs, and deployment economics before translating workflow breadth into a market-share or terminal-value assumption.[CM003, CM007, CM008, CM009, CM012, CM015, CM016]

2.5 Exhibits

3.1 The competitive set is broader than vertical AI peers

The buyer's alternative set begins with specialist products but does not end there. Rogo is positioned around finance-specific workflows and finished work products, while AlphaSense is described as starting from market-intelligence discovery and Hebbia from structured analysis of large document collections. S&P Capital IQ and Bloomberg bring incumbent data breadth and installed distribution; Sacra, however, characterizes their trade-offs as interface or contract complexity and reduced customization. Internal development and manual research-plus-Office workflows remain substitutes because a bank can preserve firm-specific controls without buying another application, although the assigned evidence does not quantify their cost, quality, or time-to-value. This means the relevant contest is for ownership of the complete research-to-deliverable workflow, not simply for the best chat response.[CP009, CP010, CP011]

3.2 Workflow depth and deployment support are the visible differentiation

Rogo's May 2026 release expanded Felix across PowerPoint, Excel, Word, dashboards, sourced research, email, scheduled monitoring, custom agents, memory, and additional connectors. The product therefore competes on orchestration, output fidelity, and permissioned context rather than on a proprietary foundation model alone. Email and native Excel access reduce interface switching, while firm templates, reusable agents, memory, and versioned revisions can embed the product in recurring work. Distribution is also part of the product: Rogo says former bankers and investors embed with customers, and its funding announcement directs capital toward forward-deployed teams and global expansion. Those elements can accelerate adoption, but public evidence does not separate repeatable software value from implementation intensity or prove superior accuracy against competing products.[CP002, CP005, CP006, CP007, CP008, CP015]

3.3 Commercial opacity and data rights complicate a clean price comparison

Rogo's reviewed public sales route did not publish a numerical tariff, billing period, seat minimum, or usage allowance. Sacra estimates approximately $3,300 per seat annually and describes multi-year enterprise contracts, but this is an analyst estimate rather than a company price list or realized contract evidence. Buyers must also separate the software subscription from licensed financial data and deployment work. Provider logos do not establish universal entitlements, and permissions may constrain what generated materials can contain or redistribute. Consequently, the defensible comparison is a proposal-level total-cost and workflow test: software, data rights, implementation, review burden, and export permissions should be measured together. Peer pricing and discount structures are not disclosed in the assigned evidence.[CP012, CP013, CP014]

3.4 The moat resides in configuration and distribution, but incumbents and upstream models remain threats

The most plausible durable layer is Rogo's harness: finance-specific evaluations, connectors, audit controls, house-style output, reusable agents, and embedded deployment knowledge that can persist when the underlying model changes. Firm-specific templates, memory, permissions, and recurring agents can create operational switching costs because replacement requires reconstructing context and governance, not merely changing a chat interface. That moat is not proven independently. Incumbents can bundle AI into existing data estates, specialist rivals can converge on document-to-deliverable workflows, and large institutions can build internally. Sacra also identifies dependence on third-party model providers as a risk if pricing, access, or performance changes. No assigned source provides a controlled benchmark of accuracy, implementation cost, retention, or multi-homing, so differentiation should be treated as promising but not yet demonstrated as durable.[CP005, CP008, CP009, CP010, CP011, CP014]

3.5 Exhibits

4.1 Revenue model and pricing

The clearest public description is a B2B SaaS model sold to financial institutions through multi-year enterprise contracts. Sacra describes a roughly $3,300 annual per-seat price and additional monetization from specialized data access and paid features; these are third-party estimates, not a company price sheet or evidence of realized contract value. The model therefore has recurring and expansion-revenue characteristics, but the retained evidence does not disclose discounting, minimum commitments, implementation fees, revenue recognition, or the mix among seats, data, and features. Rogo's platform breadth and guided implementation may strengthen adoption, yet there is no public basis for treating implementation work as a separately priced revenue stream. The revenue bridge below is a mechanism map, not a quantified gross-profit bridge.[CI001, CI002, CI007, CI015]

4.2 Traction and sales-efficiency signals

Public traction is meaningful but unevenly verified. Rogo says more than 35,000 professionals at over 250 institutions use the platform, a claim repeated by independent reporting but still originating with the company. Sacra estimates $53 million of ARR in August 2026 versus $15 million at the end of 2025, while AI Wiki reports that the company reached seven-figure ARR within five months using one sales representative. These datapoints suggest rapid adoption and potentially efficient early selling, but they do not substitute for audited revenue, bookings, cohort retention, sales-and-marketing spend, or CAC payback. User counts also cannot be converted into revenue by multiplying them by the reported seat price because deployment terms, active-seat definitions, discounts, and bundled access are unknown.[CI003, CI004, CI005, CI006, CI007]

