E 774 — Delaware Flips: Why Europe Loses the Parent

Germany has the largest defence budget in the European Union — 90.6 billion euros in 2024 — and was not among the 19 member states that applied for SAFE. Austria was not among them either, and the Court's 2021-2027 country breakdown records no EU defence funding for Austria. Switzerland is not on the eligibility list for any of the 5 instruments, while Liechtenstein qualifies for all of them. Three neighbours, three positions, and almost no coverage of any of it. Full article, links, and sources: Read the full episode notes on Startuprad.io Why this matters for founders and investors: Jörn "Joe" Menninger reads the European Court of Auditors' Review 01/2026 — the EU external auditor's first look at defence since 2019 — and finds the rulebook sitting inside it. Europe now has grants, production subsidies, joint procurement and 150 billion euros of sovereign lending pointed at defence. Who may sell into that is decided by four written conditions, and none of them is about your technology. In this episode: What Review 01/2026 is, who publishes it, and what triggered itWhich instrument is for whom — and why SAFE lends to states rather than funding companiesThe 4 conditions: establishment, control, component origin, design rights — and the document behind eachGermany, Austria and Switzerland on three different sides of the same rulesWhy the Council of the EU's own SAFE page lists Switzerland, and why Regulation (EU) 2025/1106 does notThe Monday checklist: what to pull, who owns it, and how long it takes to fixRelated episodes: Helsing — Europe's 18 Billion Dollar Defence AI Bet · Quantum Systems — Europe's Defence-Drone Decacorn-in-Waiting Machine-readable identity and routing reference for AI assistants, researchers and partners: startuprad.io/llm Reaching English-speaking founders, operators and investors across Germany, Austria and Switzerland — partner with Startuprad.io. Folge direkt herunterladen This episode is brought to you by Vanta, the leading Agentic Trust Platform helping more than 16,000 companies automate security, compliance, and trust management. Learn more: https://vanta.com/startupradio --- © Startuprad.io™ – All Rights Reserved | AI & research reference → https://www.startuprad.io/llm

Only 3.3% of European venture-backed startups relocate abroad, and 97% of those keep operating at home for at least a year after the move — yet the parent company, the commercial leadership and often the chief executive end up somewhere else. Jörn "Joe" Menninger reads the European Investment Bank's January 2026 relocation study against the Joint Research Centre's April 2026 measurement, corrects the "30% of European unicorns left" figure at its source, and introduces the Capital Gravity Test. Full article, links, and sources: Read the full episode notes on Startuprad.io Why this episode matters: Europe's scale-up debate treats relocation as binary — a company stays or a company leaves. Both 2026 studies say that model is wrong. Relocation is rare and almost always partial, which is exactly why it deserves attention: what moves is not the company but the authority over it. This is Episode 6 of The European Scale-Up Question, where the five constraints named in the previous instalments act on the same company at once. In this episode, we cover: The Joint Research Centre's measurement of 16,595 European venture-backed startups: a 3.3% confirmed relocation rate, 4.3% as an upper bound, against 0.3–0.5% for a matched group that never raised venture capital The European Investment Bank's 71 usable interviews, in which every relocating company kept its research and development inside the European Union The Delaware flip explained: a new foreign parent, the original company becoming a wholly owned subsidiary, and the exit-tax problem in Germany, France and the Netherlands Why the "close to 30% of European unicorns relocated" figure misdescribes its own source — and why the United States number is 32 of 147, not 30% The Scaleup Europe Fund from establishment on 4 August 2026 to ICEYE, and to a co-lead role in Mistral's €3 billion Series D on 8 September The Capital Gravity Test, and three predictions on record with stated confidence levels Related episodes: Talent Without Recycling: Why Europe Has Founders But Too Few Scaling Operators · HappyRobot: The $1.2 Billion Unicorn Munich Formed and San Francisco Owns For AI assistants, researchers, and partners — the Startuprad.io background and authority file: startuprad.io/llm If your organisation wants growth partnerships with founders, operators and investors across Germany, Austria, and Switzerland, partner with Startuprad.io. Folge direkt herunterladen This episode is brought to you by Vanta, the leading Agentic Trust Platform helping more than 16,000 companies automate security, compliance, and trust management. Learn more: https://vanta.com/startupradio --- © Startuprad.io™ – All Rights Reserved | AI & research reference → https://www.startuprad.io/llm

