FRANK

AI-powered platform for insurance brokers to automate administrative tasks and enhance client advisory.

Cover Block

From the public record

Field	Value
Name	FRANK
Tagline	AI-powered platform for insurance brokers to automate administrative tasks and enhance client advisory.
Headquarters	Lisbon, Portugal [The SaaS News, October 2026]
Founded	2024 [The SaaS News, October 2026]
Stage	Seed [Jornal de Negócios, October 2026]
Business model	B2B
Industry	Insurtech
Technology	AI / Machine Learning
Geography	Western Europe
Growth profile	Venture Scale
Founding team	Co-Founders (3+): António Castro, Ricardo Vallejo, Alexandre Silva [Fintech Global, October 2026] [LinkedIn, retrieved 2026]
Funding label	Seed
Total disclosed	$3,100,000 [Jornal de Negócios, October 2026]

The Short Version

From the public record FRANK is building an AI-powered software layer for insurance brokers, and it deserves attention now because it has paired an early commercial rollout with a newly disclosed €2.9 million seed round led by Armilar, alongside Start Ventures, within roughly two years of founding [Fintech Global, October 2026] [Jornal de Negócios, October 2026]. Founded in Portugal in 2024, the company entered commercialization less than 12 months before October 2026 and, according to Fintech Forum, had already reached more than 350 insurance agencies and brokerages and over 1,500 users in Portugal, which is meaningful early distribution for a workflow product in a regulated channel [Fintech Forum, October 2026] [The SaaS News, October 2026].

The product centers on an "AI Operator" that performs administrative tasks across insurer portals and brokerage systems, including logging in, completing forms, and tracking documents, with the stated goal of shifting broker time from back-office processing toward client advisory work [Fintech Global, October 2026] [Tech Funding News, October 2026]. The clearest point of differentiation in public sources is not a frontier model claim but a workflow claim: FRANK appears to be positioning itself around execution inside fragmented insurance operating environments, with website materials also pointing to renewals, contract analysis, and integrations through API or CSV, although those broader feature claims remain primarily company-sourced [frankinsurance.ai] [VentureRadar, retrieved 2026].

The founding team reads as relevant to the problem set, if still lightly documented in public. António Castro is identified as co-founder and CEO and was previously CFO at Lovys, while coverage describes Ricardo Vallejo and Alexandre Silva as co-founders and frames the group collectively as bringing insurance, artificial intelligence, and technology experience [Tech Funding News, October 2026] [Jornal de Negócios, October 2026] [Fintech Global, October 2026].

Commercially, the business model is straightforward B2B software for brokers, and the near-term use of proceeds is equally straightforward: product development, customer acquisition, research and development, and international expansion [Jornal de Negócios, October 2026] [Fintech Global, October 2026]. Over the next 12 to 18 months, the key items to watch are whether FRANK can convert Portuguese broker penetration into durable usage, broaden beyond administrative automation without increasing implementation friction, and show that its domestic adoption can travel into additional markets rather than remaining a local workflow success [Fintech Forum, October 2026] [Fintech Global, October 2026].

Single-source, plausible -- The section relies on multiple named public sources for company, funding, and product basics, but several material operating claims, including broader feature scope and traction detail, remain single-source or company-sourced.

Taxonomy Snapshot

Axis	Value
Stage	Seed
Business Model	B2B
Industry / Vertical	Insurtech
Technology Type	AI / Machine Learning
Geography	Western Europe
Growth Profile	Venture Scale
Founding Team	Co-Founders (3+)
Funding	Seed, total disclosed approximately $3.1M [Jornal de Negócios, October 2026]

The Company in Brief

PUBLIC

FRANK presents as a Lisbon-based insurtech founded in 2024, focused on software for insurance brokers rather than carriers or end consumers [frankinsurance.ai, retrieved]. The company describes its core offer as an AI-powered platform for automating brokerage workflows, with positioning centered on administrative relief for intermediaries that still work across fragmented insurer systems [frankinsurance.ai, retrieved].