4.3 Cost structure and unit economics

The visible cost structure combines cloud infrastructure, third-party model usage, licensed financial data, product development, and forward-deployed implementation teams. Sacra specifically identifies cloud and data-licensing costs and flags dependence on OpenAI and Google as exposure to price or access changes. Those drivers make gross margin sensitive to workload intensity, model routing, data entitlements, and the amount of human support bundled into deployments. No retained source quantifies gross margin, inference cost per workflow, data cost per customer, CAC, payback, NRR, logo churn, or implementation utilization. Consequently, the recurring subscription structure is visible, but neither contribution margin nor operating leverage is. The appropriate diligence focus is contract-level revenue and cost reconciliation rather than a speculative SaaS benchmark.[CI008, CI009, CI014]

4.4 Capital adequacy and financial verdict

Rogo announced a $160 million Series D in April 2026 and said the round brought total funding above $300 million. Sacra's page instead states $165 million of total funding, an internally stale-looking figure that conflicts with the later financing and should not be silently reconciled. The company says proceeds will fund international expansion, institutional partnerships, forward-deployed teams, and Felix. Funding raised is not cash available: no retained source provides cash on hand, monthly burn, debt, runway, or a next-round trigger. The financial verdict is therefore balanced: disclosed adoption and financing reduce near-term capital-access concern, but revenue quality, margin path, retention, and capital efficiency remain unobservable. Investor underwriting should remain conditional on management accounts, customer cohorts, contract economics, and a current cash-to-runway bridge.[CI010, CI011, CI012, CI013, CI016]

4.5 Exhibits

5.1 Product surface and user workflow

Rogo is positioned as a workflow system for investment bankers, private-equity investors, hedge funds, and other financial professionals rather than as a general-purpose chatbot. Public reporting describes a chat-first workspace, reusable Quick Actions, a spreadsheet agent, and Felix, an email-accessible agent intended to carry work from research through deliverable creation. The supported jobs include company research, investment memos, financial-model construction and updating, presentation drafting, deal screening, CIM generation, buyer outreach, and data-room diligence. This breadth is strategically important because the product can sit across multiple steps of a deal team's existing process, but most detailed capability descriptions ultimately originate with Rogo or secondary publications repeating its claims. The available material does not establish independently measured accuracy, completion time, or error rates for these workflows.[CE001, CE002, CE003, CE004, CE006, CE008]

5.2 Architecture, grounding, and integration

The disclosed operating model has four observable layers: user interfaces, an agent or workflow-orchestration layer, finance-oriented language-model processing, and data grounding across institutional systems and licensed financial sources. SiliconANGLE reports that custom-trained large language models use CRM and other internal-system data and can ingest FactSet, while Sacra describes connections to deal-room documents, Excel, PowerPoint, Word, data warehouses, LSEG, PitchBook, Quartr, and international filings. Outputs are designed to return to the tools bankers already use, which can improve adoption and auditability. This architecture also creates dependencies on data entitlements, integration quality, model providers, and the correctness of retrieved source material. Public evidence does not disclose model evaluation methodology, isolation design, retrieval accuracy, or contractual resilience of the external data feeds.[CE002, CE005, CE006, CE013]

5.3 Maturity, deployment, and support model

Reported adoption is consistent with a product beyond an early pilot stage, although the exact denominator changes across sources. Forbes reported more than 25,000 daily users at 150 firms in April 2026, while Rogo's April 29 announcement said more than 35,000 professionals at over 250 institutions used the platform in daily workflows. The difference may reflect rapid growth, reporting cutoffs, or differing definitions of firms and institutions; no audited cohort reconciles the two figures. Rogo also says it uses forward-deployed engineers and bankers to help customers adopt the system. That service model can reduce implementation friction in complex banks, but it may make deployment quality dependent on scarce expert labor. No reviewed source provides implementation time, support response targets, uptime history, retention, or a dated product roadmap.[CE009, CE010, CE011, CE012, CE014]

5.4 Differentiation, controls, and technical risk

Rogo's defensibility appears to come from finance-specific workflow design, integration with proprietary institutional context, licensed market data, and the ability to write outputs into familiar office and deal tools. Those are potentially stronger switching costs than model access alone, particularly where audit trails and permissions matter. SiliconANGLE reports that the platform includes audit trails, access controls, and a Rogo-developed tool called Sisyphus that scans infrastructure daily for vulnerabilities. These are relevant controls, but they remain company claims relayed by a publication; the assigned evidence contains no independent certification record, penetration-test result, incident history, service-level disclosure, or model-risk assessment. It also contains no public repository, package, API forum, engineering talk, or practitioner-community signal from which to assess developer maturity. The technical thesis is therefore credible at the product-workflow level but not independently validated at the security, reliability, or architecture level.[CE005, CE007, CE013]