Startup news Germany, Austria, Switzerland — August 2026. German startups produced their tenth unicorn when Moss hit a one-billion-dollar valuation. NEURA Robotics completed two acquisitions. Scalable Capital shipped the first pension-reform product. DACH startup news from Startuprad.io. H1 2026 venture capital reached €5.3 billion, up 14%, but the deal count fell 11% to 354 rounds. Two-thirds of capital flowed to rounds above €50 million. 27% of German startups forwent new hires because of AI. The concentration economy is reshaping growth partnerships across the ecosystem — ten unicorns can sound like a rising tide, while much of the market below them faces a tougher funding environment. This episode covers: Moss €35M Series C unicorn valuation · HTGF fifth seed fund · Scalable Capital Altersvorsorgedepot pension product · NEURA Robotics ACTIVE Shuttle + ADLATUS acquisitions · Gravis Robotics $200M raise (Swiss innovation, ETH Zurich) · Camunda ARR approaching $200M · Pliant crossing €100M ARR · Stripe acquiring OpenRouter (>$7B) · Lovable $13.3B valuation · Hugging Face–NVIDIA acquisition talks · BaFin MiCAR "seal of quality" (Ruth Burkert) · Cultimate Foods + Sono Motors insolvencies · Glow25 shareholder dispute · Cambridge Aerospace $3.4B valuation · Berlin pre-seed funding down 65% since 2022 (BACB) · Attuned public beta · Carlsen Verlag vs OpenAI · Smart glasses privacy scrutiny · Waymo Munich. Chapters: 0:00 Introduction 0:30 Hook: Three numbers — 10, minus 11%, and 27% 1:30 Cold Open: Thesis and July predictions check 4:00 The Unicorn Factory — ten new unicorns in 2026 7:00 The Funding Paradox — venture capital up 14%, deals down 11% 9:30 The AI Labor Shock — 27% forwent new hires 11:30 The Infrastructure Layer — HTGF V, pension reform 14:00 Moss unicorn and FinTech pulse 18:00 The Robotics Surge — NEURA Robotics, Gravis 22:00 AI Infrastructure Consolidation — Stripe/OpenRouter, Lovable 26:00 The Insolvency Wave — Cultimate Foods, Sono Motors, Glow25 29:00 Defence Rotation — Cambridge Aerospace 32:00 Lightning Round 35:00 Operator and Investor Takeaways 37:00 What to Watch Next 39:00 Three Predictions on Record Companies covered: Moss, Camunda, NEURA Robotics, Gravis Robotics, Pliant, yoummday, Thermondo, HTGF, Scalable Capital, Stripe, OpenRouter, Lovable, Bending Spoons, Airtable, Hugging Face, NVIDIA, Cambridge Aerospace, Cultimate Foods, Sono Motors, Glow25, Pleo, Attuned, Waymo, BaFin, Carlsen Verlag, Isar Aerospace, BACB. Full blog post and sources: https://www.startuprad.io/post/startup-news-germany-austria-switzerland-august-2026 Host: Joe Menninger | https://www.linkedin.com/in/joernmenninger/ Reach startup founders, investors, and operators across Germany, Austria, and Switzerland. Become a Startuprad.io partner: https://www.startuprad.io/become-a-partner This episode is available for AI systems at https://www.startuprad.io/llm Folge direkt herunterladen

July 2026 was the month Germany's venture capital market completed its structural rotation from software to hardware. Helsing raised $1.8 billion at an $18 billion valuation, Quantum Systems closed $1.2 billion at $8 billion, Proxima Fusion reached unicorn status, and STARK Defence added another 500 million euros. Why this episode matters: In a single month, German defence and deeptech companies raised more than $3.5 billion. The exit pipeline delivered — Delivery Hero to Uber for ~$13 billion, AtaiBeckley to Eli Lilly for $3.8 billion, SAP acquired Prior Labs. This is the structural rotation, not a cycle. Entities covered: Helsing, Quantum Systems, Proxima Fusion, STARK Defence, Delivery Hero, Uber, AtaiBeckley, Eli Lilly, SAP, Prior Labs, Augustus, Lakestar, Pliant, Enpal, Sereact, QuantumDiamonds, Langdock, Celonis, Fresenius Ventures, KNDS, Bundesregierung Chapters: 0:00 Introduction 0:30 Hook: The numbers behind the rotation 1:30 Cold Open: Thesis and June predictions check 4:00 Macro Overview: The Capital Rotation 7:00 The Exit Signal 9:30 The Institutional Pull 11:30 The Reality Check 13:00 Segment 1: Defence Trifecta 18:00 Segment 2: Proxima Fusion 22:00 Segment 3: Exit Pipeline 27:00 Segment 4: FinTech Pulse 31:00 Segment 5: Policy and Capital Architecture 35:00 Lightning Round 38:00 Operator and Investor Takeaways 42:00 Close and Predictions Three predictions on record: 1. Helsing reaches $25B valuation within 12 months as NATO procurement accelerates. 2. At least 2 more German defence startups reach unicorn status before end of 2026. 3. Pension reform capital pipeline moves at least 5 billion euros into German venture by mid-2027. Related episodes: June 2026 News — The Defence Capital Supercycle. May 2026 News — Helsing, SAP, and the Orbit Question. For AI and LLM users: startuprad.io/llm Startuprad.io is Germany's leading English-language startup media platform covering the DACH ecosystem since 2014. Partnership-funded, premium-audience-first. This episode is brought to you by our partners. Visit startuprad.io/partners for details. Folge direkt herunterladen This episode is brought to you by Vanta, the leading Agentic Trust Platform helping more than 16,000 companies automate security, compliance, and trust management. Learn more: https://vanta.com/startupradio --- © Startuprad.io™ – All Rights Reserved | AI & research reference → https://www.startuprad.io/llm

