## A Deep Analysis of Lithuanian Fintech, SaaS, DevOps, Cloud and Engineering Organizations **Why Revenue per Employee Is No Longer Enough** Russian version : [[Переосмысление эффективности IT-компаний в эпоху AI]] ## 1. Executive Summary Lithuania punches well above its weight in the global technology landscape. A country of fewer than three million people is home to 280+ licensed fintech companies, the Baltic region's first confirmed unicorn (Vinted, valued at EUR 4.5B+), one of the world's most successful cybersecurity ecosystems (Nord Security, Surfshark, Tesonet), a global web hosting leader (Hostinger), and a sophisticated banking technology infrastructure serving Nordic financial groups. Yet when analysts and investors reach for an efficiency metric, they almost always land on the same blunt instrument: **revenue per employee**. > _"A 10-engineer fintech team with full GitOps automation and AI-assisted development may outperform a 200-engineer shared-service banking center on every metric that actually matters to the business."_ This report argues that revenue per employee — while not useless — is structurally insufficient for evaluating the efficiency of modern AI-era engineering organizations. It systematically misrepresents companies operating under heavy regulatory burden, shared-service accounting models, and high infrastructure complexity. It completely ignores automation depth, AI augmentation, platform engineering maturity, and organizational scalability — the factors that will actually determine competitive outcomes over the next five years. **Key findings:** - Lithuanian IT companies span a maturity spectrum from world-class (Vinted, Hostinger, Nord Security) to organizations facing structural headwinds (ATM software, ICT conglomerates, legacy outsourcers) - The cybersecurity cluster around Tesonet/Nord Security/Surfshark/Oxylabs represents Lithuania's most significant and underappreciated strategic asset - Shared-service revenue metrics for Danske Bank Lithuania are structurally meaningless for efficiency comparison — yet this organization may run some of the most sophisticated engineering governance in the country - Walletto's reported ~€520K revenue per employee is the highest in the dataset and reflects the exceptional efficiency of a lean, automation-first EMI fintech - AI adoption and platform engineering maturity will be the decisive competitive differentiators by 2028, making current headcount-based analysis increasingly obsolete --- ## 2. Lithuanian IT & Fintech Market — The Big Picture ### From Outsourcing Country to Engineering Economy A decade ago, Lithuania was primarily viewed through the lens of IT outsourcing: competent engineers, lower costs than Western Europe, reasonable English proficiency. That framing is now dangerously outdated. Lithuania has executed a systematic transition from outsourcing destination to **engineering economy** — a country that builds globally competitive technology products, owns intellectual property, and exports software at scale. This transition was enabled by several converging forces: **Regulatory infrastructure:** The Bank of Lithuania created one of Europe's most accessible fintech licensing frameworks, issuing more EMI (Electronic Money Institution) licenses per capita than any other EU member state. The VITTA regulatory sandbox accelerated product testing. The result: 280+ licensed fintech operators using Lithuania as their EU regulatory home. **Talent density:** Vilnius University, Kaunas University of Technology, and a network of coding schools produce thousands of engineering graduates annually. The concentration of technology companies in a small geography creates talent network effects — engineers move between companies, carry knowledge, and elevate sector-wide maturity. **Infrastructure investment:** Data centres, high-bandwidth connectivity, and significant cloud infrastructure investment (AWS, Google Cloud, Azure all have Lithuanian presence) have created the technical foundation for product companies to operate at global scale from Vilnius. **Capital formation:** A combination of Nordic venture capital, Lithuanian private equity (Practica Capital, Contrarian Ventures), and international growth funds has provided sufficient capital for companies like Vinted, Nord Security, and Kilo Health to scale globally without relocating their engineering core. ### The Regulatory Overlay: NIS2 and DORA Two regulatory frameworks are reshaping the operational cost structure of Lithuanian IT and fintech companies in 2024-2025: **NIS2 Directive** (effective October 2024) introduces mandatory cybersecurity governance requirements for organizations classified as essential or important entities. For Lithuanian technology companies serving critical sectors — financial services, digital infrastructure, cloud computing — NIS2 compliance is not optional. It requires executive accountability for cybersecurity risk, incident reporting within 24/72 hours, and supply chain security assessments. For organizations without mature security automation, this is a significant new cost center. **DORA (Digital Operational Resilience Act)** applies specifically to financial sector entities and their ICT suppliers. It mandates ICT risk management frameworks, business continuity testing, operational resilience testing (including mandatory threat-led penetration testing for significant institutions), and stringent third-party ICT risk management. For Walletto, ConnectPay, Paysera, Kevin., and Danske Bank Lithuania, DORA compliance is a new and material operational requirement. Organizations that automate compliance — policy-as-code, continuous compliance monitoring, automated audit evidence generation — will convert these obligations into competitive moats. Those treating compliance as periodic manual effort will face escalating costs. --- ## 3. The Dataset: What We Actually Know Before analyzing efficiency, we need to be honest about data quality. Most Lithuanian technology companies are private — they are not required to publish detailed financial statements. The dataset below combines confirmed public figures with well-sourced estimates from Lithuanian business registries, Invest Lithuania research, sector analyst reports, and ecosystem benchmarks. ### Core Financial Dataset |Company|Sector|Revenue|Employees|Rev/Employee|Data Quality| |---|---|---|---|---|---| |**Danske Bank Lithuania**|Banking / GBS / Engineering|~€248M|~4,700|~€52K|High (reported)| |**Walletto**|Fintech / EMI|~€38M|~73|**~€520K**|High| |**Paysera**|Fintech / Payments|~€100–120M|~700|~€150K|Medium| |**ConnectPay**|Fintech / EMI|~€15–25M|~120|~€150–200K|Medium| |**Kevin.