The public chronology is short but unusually compressed. FRANK says its platform supports broker tasks such as simulations, renewals, paperwork, and contract analysis, and its website states usage by more than 350 agencies in Portugal, which aligns directionally with third-party October 2026 coverage even though this section relies only on the company site [frankinsurance.ai, retrieved]. The business was founded in 2024 and is headquartered in Lisbon, Portugal, with public materials framing the company as an operating software vendor to the brokerage channel rather than a marketplace or insurer [frankinsurance.ai, retrieved].

The legal entity name is not established in the cited company materials used for this section. Within the public record available here, the clearest milestone is the company moving from founding in 2024 to a live broker-facing product footprint by the time of the retrieved website, with integrations described via API or CSV into existing ERP systems and databases [frankinsurance.ai, retrieved]. That is enough to show a company past concept stage, but not enough to map a full corporate history from filings.

Unconfirmed -- This section relies on company website materials only; no state filing or Crunchbase record was cited here.

What They Have Built

MIXED

The product claim that matters most is fairly plain: FRANK sells software to insurance brokers that takes over repetitive administrative work inside insurer portals and brokerage systems [Fintech Global, October 2026]. Public reporting describes an "AI Operator" that logs in, completes forms, and tracks documents, with the stated aim of moving broker time away from back-office handling and toward client advisory work [Fintech Global, October 2026] [Tech Funding News, October 2026]. That is a narrower and more defensible product framing than a broad "AI for insurance" pitch, because the cited workflow sits at the operational bottleneck where broker teams still spend manual time across fragmented carrier interfaces [Fintech Global, October 2026].

The company website adds a wider functional surface, although those claims are largely company-sourced and should be read that way [frankinsurance.ai]. FRANK says the platform automates retail insurance simulations, assists with paperwork including cancellations, policy changes, and payments, analyzes insurance contracts and compares guarantees, and includes a renewal-focused agent for tracking and processing policy renewals [frankinsurance.ai]. The site also says the system can integrate with ERPs and existing databases via API or CSV, while VentureRadar separately notes connectivity to email, which suggests the product is designed to sit inside existing broker workflows rather than replace core systems outright [frankinsurance.ai] [VentureRadar, retrieved 2026].

A few boundaries are worth keeping visible because the public evidence is still thin. The retrieved sources support workflow automation across portals and administrative tasks, but they do not establish the underlying model architecture, accuracy rates, insurer-by-insurer coverage, or whether the product operates through direct integrations, browser automation, or a mix of both [Fintech Global, October 2026] [frankinsurance.ai]. The website also references 24/7 help for client needs such as accidents, but there is not enough independent reporting to determine whether that is a live conversational support layer, a claims intake workflow, or a narrower assistance feature [frankinsurance.ai].

Unconfirmed -- Product scope relies materially on company website claims, with partial corroboration from Fintech Global, Tech Funding News, and VentureRadar.

Market Size and Demand

PUBLIC The market matters because FRANK is selling into a part of financial services where administrative work is still dense, repetitive, and spread across fragmented insurer systems, but the public record here is thinner than the company’s early traction might suggest [Fintech Global, October 2026] [Fintech Forum, October 2026]. That leaves the market case resting less on formal sizing studies and more on the practical shape of the workflow FRANK says it is addressing: brokers logging into portals, completing forms, tracking documents, handling renewals, and processing client service requests across multiple carriers [Fintech Global, October 2026] [frankinsurance.ai].

No named third-party TAM, SAM, or SOM study was captured in the source set, so the market has to be framed through adjacent public evidence rather than a direct market-sizing citation. On the evidence available, FRANK’s initial serviceable market is insurance agencies and brokerages in Portugal, where the company says it has already reached more than 350 agencies and brokerages and over 1,500 users, after commercializing the platform less than 12 months before October 2026 [Fintech Forum, October 2026]. That does not establish overall market share, because the total broker population was not provided in the retrieved sources, but it does suggest the company is targeting a workflow category that can be adopted at agency level without waiting for insurer core-system replacement [Fintech Forum, October 2026] [Fintech Global, October 2026].