5.5 Exhibits

6.1 Institutional footprint and customer segments

Rogo's disclosed footprint is enterprise-first: the company reported more than 35,000 financial professionals and more than 250 institutions in April 2026. The named set is concentrated in global investment banking and advisory, while independent workflow analysis also identifies private- market and public-market investment teams as intended users. Buyers are therefore likely to include technology, operations, compliance, and business leadership, while day-to-day users include bankers, investors, and analysts. The platform is positioned across origination, execution, advisory, and portfolio intelligence rather than as a single research utility. That breadth expands potential seat and workflow penetration, but neither the user count nor institution count discloses geography, paid seats, active-user methodology, or revenue contribution by segment.[CU001, CU002, CU003, CU004, CU006]

6.2 Adoption scale and named customer proof

The strongest public adoption evidence is the company's current scale disclosure, repeated by independent publishers, plus named institutions including Rothschild & Co, Jefferies, Lazard, Moelis, and Nomura. A separate platform testimonial attributes a successful integration, productivity improvement, lower risk, and added banker capacity to Truist Securities chief executive Tom Hackett. These references are meaningful enterprise proof, but they do not establish comparable deployment scope: public materials do not state paid seats, contract value, rollout dates, renewal status, or measured outcomes for each named institution. Felix use cases such as deal screening, CIM generation, buyer outreach, and data room diligence indicate production-oriented workflows, yet most named customer evidence remains vendor-published rather than customer-issued.[CU001, CU002, CU003, CU005, CU011]

6.3 Commercial durability and expansion motion

Sacra characterizes Rogo as an enterprise subscription sold through multi-year contracts and estimates pricing at about $3,300 per seat per year, with additional revenue from specialized data access and paid features. Sacra also estimates $53 million of ARR in August 2026 versus $15 million at year-end 2025; these are analyst estimates, not company- confirmed revenue or cash metrics. The reported model implies expansion through seat growth and additional workflows, while Rogo's broad workflow catalog creates multiple adoption surfaces within one institution. However, Felix has a separate request-access route, and financial-data entitlements may require separate commercial rights. Expansion therefore depends on product access, implementation, permission mapping, data licensing, and sustained human review rather than on seat availability alone.[CU007, CU008, CU009, CU010, CU011, CU012, CU014]

6.4 Retention, concentration, and procurement risk

None of the eight reviewed sources discloses NRR, GRR, churn, renewal rate, contract renewal outcomes, or top-customer revenue share. The public customer count therefore cannot establish durability, and the five named banks do not reveal whether revenue is diversified or concentrated. Public pricing is also incomplete: an independent review found no numerical tariff on the sales route, while Sacra's per-seat figure remains an estimate. Adoption carries review friction because financial work must preserve source periods, units, permissions, and human approval; one independent source explicitly warns that hallucination or data inaccuracy has an extreme cost in finance. Underwriting should require cohort retention, institution-level ARR bands, paid-versus-active seats, expansion by vintage, and references issued directly by customers.[CU003, CU008, CU013, CU014, CU015, CU016]

7.1 Ranked residual risk posture

The highest residual exposure is cross-deal confidentiality and entitlement enforcement. Rogo says existing entitlements carry through to agents, while independent reviews emphasize that a deployment must prove that inaccessible source material cannot leak through generated tables or reusable workflows. The second cluster is output integrity: Felix is described as autonomously executing multi-step deal workflows, so a wrong assumption can travel farther than in a single-answer research tool before a human catches it. Third-party model and financial-data dependencies add both availability and gross-margin sensitivity. The heatmap therefore ranks residual risk from evidence strength and workflow consequence, not from an asserted incident history; no reviewed source establishes incident frequency or loss rates.[CR001, CR006, CR007, CR009, CR011, CR016]

7.2 Regulatory, legal, privacy, and assurance risk

Rogo publicly lists SOC 2 Type II, ISO 27001, ISO 42001, GDPR, CCPA, and EU AI Act credentials or alignment. Those are meaningful mitigation signals, but the assigned evidence does not include certificates, auditor reports, control scope, test periods, legal opinions, or jurisdiction-specific deployment assessments. Vantaige explicitly recommends obtaining current attestation letters and scope, while Sequenced states that it did not inspect private assurance reports or verify a particular deployment. The investment implication is a conditions-precedent approach: verify certificate issuers and scope, map data residency and retention to each customer entity, and test permission inheritance before treating the compliance posture as mature. Public-source silence on litigation, enforcement, or incidents remains an evidence gap, not evidence that none exist.[CR001, CR002, CR003, CR004, CR007]