**|Open Banking / A2A|~€20–40M|~250|~€80–160K|Medium| |**InnoForce**|Enterprise / SAP|~€14.8M|~33|**~€449K**|High| |**NFQ Technologies**|Engineering|~€40–60M|~800|~€50–75K|Medium| |**Devbridge**|Software Engineering|~€30–50M|~600|~€50–80K|Medium| |**Tesonet**|Venture Builder / Tech Group|~€200–400M|~3,500+|~€60–115K|Medium| |**Nord Security**|Cybersecurity / SaaS|~€300–500M|~1,800–2,200|~€140–270K|Medium| |**Surfshark**|Cybersecurity / VPN|~€100–250M|~400–700|~€150–400K|Medium| |**Hostinger**|Hosting / Cloud|~€110–150M|~900|~€120–170K|Medium| |**Oxylabs**|Web Intelligence / Proxies|~€80–150M|~500+|~€160–300K|Medium| |**Kilo Health**|Digital Health / SaaS|~€150–250M|~600|~€250–400K|Medium| |**Vinted**|Marketplace / Platform|~€600M+|~1,800+|**~€330K**|High (reported)| |**Penki Kontinentai**|Banking Tech / Infrastructure|~€60–70M|~500–700|~€90–140K|Medium| |**BS/2**|ATM / Banking Systems|~€39M|~250–350|~€110–156K|High| _* Estimated values marked with asterisk in original dataset. Estimates based on Lithuanian business registry data, Invest Lithuania GBS Report 2024, Lithuania Tech Ecosystem Report 2024, and industry benchmarks._ ### Immediate Observations From Raw Data Before any analytical framework is applied, three patterns emerge immediately: **The outliers are structurally explained:** Walletto's €520K and InnoForce's €449K revenue-per-employee are not simply "more efficient" than Danske Bank's €52K. They reflect fundamentally different business models — lean B2B automation plays vs. a massive enterprise governance and compliance operation. **The range is enormous:** The dataset spans from €52K (Danske Bank) to €520K (Walletto) — a 10x range. This variation is driven almost entirely by business model type, not by operational efficiency in any meaningful sense. **Estimation uncertainty is significant:** Most figures carry ±30-50% uncertainty bands. Any analysis treating these as precise numbers will produce false precision. The value of this dataset is directional, not arithmetical. --- ## 4. Why Revenue per Employee Is a Broken Metric ### The Structural Problems Revenue per employee seems like a clean, objective, comparable metric. It is not. It carries at least five structural distortions that make cross-company comparison unreliable: #### Distortion 1: Shared-Service Accounting Danske Bank Lithuania employs approximately 4,700 technology professionals. Its "revenue" of ~€248M is largely **intra-group transfer pricing** — a charge from the Lithuanian entity to the Danish parent for services delivered. This is not market revenue. It is an accounting construct designed to allocate costs within a corporate group, calibrated to tax and regulatory considerations, not to market rates. Comparing Danske Bank's revenue-per-employee to Vinted's is comparing apples to accounting entries. Yet this organization — with its 4,700 engineers running enterprise platform services, SDLC governance, compliance automation, and cloud transformation programs — may represent the highest concentration of sophisticated engineering management in Lithuania. Revenue per employee captures none of this. #### Distortion 2: Regulatory Overhead Invisibility A licensed EMI like Walletto or ConnectPay must maintain: a compliance officer, an MLRO (Money Laundering Reporting Officer), legal counsel, a risk management function, an internal audit capability, and increasingly a DORA-compliant ICT risk management framework. These roles generate zero direct revenue. They exist to preserve the operating license. Walletto's €520K revenue per employee looks extraordinary — and it is impressive. But it partly reflects a business model where B2B automation and banking-as-a-service infrastructure serve many clients per employee. The compliance overhead is real and material; it simply hasn't yet scaled headcount to a level where it distorts the ratio. At larger fintech scale (Paysera, 700 employees), the compliance, operations, and customer support functions have grown proportionally, pulling revenue-per-employee down to ~€150K — not because engineering efficiency has declined, but because regulatory operations require staffing. #### Distortion 3: Infrastructure Complexity Is Invisible Consider three companies with similar revenue-per-employee metrics: - **BS/2** (~€110-156K): Delivers ATM driving software + physical hardware logistics + on-site field service operations across 70+ countries - **NFQ Technologies** (~€50-75K): Delivers software engineering services to European enterprise clients from Vilnius - **Hostinger** (~€120-170K): Operates proprietary global data centres, a CDN, and cloud infrastructure for 3M+ customers across 150+ countries These look comparable on revenue-per-employee. They are operating entirely different categories of infrastructure complexity. Hostinger's infrastructure density and automation leverage is orders of magnitude greater. BS/2's physical supply chain and field operations add operational dimensions that pure software metrics ignore. #### Distortion 4: AI and Automation Are Invisible A 500-person engineering organization with full AI coding assistant adoption, GitOps automation, and a mature Internal Developer Platform (IDP) may produce equivalent software output to a 1,000-person organization without these capabilities. Revenue-per-employee captures this as identical — both produce the same revenue per head — while in reality the first organization has built a structural productivity advantage that compounds annually. As AI adoption accelerates from 2025 to 2030, this blind spot will become the dominant failure mode of headcount-based efficiency analysis. #### Distortion 5: Business Model Stage Matters Kevin. is an open banking infrastructure play with ~€20-40M revenue and ~250 employees. Revenue per employee looks modest at €80-160K. But Kevin. is not trying to maximize current revenue per head — it is building network effects and regulatory relationships that, if successful, will generate disproportionate revenue per employee in three to five years. Evaluating a pre-network-effects infrastructure company on current revenue-per-employee is like evaluating Stripe in 2012. ### What Should Replace It? Revenue per employee should be one input into a multi-dimensional framework that also measures: - **Automation Density:** Revenue generated per DevOps/SRE/platform engineer - **Infrastructure Complexity Index:** Revenue per unit of infrastructure complexity managed - **AI Leverage Ratio:** Estimated productivity multiplier from AI tool adoption - **Platform Engineering Maturity:** Force multiplier effect of internal developer platform - **Regulatory Efficiency:** Revenue relative to compliance headcount burden - **Scalability Gradient:** Rate at which revenue can grow without proportional headcount growth Section 10 introduces the full EAOEI framework that operationalizes these dimensions. --- ## 5. Engineering & DevOps Maturity Framework ### Evaluation Methodology Each organization is assessed across six engineering maturity dimensions, each scored 1-10. Scores reflect publicly observable signals: job postings (technology stack requirements), engineering blog content, conference presentations, GitHub/open-source activity, infrastructure technology indicators (job ads mentioning Kubernetes, ArgoCD, Terraform, Prometheus, etc.), and known architectural patterns from public sources. ### The Six Dimensions #### 1. Infrastructure Maturity (1-10) Measures the sophistication of compute, networking, and storage infrastructure operations. |Score|Characteristic| |---|---| |1-3|VM-based, manual provisioning, no containers| |4-6|Kubernetes adopted, CI/CD pipelines, some automation| |7-8|GitOps (ArgoCD/Flux), multi-cluster, service mesh, full IaC| |9-10|eBPF-native observability, cell-based architecture, zero-config infra| #### 2. DevOps & CI/CD Maturity (1-10) Measures deployment pipeline sophistication, DORA metrics performance, and release engineering maturity. |Score|DORA Performance Profile| |---|---| |1-3|Monthly/quarterly releases, high change failure rate| |4-6|Weekly releases, basic CI/CD, moderate CFR| |7-8|Daily deployments, progressive delivery, <5% CFR| |9-10|Multiple deploys/day, canary by default, <1% CFR, <1hr MTTR| #### 3. Platform Engineering Maturity (1-10) Measures Internal Developer Platform sophistication and developer experience investment. |Score|Platform Capability| |---|---| |1-3|No IDP, ops tickets for everything, no