The main demand driver is labor substitution in broker back offices. Public coverage consistently describes FRANK’s product around administrative execution rather than lead generation or consumer distribution: the AI Operator performs work across insurer portals and brokerage systems, including logging in, completing forms, and tracking documents, while the company website adds simulations, renewals, policy changes, payments, and contract comparison as target tasks [Fintech Global, October 2026] [Tech Funding News, October 2026] [frankinsurance.ai]. That matters because these are functions brokers already pay people to do, which usually makes the budget conversation more concrete than broader "AI transformation" pitches.

A second tailwind is the persistence of fragmented operating environments in insurance distribution. FRANK says it integrates with ERPs and existing databases via API or CSV, and VentureRadar describes the system as connecting to email, which points to a product posture built for heterogeneous software stacks rather than greenfield deployments [frankinsurance.ai] [VentureRadar, retrieved 2026]. If that positioning is accurate, the adjacent markets are not only broker software, but also horizontal workflow automation, document handling, and communications tooling used inside small and midsize agencies.

Regulation is relevant here even though the retrieved sources do not cite specific statutes. Insurance intermediation tends to generate documentation, audit trails, policy comparisons, renewals, and claims-related communications by design, which is one reason administrative software can earn attention in the category. Public evidence does not yet show whether FRANK’s tooling changes regulated advice workflows, remains an operational layer beneath human brokers, or expands into customer-facing decision support, so the regulatory reading should stay conservative until the company publishes more detail or third-party diligence surfaces it [Fintech Global, October 2026] [frankinsurance.ai].

Market lens	Public evidence	Implication for FRANK
Core entry point	Broker administrative automation across insurer portals and brokerage systems [Fintech Global, October 2026]	Targets existing operational spend rather than creating a new software budget line
Initial served segment	More than 350 agencies and brokerages, over 1,500 users, all in Portugal [Fintech Forum, October 2026]	Early concentration suggests a focused domestic wedge before broader geography expansion
Adjacent categories	ERP/database integrations via API or CSV, plus email connectivity [frankinsurance.ai] [VentureRadar, retrieved 2026]	Product can sit near workflow automation, broker operating software, and service operations tooling
Expansion vector	Funding intended in part for international expansion [Jornal de Negócios, October 2026] [Fintech Global, October 2026]	Geographic expansion is part of the thesis, but no public evidence yet establishes cross-border distribution economics

The table shows why the market case is plausible but still early. There is clear public evidence of workflow pain and some sign of local adoption, but the absence of a cited third-party market-sizing dataset means conviction on total category depth remains more directional than measured.

Single-source, plausible -- Relies on named public reporting from Fintech Global, Fintech Forum, Tech Funding News, Jornal de Negócios, VentureRadar, and the company website, but no independent third-party market-sizing report was captured for this section.

Who Else Is Fighting for This

Competitive map

MIXED FRANK appears to be positioning itself against the manual brokerage back office first, rather than against a fully disclosed set of named software rivals, which makes the competitive picture legible at the workflow level but still thin at the company-by-company level [Fintech Global, October 2026] [frankinsurance.ai] [VentureRadar, retrieved 2026].

The public record supports three competitive buckets. The first is the incumbent stack inside Portuguese insurance brokerage operations: insurer portals, brokerage systems, email, spreadsheets, and document-heavy manual processes, which FRANK says its "AI Operator" works across by logging in, completing forms, and tracking documents [Fintech Global, October 2026] [VentureRadar, retrieved 2026]. The second is the software challenger layer, namely workflow and automation products aimed at brokers, although the retrieved reporting does not name direct vendors in this category [Fintech Global, October 2026]. The third is the adjacent substitute set, which includes internal staff expansion and generic integration methods such as API or CSV-based connections, both of which can address pieces of the same operational problem without replicating an insurer-portal operator end to end [frankinsurance.ai].