7.3 Operational, quality, security, and dependency risk

The mitigation architecture is directionally appropriate: SiliconANGLE reports audit trails, access controls, and daily infrastructure vulnerability scanning, and Rogo says source permissions carry into agents. However, these controls are predominantly vendor-described and do not quantify false outputs, review effort, availability, recovery objectives, or isolation failures. Operational diligence should use adversarial production-like tests: conflicting source documents, broken spreadsheet links, stale market data, and two segregated mock deal rooms. Dependency risk is also structural. External model providers, licensed financial-data feeds, and firm-specific implementation support are integral to the offering, so adverse pricing, access, or service changes can affect quality, availability, or economics even if Rogo's own application layer performs as designed.[CR001, CR005, CR006, CR007, CR008, CR009, CR010, CR011]

7.4 Financial exposure, monitoring indicators, and thesis-break triggers

Rogo announced a $160 million Series D and more than $300 million of total funding, while FinTech Futures reported that the round followed a $75 million Series C by roughly four months. This supports substantial financing access, but funding raised is not cash available and does not disclose burn, runway, gross margin, or working-capital needs. Sacra's $53 million August 2026 ARR figure is explicitly an estimate, and Vantaige reports that pricing is not public. Underwriting should therefore monitor independently reconciled ARR, gross retention, model and data costs as a share of revenue, production error rates, severity-one security events, and implementation time. Thesis-break triggers are a material permission-isolation failure, inability to reproduce cited financial outputs, loss of critical data rights, or financing needs that arrive before durable unit economics are demonstrated.[CR012, CR013, CR014, CR015]

7.5 Exhibits

8.1 Investment call and decision logic

The recommendation is research-more with medium confidence, high underwriting risk, and an indeterminate valuation stance. Rogo announced a $160 million Series D and said total funding exceeded $300 million, while the company reported more than 35,000 users across over 250 institutions. Those signals support product demand and financing capacity, but funding raised is not cash available and reported adoption is not revenue quality. The retained evidence provides no Series D price, preference stack, gross margin, net retention, cash balance, or burn rate. Without both entry price and verified recurring economics, a target return, hold period, or exit value would be false precision rather than underwriting.[CV001, CV002, CV007, CV014, CV015, CV016]

8.2 Financing context and operating anchors

The strongest historical price anchor is the January 2026 Series C: Forbes and Sacra report a $75 million round at a $750 million post-money valuation. Forbes separately reports that revenue rose from $2 million in 2024 to more than $15 million in 2025, implying that the Series C was priced at less than 50 times the disclosed prior-year revenue. Sacra later estimates $53 million of ARR in August 2026 and approximately $3,300 per seat annually, but ARR is not accounting revenue and the estimate is not company-confirmed. These figures indicate rapid expansion while leaving revenue recognition, contract value, gross margin, retention, and the Series D valuation unresolved.[CV003, CV004, CV005, CV006, CV009, CV014]

8.3 Thesis, anti-thesis, and scenarios

The thesis is that domain-specific workflows, financial-data integrations, and reported institutional usage can support a durable vertical software franchise. Rogo says Felix can automate multi-step work from screening through diligence, and its team emphasizes experience across banking, investing, and AI. The anti-thesis is equally material: Sacra identifies reliance on third-party model providers and tightening AI governance, while the public evidence does not establish durable margins, retention, or pricing power. The bull, base, and bear cases therefore use observable proof points and diligence triggers rather than invented valuation ranges. A higher price is defensible only after the estimated ARR and contract economics are verified.[CV007, CV009, CV010, CV011, CV012, CV013, CV014]

8.4 Exit readiness, diligence asks, and thesis breaks

Public evidence is insufficient to call Rogo exit-ready or to select a defensible public-company, M&A, or private-round multiple. The Series C is a historical transaction, not a current fair-value opinion; Sacra's ARR estimate is an analytical input, not audited performance; and competitor valuations are not independent benchmarks for Rogo. Before an investment decision, diligence should obtain the Series D cap table and preference waterfall, audited or board-approved ARR and revenue bridges, gross margin, cohort retention, cash and burn, customer concentration, contract duration, and renewal data. Thesis breaks include materially lower verified ARR, weak retention, uneconomic data or model costs, regulatory limits on core workflows, or an entry price that cannot produce the required return under a conservative exit case.[CV003, CV005, CV009, CV012, CV013, CV014, CV015]

8.5 Exhibits