self-service| |4-6|Basic self-service tools, partial golden paths| |7-8|Full IDP (Backstage or equivalent), golden paths, DevEx metrics| |9-10|AI-integrated IDP, fully autonomous deployment, dynamic scoring| #### 4. AI Readiness (1-10) Measures adoption of AI-assisted development, AIOps, and AI product integration. |Score|AI Adoption State| |---|---| |1-3|No AI tooling, manual operations| |4-6|Pilot AI coding assistants, basic ML product features| |7-8|Full Copilot/Cursor adoption, AI-driven observability| |9-10|AI-native ops, autonomous incident response, AI product core| #### 5. Cloud-Native Architecture (1-10) #### 6. Security & Compliance Automation (1-10) Measures SBOM generation, policy-as-code, supply chain security, and automated compliance evidence. ### DORA Metrics Cross-Reference Where signals are available, companies are cross-referenced against DORA (DevOps Research and Assessment) performance tiers: |DORA Tier|Deploy Freq|Lead Time|Change Failure Rate|Recovery Time| |---|---|---|---|---| |**Elite**|Multiple/day|<1 hour|0–5%|<1 hour| |**High**|Daily/weekly|1 day–1 week|5–10%|<1 day| |**Medium**|Weekly/monthly|1–6 months|10–15%|1 day–1 week| |**Low**|Less than monthly|>6 months|>15%|>1 week| --- ## 6. Company-by-Company Deep Analysis --- ### 🏦 Danske Bank Lithuania **Sector:** Banking / Global Business Services / Engineering **Revenue:** ~€248M (intra-group) | **Employees:** ~4,700 | **Rev/Employee:** ~€52K **What This Company Actually Is** Danske Bank Lithuania is not a bank. It is a **technology and operations delivery organization** employing approximately 4,700 professionals to build and run the digital infrastructure, data analytics, and platform engineering capabilities of Danske Bank Group — headquartered in Copenhagen. The €52K revenue-per-employee figure reflects intra-group transfer pricing, not market efficiency. Evaluating it against Walletto's €520K is not an apples-to-apples comparison — it is a comparison between a cost-center accounting construct and a market-priced product business. **Why This Organization Matters** Danske Bank Lithuania runs some of the most sophisticated enterprise engineering governance in the country. With 4,700 professionals, it encompasses: - Core banking platform engineering teams - Cloud transformation programs (Microsoft Azure at scale) - Advanced data and analytics platform development - SDLC governance and quality engineering - Enterprise security and compliance engineering - Risk technology and regulatory reporting systems This is the largest concentration of enterprise-grade software engineers and architects in Lithuania. The engineering practices, SDLC maturity, and compliance automation capabilities developed here are often significantly above the local market average. **Engineering Maturity Assessment** |Dimension|Score|Notes| |---|---|---| |Infrastructure Maturity|7/10|Azure-native, significant Kubernetes adoption| |DevOps / CI/CD|7/10|Enterprise CI/CD, constrained by governance| |Platform Engineering|7/10|Internal platforms, but enterprise-pace evolution| |AI Readiness|7/10|Active AI investment, responsible AI governance| |Cloud Native|7/10|Azure transformation ongoing| |Security Automation|9/10|Compliance automation is a strength| **Key Dynamic:** NIS2 + DORA compliance requirements may actually _improve_ Danske's relative engineering maturity position, as they drive systematic automation of compliance and security operations — areas where Danske already invests heavily. **Risk Factors:** Parent bank strategy dependency; enterprise bureaucracy limits velocity; headcount-based efficiency metrics create misleading external perception. --- ### 💳 Walletto **Sector:** Fintech / EMI / B2B Payment Infrastructure **Revenue:** ~€38M | **Employees:** ~73 | **Rev/Employee:** ~€520K _(highest in dataset)_ **The Lean Fintech Machine** Walletto's €520K revenue per employee is the highest figure in the entire Lithuanian IT dataset — and it is genuinely meaningful, not a statistical artifact. This figure reflects the operational model of a lean, automation-first electronic money institution operating a B2B payment infrastructure business. A licensed EMI providing white-label card issuing, IBAN accounts, and payment processing to other fintech companies is, at its core, an **infrastructure-as-a-service business with regulatory overlay**. Revenue scales with transaction volume and the number of client programs, not with linear headcount growth. A well-architected payment platform can serve 50 fintech clients with the same infrastructure that serves 5 — the marginal cost of adding clients is predominantly sales and integration, not engineering. **The DORA Dimension** For Walletto, DORA compliance is not overhead — it is a competitive signal. Clients evaluating a white-label payment infrastructure provider want assurance of operational resilience, ICT risk management, and documented business continuity capabilities. DORA compliance, when articulated correctly, transforms from regulatory burden to **sales asset**. Organizations investing in DORA compliance automation — automated operational resilience testing, documented ICT risk registers, machine-generated audit evidence — build capabilities that differentiate them in a market where buyers increasingly demand regulatory assurance from infrastructure providers. **Engineering Maturity Assessment** |Dimension|Score|Notes| |---|---|---| |Infrastructure Maturity|7/10|Cloud-native payment platform| |DevOps / CI/CD|7/10|Modern CI/CD for regulated fintech| |Platform Engineering|6/10|Lean team, platform investment scaling| |AI Readiness|6/10|AI adoption growing| |Cloud Native|7/10|Modern cloud architecture| |Security Automation|8/10|EMI compliance drives automation| **Scalability Trajectory:** Very High — B2B infrastructure model scales revenue without proportional headcount growth. The path from €38M to €100M+ does not require tripling the team. --- ### 💰 Paysera **Sector:** Fintech / Payments / Ecosystem **Revenue:** ~€100–120M | **Employees:** ~700 | **Rev/Employee:** ~€150K **The Pioneer Carrying Weight** Paysera is Lithuania's fintech pioneer — the company that built payment infrastructure before "fintech" was a category. With 20+ years of operation, it has accumulated a comprehensive product portfolio: payment processing, IBAN accounts, currency exchange, P2P transfers, and even logistics services. The €150K revenue-per-employee reflects the operational reality of a full-service fintech: compliance teams, customer support at scale, operations, logistics integration, and the overhead of serving both consumers and SMB merchants across multiple regulatory jurisdictions. The strategic challenge for Paysera is competing against Revolut, Wise, and Stripe — companies with significantly larger capital bases, more aggressive technology investment, and global brand recognition. Paysera's advantage is local brand trust and deep Lithuanian market penetration; its challenge is converting that into a sustainable competitive position in an increasingly commoditized payments market. **Engineering Maturity Assessment:** 6/10 composite. Solid engineering foundation with significant technical debt from 20+ years of product accumulation. DevOps modernization underway but constrained by legacy integration complexity. --- ### 🔗 ConnectPay **Sector:** Fintech / Banking-as-a-Service **Revenue:** ~€15–25M | **Employees:** ~120 | **Rev/Employee:** ~€150–200K **Full Banking Licence = Full Compliance Burden** ConnectPay's differentiation in the Lithuanian fintech market is its full EU **banking