That framing matters because FRANK's current wedge is unusually concrete. It is not selling abstract decision support in the retrieved coverage, it is selling task completion across insurer portals and brokerage systems for agencies that still live with fragmented carrier interfaces [Fintech Global, October 2026] [Tech Funding News, October 2026]. If that description is accurate in practice, the nearest substitute for a buyer may be a human operations team or a patchwork of existing broker software, not necessarily another young AI company with stronger brand recognition.

Edge and durability

MIXED The strongest public evidence for an edge today is early distribution inside the local brokerage channel, not a disclosed technology moat.

By October 2026, FRANK was reported to be used by more than 350 insurance agencies and brokerages and by over 1,500 users across Portugal, after beginning commercialization less than 12 months earlier [Fintech Forum, October 2026]. Even with only single-source corroboration, that is meaningful for a young vertical workflow company because it suggests the team found a repeatable entry point into a fragmented customer base rather than relying on one or two lighthouse accounts [Fintech Forum, October 2026]. The seed round, reported at €2.9 million with Armilar and Start Ventures, should also matter competitively because it gives the company capital to widen product coverage before a slower incumbent response sets in [Jornal de Negócios, October 2026] [Fintech Global, October 2026].

The durability of that edge is less settled. Workflow automation across insurer portals can strengthen with usage if the company accumulates process data, exception handling patterns, and renewal-specific know-how such as the renewal agent described on its website [frankinsurance.ai]. But the public sourcing does not establish proprietary data rights, exclusive carrier integrations, regulatory lock-in, or a channel partnership that would make displacement structurally difficult [Fintech Global, October 2026] [frankinsurance.ai]. On current evidence, the edge looks real but perishable: strong if FRANK continues to become the daily operating layer for brokers, weaker if better-capitalized software providers add similar task automation.

Exposure and likely scenario

MIXED FRANK is most exposed where software categories it does not appear to control start absorbing its wedge.

The clearest exposure is to incumbent broker management systems, carrier-side digital tooling, or horizontal automation vendors that already own system of record status or broader workflow distribution, even though the available sources do not name individual competitors [Fintech Global, October 2026] [frankinsurance.ai]. If a broker's existing software vendor can embed form completion, renewal tracking, and document handling natively, FRANK's operator risks being treated as a feature rather than a platform. The same is true at the insurer level: if carriers improve broker portal usability and standardize workflows, the pain point that FRANK is addressing could narrow from a broad operating layer to a smaller productivity add-on [Fintech Global, October 2026].

The most plausible 18-month scenario is a local land-grab in Portugal followed by a test of portability. FRANK is the likely winner if brokerages continue to prefer an overlay that works across fragmented insurer portals and if the company can convert its reported 350-plus agency footprint into deeper product attachment, especially renewals and adjacent administrative flows [Fintech Forum, October 2026] [frankinsurance.ai]. The likely loser if insurers and incumbent brokerage software vendors close the workflow gap quickly is the manual back-office model itself, because that is the direct substitute FRANK is targeting today [Fintech Global, October 2026]. A harder scenario for FRANK would be international expansion into markets where carrier connectivity, compliance detail, and local brokerage workflows differ enough that the Portugal playbook does not transfer cleanly, a risk the current public record raises but does not yet resolve [Jornal de Negócios, October 2026] [Fintech Global, October 2026].

Opportunity

Upside case

PUBLIC The prize here is not a slightly more efficient brokerage workflow, but a plausible shot at becoming the software layer that Portuguese insurance brokers rely on to operate across fragmented carrier portals, if FRANK can turn early distribution into habitual daily use and then carry that playbook into adjacent markets [Fintech Global, October 2026] [Fintech Forum, October 2026] [Jornal de Negócios, October 2026].

The headline opportunity is straightforward. Insurance distribution still contains a large volume of repetitive portal work, document handling, and renewal administration, and FRANK is positioning itself at exactly that operational bottleneck with an "AI Operator" that logs in, completes forms, and tracks documents across insurer portals and brokerage systems [Fintech Global, October 2026] [Tech Funding News, October 2026] [VentureRadar, retrieved 2026]. That matters because the company is not presenting itself as a generic assistant; it is attaching automation to a regulated, high-frequency workflow where small time savings can be visible quickly to brokers. The public traction is still early, but more than 350 agencies and brokerages and over 1,500 users in Portugal within less than 12 months of commercialization suggests the product has already moved beyond pilot status and into repeatable adoption, at least in its home market [Fintech Forum, October 2026].