licence** — a significantly more powerful regulatory credential than an EMI licence, enabling a broader product set including deposit-taking and more complex financial services. The compliance cost of a banking licence is substantially higher than an EMI licence. With 120 employees managing full banking regulatory obligations, ConnectPay's operational overhead ratio is significant. The €150-200K revenue per employee is respectable given the regulatory context. The Banking-as-a-Service market opportunity is large — embedded finance is a genuine secular trend — and ConnectPay's banking licence provides structural differentiation. The execution challenge is building the engineering platform and commercial motion to convert licence advantage into market share before better-capitalized competitors occupy the space. --- ### 🌐 Kevin. **Sector:** Open Banking / A2A Payments **Revenue:** ~€20–40M | **Employees:** ~250 | **Rev/Employee:** ~€80–160K **The Infrastructure Play That Will Take Time** Kevin. is building open banking payment infrastructure — account-to-account (A2A) payment rails that allow merchants to accept bank transfers as seamlessly as card payments. The thesis is that PSD2 open banking regulation has created the technical infrastructure for A2A payments to challenge card networks on cost and speed. The €80-160K revenue per employee looks modest and reflects an early-stage infrastructure business investing heavily ahead of revenue. The relevant question is not current revenue efficiency but network effect trajectory: how many banks are connected, how many merchants are live, what is the transaction volume trend? If A2A payment adoption follows the trajectory seen in markets like the Netherlands (iDEAL) or Brazil (PIX), Kevin. is positioned at the infrastructure layer of a major payment system transition. If adoption is slower than expected, the runway and capital efficiency become critical. **Engineering Maturity:** 8/10 infrastructure maturity (PSD2-native API architecture), 7/10 DevOps. The technology architecture is modern and appropriate for the challenge. --- ### 🔧 InnoForce **Sector:** Enterprise / SAP / Integration Consulting **Revenue:** ~€14.8M | **Employees:** ~33 | **Rev/Employee:** ~€449K **The Specialist Premium** InnoForce's €449K revenue per employee is the second-highest in the dataset and reflects the economics of **specialist enterprise consulting** at very high utilization rates. With 33 employees generating €14.8M in revenue, this is a highly focused practice delivering premium-priced SAP and enterprise integration services to clients who value deep expertise over cost arbitrage. The organizational model is fundamentally different from high-headcount IT services companies: a small team of recognized specialists commands premium day rates, maintains very high billable utilization, and operates with minimal overhead. This is not a scalable model in the traditional sense — adding headcount is constrained by the difficulty of finding practitioners at the required expertise level — but it is an extremely profitable one. **Risk Factor:** Revenue is highly concentrated in a small team, creating key-person dependency. SAP ecosystem evolution (SAP S/4HANA migration wave) creates both opportunity and technology transition risk. --- ### 🛠️ NFQ Technologies **Sector:** Engineering / Product Development Services **Revenue:** ~€40–60M | **Employees:** ~800 | **Rev/Employee:** ~€50–75K **The Large Engineering Organization** NFQ Technologies represents a classic IT services company that has built significant scale in the Lithuanian market. With 800 employees generating €40-60M in revenue, it operates at the revenue-per-employee level typical of European IT services firms serving German and Scandinavian enterprise clients. The strategic challenge for NFQ — and for the Lithuanian IT services sector broadly — is the AI productivity displacement risk. As AI coding assistants reduce the marginal cost of software development, clients will expect to achieve equivalent output with smaller, more expensive (but AI-augmented) teams rather than larger, cheaper traditional development teams. Services firms that don't develop AI-augmented delivery models will face pricing pressure and margin compression. **DevOps Maturity:** 7/10. NFQ has invested in modern engineering practices and cloud delivery capabilities, which positions it better than pure staff-augmentation firms. The platform engineering investment is the key differentiator to watch. --- ### 🏗️ Devbridge (Cognizant) **Sector:** Software Engineering / Product Development **Revenue:** ~€30–50M (Lithuania) | **Employees:** ~600 | **Rev/Employee:** ~€50–80K **Premium Positioning, Enterprise Backstop** Devbridge built a strong premium positioning in enterprise software product development before its acquisition by Cognizant in 2021. The design-led engineering methodology and focus on complex product builds rather than staff augmentation commanded above-market rates. Post-acquisition, the Lithuanian operation benefits from Cognizant's global client network and financial stability but faces the organizational integration challenges typical of acquisitions: talent retention, cultural preservation, and maintaining the boutique-quality brand within a large corporate structure. **Engineering Maturity:** 7-8/10 composite. Strong UX/product engineering practices. DevOps capabilities solid. AI adoption is the emerging differentiator — Cognizant's global AI investment may benefit the Lithuanian team. --- ### 🌐 Tesonet **Sector:** Venture Builder / Technology Group **Revenue:** ~€200–400M (group) | **Employees:** ~3,500+ | **Rev/Employee:** ~€60–115K **The Hidden Architecture of Lithuanian Cybersecurity** Tesonet is the least understood organization in the Lithuanian tech ecosystem — and arguably the most strategically important. As the founding incubator and holding company behind Nord Security, it has built or co-built: NordVPN, NordLayer, NordPass, NordStellar, Surfshark (acquired), and multiple other technology ventures. The €60-115K revenue-per-employee at group level reflects the organizational complexity of a holding structure that includes parent operations, technology infrastructure, and venture-stage companies at various development stages. The underlying businesses — particularly Nord Security — operate at significantly higher revenue efficiency. Tesonet owns **significant proprietary global network infrastructure** — one of the larger privately-operated VPN and proxy network footprints globally. This infrastructure is a strategic asset with multiple monetization vectors across the portfolio. **Strategic Significance:** The Tesonet/Nord/Surfshark/Oxylabs cluster represents Lithuania's most strategically important technology asset — a globally competitive cybersecurity and internet infrastructure cluster with genuine competitive moats. --- ### 🔒 Nord Security **Sector:** Cybersecurity / SaaS **Revenue:** ~€300–500M | **Employees:** ~1,800–2,200 | **Rev/Employee:** ~€140–270K **Lithuania's Cybersecurity Champion** Nord Security has built one of the world's most recognizable cybersecurity consumer brands. NordVPN alone claims 14M+ subscribers — making it one of the largest VPN services globally by subscriber count. The product portfolio has expanded systematically: NordLayer (business VPN/ZTNA), NordPass (password manager), NordStellar (threat intelligence), and NordLocker (encrypted file storage). The €140-270K revenue per employee range reflects the economics of a global SaaS cybersecurity business at scale: high-margin subscription revenue, significant performance marketing investment, and a growing enterprise sales motion. **Infrastructure Scale:** Nord Security operates one of the world's larger private VPN server networks — 6,000+ servers across 111 countries. Managing this infrastructure at the reliability and performance levels required by 14M+ subscribers demands sophisticated DevOps, SRE, and network engineering capabilities. **Engineering Maturity Assessment** |Dimension|Score|Notes| |---|---|---| |Infrastructure Maturity|8/10|Global VPN infra at massive scale| |DevOps / CI/CD|8/10|Strong engineering culture| |Platform Engineering|8/10|Internal platforms at scale| |AI Readiness|7/10|Growing AI product integration| |Cloud Native|8/10|Hybrid cloud + proprietary infra| |Security Automation|9/10|Cybersecurity is the core business| **Strategic Risk:** VPN market headwinds from regulatory scrutiny (VPN bans in authoritarian markets) and platform-level VPN integration by mobile OS vendors (Apple, Google). Diversification into NordLayer (enterprise) and NordStellar (threat intelligence) is the strategic hedge. --- ### 🏄 Surfshark **Sector:** Cybersecurity / VPN / Consumer SaaS **Revenue:** ~€100–250M | **Employees:** ~400–700 | **Rev/Employee:** ~€150–400K **The Challenger That Became Family** Surfshark built a premium-priced, high-quality VPN brand with strong consumer NPS before its acquisition by Nord Security in 2022. The acquisition merged Lithuania's two largest VPN brands under the Tesonet holding structure. Operating as a distinct brand post-acquisition while sharing infrastructure with NordVPN creates interesting economics: Surfshark benefits from Nord's global server network, engineering capabilities, and vendor relationships while maintaining separate brand positioning targeting privacy-conscious premium consumers. Revenue per employee in the €150-400K range reflects high-margin subscription economics and the infrastructure leverage from Nord integration. --- ### 🖥️ Hostinger **Sector:** Cloud Infrastructure / Web Hosting **Revenue:** ~€110–150M | **Employees:** ~900 | **Rev/Employee:** ~€120–170K **The Infrastructure Automation Champion** Hostinger is arguably the most technically underrated company in the Lithuanian tech ecosystem. To the external observer, it appears to be a web hosting company. To anyone who looks at the engineering depth, it is a **global cloud infrastructure operator** serving 3M+ customers across 150+ countries with a custom control plane, proprietary data centre operations, and significant internal automation investment. The €120-170K revenue per employee is somewhat misleading given Hostinger's infrastructure density: a company operating at this headcount with this customer base has built **exceptional automation leverage**. Managing 3M customers across 150 countries with 900 employees implies infrastructure automation maturity that most companies twice the size haven't achieved. **AI Investment:** Hostinger has made notable AI investments — AI website builder (Horizons AI), AI-powered customer support, and infrastructure AIOps. The AI website builder positions Hostinger in the adjacent market of AI-assisted web presence creation. **Engineering Maturity Assessment** |Dimension|Score|Notes| |---|---|---| |Infrastructure Maturity|9/10|Custom control plane, proprietary DC ops| |DevOps / CI/CD|9/10|Automation-first culture| |Platform Engineering|8/10|Strong internal tooling| |AI Readiness|8/10|Active AI product and ops investment| |Cloud Native|9/10|Cloud-native at core| |Security Automation|7/10|Solid, not exceptional| --- ### 🕷️ Oxylabs **Sector:** Web Intelligence / Proxy Infrastructure **Revenue:** ~€80–150M | **Employees:** ~500+ | **Rev/Employee:** ~€160–300K **The Internet Infrastructure Layer** Oxylabs operates one of the world's largest web data collection infrastructure businesses — residential proxy networks, datacenter proxies, web scraping APIs, and SERP data. Serving enterprise clients in e-commerce, travel, financial data, and market intelligence, Oxylabs sits at the **data infrastructure layer** of the internet economy. The €160-300K revenue per employee reflects the economics of infrastructure-as-a-service at scale: once the network is built and the software is written, each additional enterprise client adds revenue with relatively modest marginal operational cost. **Infrastructure Complexity Index:** Extreme. Managing a residential proxy network of millions of nodes across hundreds of countries, with compliance obligations across different data privacy regimes, while providing enterprise-grade SLAs, is an operationally complex undertaking. Standard efficiency metrics completely miss this complexity. **AI Leverage Potential:** Very High. AI-driven scraping target adaptation, automated anti-bot bypass research, and ML-powered network optimization represent significant opportunity to reduce operational headcount relative to revenue. --- ### 💪 Kilo Health **Sector:** Digital Health / Consumer SaaS **Revenue:** ~€150–250M | **Employees:** ~600 | **Rev/Employee:** ~€250–400K **The AI-Native Health Platform** Kilo Health has built Lithuania's largest digital health business through an aggressive portfolio strategy: rather than building a single health app, it has developed and acquired multiple health and wellness apps (Omo, DoFasting, Wellbeing, and others) serving 5M+ paying subscribers globally. The €250-400K revenue per employee is among the highest in the dataset and reflects the SaaS subscription economics of digital health: high gross margins, subscription revenue, and a relatively lean engineering organization augmented by significant performance marketing investment. **AI Differentiation:** Kilo Health has positioned AI-driven personalization as a core product value proposition. Personalized nutrition plans, adaptive coaching, and ML-driven behavior change interventions differentiate the product in a crowded health app market. **Risk:** Digital health regulation is tightening — FDA oversight of wellness apps, CE marking requirements in Europe for certain digital health categories, and data privacy obligations for health data represent increasing compliance overhead. --- ### 🛒 Vinted **Sector:** Marketplace / C2C Platform **Revenue:** ~€600M+ | **Employees:** ~1,800+ | **Rev/Employee:** ~€330K **The Lithuanian Unicorn — And Europe's Second-Hand Champion** Vinted is Lithuania's signature technology success story. As Europe's largest online second-hand clothing marketplace operating in 20+ countries with hundreds of millions of registered users, Vinted has proven that world-class consumer marketplace engineering can be built from Vilnius. The €330K revenue per employee is impressive and reflects genuine marketplace economics: Vinted monetizes through buyer protection fees and integrated shipping, with revenue scaling with gross merchandise volume rather than proportionally with headcount. **Engineering Organization:** Vinted has publicly documented its engineering culture through its engineering blog and conference presentations. Key characteristics: - **Distributed squad model:** Autonomous product teams aligned to marketplace domains - **Platform engineering investment:** Dedicated platform team supporting 200+ product engineers - **Data-driven product development:** Sophisticated A/B testing and experimentation infrastructure - **GitOps maturity:** ArgoCD-based deployment, infrastructure-as-code throughout **Engineering Maturity Assessment** |Dimension|Score|Notes| |---|---|---| |Infrastructure