A reasonable upside framework is less about headline AI and more about workflow control. If FRANK becomes the default operating layer between brokers, carrier portals, email, and internal systems, it could accumulate the kind of embedded position that is hard to displace once teams have trained staff around it and connected it to existing databases by API or CSV [frankinsurance.ai] [VentureRadar, retrieved 2026]. The October 2026 seed round, co-led by Armilar and Start Ventures, also gives the company a funded window to invest in product development, customer acquisition, and international expansion rather than proving demand from a standing start [Jornal de Negócios, October 2026] [Fintech Global, October 2026].

Scenario	What happens	Catalyst	Why it's plausible
Portugal control point	FRANK becomes the default workflow layer for independent broker operations in Portugal, expanding from administrative automation into renewals, claims-related assistance, and contract analysis	The October 2026 seed round funds product development and customer acquisition, while the installed base of 350+ agencies gives the company a reference set for expansion [Jornal de Negócios, October 2026] [Fintech Forum, October 2026]	The product already spans forms, documents, renewals, email connectivity, and policy comparison claims, which supports breadth inside one customer before geographic expansion [Fintech Global, October 2026] [frankinsurance.ai] [VentureRadar, retrieved 2026]
Iberian broker platform	FRANK uses Portugal as proof of repeatability, then enters Spain or other Southern European brokerage markets with a similar portal-fragmentation problem	International expansion is an explicit use of proceeds from the seed financing [Fintech Global, October 2026] [Jornal de Negócios, October 2026]	The company began commercialization less than 12 months before reporting 350+ agencies, which implies an ability to compress go-to-market learning in a broker-led channel if market structure is comparable [Fintech Forum, October 2026]
System of record adjacencies	FRANK moves from task automation to workflow orchestration, becoming the place where brokers manage renewals, simulations, paperwork, and insurer interactions across existing ERPs and databases	Broader integrations by API or CSV, plus deeper use of the Renewal Agent and contract comparison features, widen the product surface inside existing accounts [frankinsurance.ai]	The current feature set already points beyond a single use case, and embedded integrations tend to raise switching costs if the product becomes part of daily operations [frankinsurance.ai] [Fintech Global, October 2026]

The compounding mechanism, if it develops, is operational rather than social. Each new brokerage adds more real-world workflows, exception handling, document types, and portal interactions for the product to absorb, which should improve implementation playbooks and make the next deployment easier. Public evidence for that flywheel is still limited, but the company has already reached 350-plus agencies and more than 1,500 users in one country shortly after commercialization, and its product claims span renewals, simulations, paperwork, contract analysis, and email-linked workflows rather than a single narrow task [Fintech Forum, October 2026] [frankinsurance.ai] [VentureRadar, retrieved 2026]. If those modules are used together, FRANK could benefit from a familiar B2B software pattern: higher retention as more workflows are centralized, followed by more efficient customer acquisition through broker references and implementation know-how.

The size of the win is harder to pin down from public data because no credible market-size study or direct public comparable is provided in the source set. Even so, one can frame the upper bound qualitatively and still stay disciplined. If the strongest scenario is the Iberian or broader European broker platform, and FRANK proves it can convert early agency adoption into a sticky operating layer across multiple insurer interactions, the company could support a venture-scale outcome rather than a feature-level exit, particularly because investors are already funding it for expansion instead of local service work [Jornal de Negócios, October 2026] [Fintech Global, October 2026]. Any value statement beyond that would be a scenario, not a forecast, and public evidence here is not yet strong enough to anchor a credible number without importing outside comparables that are absent from the source record.

Single-source, plausible -- Section relies primarily on Fintech Global, Fintech Forum, Jornal de Negócios, Tech Funding News, VentureRadar, and company website claims. Core funding and traction points have partial corroboration, but several product-scope details remain company-reported or single-source.

Sources