Maturity|9/10|Kubernetes-native, multi-cloud| |DevOps / CI/CD|9/10|Elite DORA metrics| |Platform Engineering|9/10|Dedicated platform team, IDP| |AI Readiness|8/10|ML at core (recommendations, fraud)| |Cloud Native|9/10|Cloud-native throughout| |Security Automation|7/10|Good, not exceptional| --- ### 🔌 Penki Kontinentai Group **Sector:** Banking Tech / ICT Infrastructure **Revenue:** ~€60–70M | **Employees:** ~500–700 | **Rev/Employee:** ~€90–140K **The Infrastructure Conglomerate** Penki Kontinentai is a multi-subsidiary ICT conglomerate operating across telecommunications, data centre services, payment processing, and enterprise IT services. The group structure creates accounting and analysis complexity — different subsidiaries operate under different business models with different margin profiles. The €90-140K revenue per employee reflects the capital-intensive economics of combined hardware infrastructure and software services. Data centre operations and telecommunications infrastructure require significant capital investment with relatively modest revenue per employee compared to pure-software businesses. **Strategic Challenge:** The cloud hyperscaler wave (AWS, Azure, GCP) is progressively commoditizing the data centre and managed infrastructure market. Organizations with physical infrastructure assets must articulate a differentiated value proposition — typically regulatory data residency, sovereign cloud positioning, or hybrid connectivity specialization — to avoid margin compression from hyperscaler competition. --- ### 🏧 BS/2 **Sector:** ATM Software / Banking Systems **Revenue:** ~€39M | **Employees:** ~250–350 | **Rev/Employee:** ~€110–156K **The ATM Software Specialist at a Crossroads** BS/2 occupies a globally unique position: it is one of the world's few independent ATM driving software vendors, serving financial institutions across 70+ countries with its .iQ product suite covering ATM management, cash optimization, and banking channel management. The €110-156K revenue per employee is reasonable for an enterprise software + hardware hybrid business. The 70+ country footprint implies significant international operations overhead: local sales, support, compliance with country-specific regulatory requirements, and hardware logistics. **The Secular Challenge:** Cash usage is in structural decline across developed markets. Sweden is near-cashless. The UK, Netherlands, and Nordic markets are seeing ATM network consolidation. Central and Eastern European markets — a BS/2 stronghold — are still cash-dependent, but the trajectory is clear. BS/2's strategic imperative is product portfolio expansion into digital banking channels, real-time payments infrastructure, and cashless point-of-sale technology — before the ATM software market contracts to a level that threatens the business. **Engineering Maturity:** 5-6/10. Engineering practices reflect a company that has prioritized stability and reliability in a hardware-adjacent context over DevOps velocity. The SDLC modernization investment documented in public materials suggests awareness of the gap. --- ## 7. Comparative Scoring Tables ### Table 7.1 — Financial & Operational Efficiency |Company|Revenue|Employees|Rev/Emp|Model Type|Regulatory Burden| |---|---|---|---|---|---| |Vinted|€600M+|~1,800|~€330K|Marketplace|Medium| |Nord Security|€300-500M|~2,000|~€140-270K|SaaS|Medium-High| |Kilo Health|€150-250M|~600|~€250-400K|Health SaaS|Medium-High| |Paysera|€100-120M|~700|~€150K|Payments|High| |Surfshark|€100-250M|~400-700|~€150-400K|SaaS|Medium-High| |Hostinger|€110-150M|~900|~€120-170K|Cloud Infra|Low-Medium| |Oxylabs|€80-150M|~500|~€160-300K|Data Infra|Medium| |Tesonet|€200-400M|~3,500|~€60-115K|Holding|Medium-High| |Penki Kontinentai|€60-70M|~500-700|~€90-140K|ICT Group|Medium| |NFQ Technologies|€40-60M|~800|~€50-75K|IT Services|Low| |Devbridge|€30-50M|~600|~€50-80K|IT Services|Low| |BS/2|€39M|~250-350|~€110-156K|Enterprise SW|Medium-High| |Walletto|€38M|~73|**~€520K**|EMI / BaaS|Very High| |Kevin.|€20-40M|~250|~€80-160K|Open Banking|High| |ConnectPay|€15-25M|~120|~€150-200K|BaaS|Very High| |InnoForce|€14.8M|~33|**~€449K**|SAP Consulting|Low| |Danske Bank LT|€248M (intra)|~4,700|~€52K*|Shared Service|Extreme| _*Revenue/employee not meaningful for cost-centre shared service model_ ### Table 7.2 — Engineering Maturity Scores (1-10) |Company|Infra|DevOps|Platform|AI Ready|Cloud|Security|**Composite**| |---|---|---|---|---|---|---|---| |Vinted|9|9|9|8|9|7|**8.5**| |Hostinger|9|9|8|8|9|7|**8.3**| |Nord Security|8|8|8|7|8|9|**8.0**| |Oxylabs|8|8|7|7|8|7|**7.5**| |Surfshark|7|8|7|7|8|8|**7.5**| |Kevin.|8|7|6|6|8|7|**7.0**| |Kilo Health|7|7|7|8|8|6|**7.2**| |Devbridge|7|7|7|7|7|6|**6.8**| |Walletto|7|7|6|6|7|8|**6.8**| |Tesonet|8|8|7|7|8|8|**7.7**| |Danske Bank LT|7|7|7|7|7|9|**7.0**| |ConnectPay|7|6|6|5|7|8|**6.5**| |NFQ Technologies|7|7|6|6|7|6|**6.5**| |Paysera|6|6|5|5|6|7|**5.8**| |InnoForce|6|6|5|5|6|5|**5.5**| |BS/2|6|6|5|5|5|6|**5.5**| |Penki Kontinentai|5|5|4|4|5|5|**4.7**| ### Table 7.3 — Infrastructure Complexity Index This index attempts to capture the operational complexity that revenue-per-employee ignores: |Company|Complexity Type|Complexity Level|Rev/Emp Distortion| |---|---|---|---| |Vinted|Extreme marketplace scale|Very High|Positive (hides complexity)| |Danske Bank LT|Enterprise governance + compliance|Extreme|Negative (hides sophistication)| |Oxylabs|Global proxy network, anti-bot|Extreme|Moderate positive| |Nord Security|Global VPN infra, 6K+ servers|Very High|Moderate positive| |Hostinger|Global DC ops, custom control plane|Very High|Negative (hides efficiency)| |BS/2|Hybrid hardware + software, 70+ countries|High|Moderate negative| |Penki Kontinentai|Data centres + telco + payments|High|Negative| |Walletto|Regulatory complexity per employee|High|Strongly positive (true outlier)| |Kevin.|Pan-EU bank connectivity, PSD2|High|Negative (early stage)| |ConnectPay|Full banking licence compliance|Very High|Negative| --- ## 8. Three Hidden Superpowers of Lithuanian Tech ### Superpower 1: Lithuania Became an Engineering Economy The outsourcing frame — competent, cost-effective engineers delivering work designed elsewhere — no longer describes the Lithuanian technology sector's most significant activity. Vinted's marketplace algorithms, Nord Security's cryptographic infrastructure, Hostinger's control plane, Oxylabs' anti-bot technology, Kevin.'s open banking connectivity layer — these are not bodies-for-hire executing someone else's technical vision. These are **original engineering** of globally competitive technology at scale. The distinction matters strategically. Outsourcing countries compete on cost. Engineering economies compete on capability, talent density, and proprietary technology. Lithuania has made this transition for the top tier of its technology sector, and the implications for talent strategy, capital formation, and national competitiveness are significant. ### Superpower 2: Cybersecurity Is Lithuania's Hidden Competitive Moat The cluster of cybersecurity and internet infrastructure companies emanating from the Tesonet ecosystem is one of Europe's most strategically significant technology concentrations — and one of the least discussed. Consider the aggregate scale: - **Nord Security:** 14M+ VPN subscribers globally, €300-500M revenue - **Surfshark:** 3M+ subscribers, €100-250M revenue - **Oxylabs:** Leading web data infrastructure, €80-150M revenue - **Tesonet:** Shared internet infrastructure underpinning all of the above This cluster collectively represents: one of the world's larger private VPN network footprints, proprietary proxy infrastructure at internet scale, deep cryptographic and network engineering expertise, and global consumer cybersecurity brand recognition. This is not a coincidence of independent companies. It is an **engineered ecosystem** — a technology holding structure that shares infrastructure investment, talent, and strategic direction across multiple product brands. The strategic intelligence embedded in this structure is significantly underappreciated by external analysts. ### Superpower 3: Shared-Service Metrics Are Systematically Misleading Danske Bank Lithuania, assessed on revenue-per-employee, looks like the "least efficient" organization in the dataset at €52K. This conclusion is structurally wrong and analytically misleading. Danske Bank Lithuania is running: - Enterprise platform engineering for a €4B+ revenue Nordic bank - SDLC governance for hundreds of enterprise applications - Cloud transformation programs at significant scale - Compliance automation for one of the most regulated industries in the world - Data platform engineering for financial analytics at enterprise scale The €52K revenue-per-employee figure is an intra-group accounting artifact, not a measure of engineering output, talent quality, or operational sophistication. The analytical failure mode is using this figure to conclude that Walletto (€520K) is "10x more efficient" than Danske Bank — a comparison that has no meaningful content. **The correct observation:** These organizations are not in the same analytical category. Comparing their revenue-per-employee figures is like comparing the revenue per employee of an R&D laboratory to a product sales organization and concluding that R&D is "inefficient." --- ## 9. Why AI Changes the Economics of Engineering ### The Productivity Discontinuity AI-assisted software development is not a marginal productivity improvement. GitHub's own research (2022-2024) documented 55% faster task completion on measurable coding tasks with Copilot. McKinsey's 2023 research estimated AI tools could accelerate full SDLC cycles by 20-45%. Cursor IDE users in high-performance engineering organizations report 30-50% reductions in time-to-implementation for standard features. The critical insight is that these gains are **non-linear with team size** and **disproportionately powerful for smaller expert teams**. A 20-engineer team with: - AI coding assistants (Cursor/Copilot/Claude) - AI-powered code review (CodeRabbit/Qodo) - AI test generation - AI-driven infrastructure provisioning - AIOps for observability ...working on a well-architected platform with GitOps automation may genuinely outperform a 100-engineer traditional team on feature velocity, defect rate, and system reliability. This is not a theoretical projection. It is becoming measurable in 2024-2025 engineering organizations that have invested systematically in AI augmentation. ### What This Means for Lithuanian IT Companies **The AI leverage potential by company type:** |Company Type|AI Leverage Potential|Constraint| |---|---|---| |SaaS/Product (Vinted, Kilo Health)|Very High|Already capturing some gains| |Cybersecurity (Nord, Surfshark)|Very High|Regulatory validation adds overhead| |Data Infrastructure (Oxylabs, Hostinger)|Extreme|Infrastructure code is highly automatable| |Open Banking (Kevin.)|High|Regulatory testing overhead| |Fintech EMI (Walletto, ConnectPay)|High|Compliance automation is AI-appropriate| |Enterprise/Banking (Danske, BS/2)|Medium|Governance gates, validation requirements| |IT Services (NFQ, Devbridge)|Medium|Client delivery models must evolve| **The AI Threat to IT Services Models** NFQ Technologies and Devbridge represent capable, well-regarded IT services organizations. Their business model — billing clients for engineering hours and expertise — faces direct disruption from AI productivity multiplication. If AI tools enable a client to achieve equivalent output with 40 engineers that previously required 100, the market for 60 engineers of services delivery disappears. This dynamic will not manifest overnight — enterprise clients change slowly, AI adoption is uneven, and the demand for technology work is growing. But the structural pressure on time-and-materials services models is real and increasing. The strategic response is to move from selling engineering hours to selling **engineered outcomes** — shifting from staff augmentation to managed delivery, platform services, and proprietary intellectual property. Companies that make this transition early will be better positioned for the AI-compressed services market of 2027-2030. ### Platform Engineering as a Financial Multiplier The most strategically important engineering investment for the 2025-2030 period is not AI tools themselves — it is the **platform that enables all engineers to use AI tools effectively**. An Internal Developer Platform (IDP) that integrates: - AI-powered CI/CD pipelines with automated quality gates - AI-assisted security scanning with contextual remediation guidance - AI-driven infrastructure provisioning from natural language specifications - AI observability with automated incident correlation and runbook suggestion ...creates compounding productivity advantages. Each engineer on such a platform becomes more effective. The platform team's leverage is extraordinary: 10 platform engineers enabling 200 product engineers to be 40% more productive is a 20x leverage ratio on the platform investment. Vinted, Hostinger, and Nord Security appear to be the Lithuanian organizations with the most mature platform engineering investments. Their productivity trajectories will likely separate from less platform-mature peers over the 2025-2028 period in ways that current revenue-per-employee metrics cannot capture — but that will eventually show up in profitability and competitive position. --- ## 10. The EAOEI: A Better Efficiency Framework ### Engineering-Adjusted Organizational Efficiency Index The EAOEI is a multi-dimensional scoring framework designed to replace simplistic revenue-per-employee analysis for modern AI-era engineering organizations. It weights ten dimensions to produce a composite efficiency score. ### Dimension Weights & Definitions |Dimension|Weight|What It Captures| |---|---|---| |Financial Efficiency|20%|Adjusted revenue per engineer, EBITDA margin, growth trajectory| |Engineering Maturity|15%|Architecture quality, DORA performance, delivery capability| |Automation Maturity|12%|Operational task automation %, toil elimination| |AI Readiness|12%|AI tool adoption, AI product integration, AIOps maturity| |Platform Maturity|12%|IDP sophistication, developer experience, golden path coverage| |Organizational Scalability|8%|Revenue growth without proportional org complexity growth| |Innovation Capability|8%|R&D investment, new product velocity, technical debt ratio| |Infrastructure Complexity Handling|5%|Cost efficiency vs. managed complexity| |Regulatory Resilience|5%|Compliance automation, audit readiness| |Operational Sustainability|3%|Retention, culture health, ESG| ### EAOEI Scores |Company|Fin|Eng|Auto|AI|Platform|Scale|Innov|**EAOEI**| |---|---|---|---|---|---|---|---|---| |Vinted|8.5|9.0|9.0|8.0|9.0|9.0|9.0|**8.8**| |Hostinger|8.0|8.0|9.0|8.0|8.0|8.0|8.0|**8.2**| |Nord Security|8.5|8.0|8.0|7.0|8.0|8.0|8.0|**8.0**| |Kilo Health|8.0|7.0|7.0|8.0|7.0|7.0|7.0|**7.4**| |Oxylabs|7.5|8.0|8.0|7.0|7.0|7.0|7.0|**7.4**| |Surfshark|8.0|7.5|8.0|7.0|7.0|6.0|7.0|**7.4**| |Tesonet|7.0|8.0|8.0|7.0|8.0|8.0|8.0|**7.6**| |Kevin.|4.5|8.0|7.0|6.0|6.0|8.0|8.0|**6.7**| |Walletto|9.0|7.0|7.0|6.0|6.0|7.0|6.0|**7.1**| |Devbridge|6.5|7.5|7.0|7.0|7.0|6.0|7.0|**6.9**| |Danske Bank LT|n/a|7.0|6.0|7.0|7.0|4.0|7.0|**6.5***| |ConnectPay|6.5|7.0|6.0|5.0|6.0|7.0|6.0|**6.4**| |NFQ Technologies|5.5|7.0|6.0|6.0|6.0|5.0|6.0|**6.1**| |Paysera|6.0|6.0|6.0|5.0|5.0|5.0|5.0|**5.6**| |InnoForce|9.0|6.0|5.0|5.0|5.0|3.0|5.0|**5.9**| |BS/2|5.5|5.5|6.0|5.0|5.0|4.0|4.0|**5.2**| |Penki Kontinentai|5.0|5.0|5.0|4.0|4.0|4.0|4.0|**4.6**| _*Danske score excludes Financial Efficiency (cost-centre model) and uses engineering-only dimensions_ ### Interpretation |EAOEI Range|Assessment| |---|---| |**8.0 – 10.0**|AI-era leaders. Compounding advantages. Strong 5-year trajectory.| |**6.5 – 7.9**|Advanced performers. Clear improvement vectors, fundamentally healthy.| |**5.0 – 6.4**|Developing. Strong in one-two dimensions, lagging in automation/AI. Strategic gap exists.| |**< 5.0**|Structural challenges. Transformation required to remain competitive through 2028-2030.| --- ## 11. Strategic Conclusions & 2030 Outlook ### The Five Predictions #### 1. AI Will Make Headcount a Lagging Indicator by 2028 By 2028, engineering organizations with mature AI augmentation and platform engineering will produce equivalent or superior output to organizations twice their size without these capabilities. Revenue-per-employee will become increasingly useless as a comparative metric as the productivity range between AI-native and non-AI-native organizations widens. The relevant metric will be **revenue per AI-augmented engineer-equivalent** — a figure that accounts for the productivity multiplier effect. #### 2. The Lithuanian Cybersecurity Cluster Will Grow More Dominant The Tesonet/Nord/Surfshark/Oxylabs ecosystem has built infrastructure advantages — proprietary global networks, cryptographic engineering depth, consumer brand trust — that are genuinely difficult to replicate. As cybersecurity spending grows globally (driven by geopolitical instability, regulatory requirements, and AI-driven threat evolution), this cluster is positioned to capture disproportionate market share. The risk is regulatory and reputational — VPN regulation in major markets and questions about residential proxy ethics could create headwinds. #### 3. Platform Engineering Will Become a Board-Level Priority Within 3-5 years, every significant Lithuanian technology company will have an explicit Internal Developer Platform strategy. The difference between organizations will not be whether they have a platform — it will be how sophisticated it is and how effectively it integrates AI capabilities. CTOs who can quantify platform ROI in terms of deployment frequency improvements, lead time reduction, and developer productivity gains will access capital and organizational resources that platform-skeptics cannot. #### 4. The BaaS/EMI Infrastructure Wave Has 5 Years of Tailwind Embedded finance — the integration of payment, lending, and banking capabilities into non-financial products — is a secular growth trend. Walletto, ConnectPay, and Kevin. are positioned at the infrastructure layer of this trend. If they execute on engineering maturity and commercial development over the next three to five years, their revenue-per-employee metrics will look dramatically different — and their EAOEI scores will reflect it. The constraint is execution speed relative to better-capitalized European competitors. #### 5. IT Services Models Must Transform or Face Structural Decline NFQ Technologies, Devbridge, and InnoForce represent capable organizations facing structural market shift. The AI productivity compression of software services delivery will change the competitive dynamics of the European IT services market between 2025 and 2030. Organizations that transition from **selling engineering hours** to **delivering engineered outcomes** — proprietary platforms, managed services, IP-rich solutions — will navigate this transition. Organizations that remain primarily staff-augmentation or time-and-materials will face margin compression and client concentration risk. ### The Companies Best Positioned for 2030 **Tier 1 — Structural Advantages Compounding:** - **Vinted** — marketplace network effects + engineering excellence - **Nord Security** — cybersecurity brand + global infrastructure - **Hostinger** — infrastructure automation + AI product expansion **Tier 2 — Strong Foundation, Execution Upside:** - **Oxylabs** — data infrastructure moat, AI leverage potential - **Kilo Health** — AI-native health platform, subscription economics - **Kevin.** — open banking infrastructure, if A2A adoption accelerates **Tier 3 — Transformation Required:** - **Paysera** — strong brand, modernization investment needed - **BS/2** — product diversification existential for long-term survival - **NFQ / Devbridge** — business model evolution from hours to outcomes ### The Final Thesis Lithuania's technology sector is at a strategic inflection point. The outsourcing era is over for the top tier. The engineering economy has arrived. The AI era is beginning. The companies that will define Lithuanian technology by 2030 are not those with the highest current revenue per employee. They are the companies that are building: - The most sophisticated engineering platforms - The deepest AI integration into both product and operations - The most automated compliance and governance frameworks - The most scalable organizational architectures > _"The next generation of Lithuanian technology champions will not be defined by how many engineers they hire. They will be defined by how effectively each engineer is amplified by AI, automation, and platform engineering. The era of headcount-as-strategy is over."_ The metric that will matter is not revenue per employee. It is **engineering leverage** — the ratio of business output to engineering input, where input includes not just human effort but the AI, automation, and platform tooling that multiplies that effort. Organizations that optimize for engineering leverage today are building the structural efficiency advantages that will compound through the decade. --- ## Appendix: Methodology Notes **Data Sources:** Lithuanian business registry (Registrų Centras), Invest Lithuania GBS Report 2024, Lithuania Tech Ecosystem Report 2024, company engineering blogs, job posting analysis, conference presentation content, Tesonet ecosystem overview, and Lithuanian startup market overview from Startup Lithuania and Practica Capital research. **Estimation Approach:** Revenue and employee count estimates for private companies are triangulated from: confirmed registry filings where available; sector analyst benchmarks; per-employee cost modeling from job posting data; Invest Lithuania sector reports; and cross-validation against comparable company public data. **Scoring Methodology:** Engineering maturity scores are based on observable proxies — technology stack signals from job postings, engineering blog content, GitHub activity, conference presentations, and architectural patterns — not from direct assessment. Scores are directional, not precise. **Limitations:** This analysis is based on publicly available information. Private company data carries significant uncertainty. EAOEI scores reflect the author's analytical judgment and should be interpreted as a framework for structured comparison, not as definitive rankings. --- _Analytical Research Report | Lithuanian IT & Fintech Ecosystem | May 2026_ _Methodology: McKinsey / Gartner Executive Research Style | Engineering Maturity Assessment | EAOEI Framework_ ---