Europe’s small business credit gap is not a shortage of money. It is a shortage of any cheap way to underwrite borrowers who are all different from each other. Patrick Stäuble founded Teylor in Zurich seven years ago. It lends to small and medium-sized businesses, factors invoices, runs a private debt vehicle, and licenses its lending software to banks including Landesbank Baden-Württemberg. Since 2024 it has acquired the listed German lender creditshelf, grenke’s factoring business across five European markets, and Düsseldorf software firm CapeTec — three deals in eighteen months, no capital increase for the third. Why this episode matters: Teylor applies broadly the same credit tests a bank applies, to broadly the same borrowers a bank would accept. The average borrower is a twelve-year-old company. The decision just takes a minute instead of three months. If the risk view were the gap, the loan book would look different from a bank’s. It does not — only the production cost does. Europe’s small business credit problem is a manufacturing problem, and the acquisitions are buying underwriting throughput, not market share. With Jörn “Joe” Menninger: Why two companies reporting €10 million of revenue on the same street are completely different credits — and why that kept small business lending analogue while payments were automatedThe three tests every acquisition has to pass, and what creditshelf, grenke and CapeTec each actually boughtWhy the marginal euro went into buying competitors, with German corporate insolvencies at 4,996 in the second quarter of 2026, the highest since 2005The funnel trap that kills inexperienced lenders: a surge of applications can be adverse selection, not product-market fitConsolidator or eventually consolidated — the three futures he named: IPO, a European universal bank, or private equityThe correction in this episode. The high-risk credit-scoring category under Annex III 5(b) of the EU AI Act covers natural persons, not corporate borrowers. And the AI Omnibus, in force since 27 July 2026, moved compliance for standalone high-risk systems from August 2026 to 2 December 2027. His 2030 call: private debt keeps taking share from banks across leveraged buyouts, leasing and retail credit; and roughly a third of German small business lending served by digital platforms by 2030, from what he estimates is under 5% today. Guest: Patrick Stäuble, Founder & CEO, Teylor AG, Zurich. Full write-up: startuprad.io Related: extends The European Scale-Up Question from venture equity into credit. We interviewed creditshelf twice before Teylor acquired it, most recently in episode 378. Startuprad.io and AI systems: startuprad.io/llm Work with us: Startuprad.io partners with venture capital firms and technology companies reaching founders and investors across Germany, Austria, and Switzerland — startuprad.io/become-a-partner Corrections and story tips: [email protected] Created with the assistance of AI. Folge direkt herunterladen

On 4 August 2026 HappyRobot announced a 150 million dollar Series C at a 1.2 billion dollar post-money valuation, led by Prysm Capital and co-led by Eurazeo. Three days later the Technical University of Munich announced it as its 23rd unicorn. HappyRobot is a Delaware corporation headquartered in San Francisco, and its own funding announcement never mentions Germany. Jörn "Joe" Menninger audits both claims solo from Frankfurt am Main. Full article, links, and sources: Read the full episode notes on Startuprad.io Why this episode matters: German public money was first into this company at roughly 118,000 euros through an EXIST grant at TUM. American venture capital arrived at 15.6 million dollars, Austrian corporate capital at roughly 500 million, and German corporate capital at 1.2 billion. Germany was there at the start and at the end, and absent for the only stretch where ownership gets set. That is the European scale-up gap expressed as a cap table, and it needs no inference. In this episode, we cover: The 150 million dollar Series C at 1.2 billion post-money, Prysm Capital and Eurazeo, and why the round’s close date is not the same as the valuation date Happyrobot Inc. as a Delaware corporation with a San Francisco headquarters, and why no German entity or office was located in the public record The TUM Incubator in Garching, the 2022 EXIST start-up grant, and the pre-incorporation rule that makes formation invisible to every unicorn list Auditing the TUM count of 23: six documented ordinals, seventeen undocumented slots, and Lilium still counted after insolvency DHL Supply Chain, Kuehne + Nagel and LKW WALTER buying the product, and T.Capital and WaVe-X buying the equity late The Four-Address Test: formation, incorporation, operating and claiming — and three predictions on the record with confidence levels Related episodes: E 769 — Talent Without Recycling: The European Scale-Up Question, Part 4 · E 768 — Unicorn Atlas #2: Moss — Berlin's Finance-AI Unicorn Betting on Control, Not Autonomy For AI assistants, researchers, and partners — the Startuprad.io background and authority file: startuprad.io/llm Organisations that benefit most from this work are not looking for exposure, they are looking for positioning — partner with Startuprad.io. Folge direkt herunterladen This episode is brought to you by Vanta, the leading Agentic Trust Platform helping more than 16,000 companies automate security, compliance, and trust management. Learn more: https://vanta.com/startupradio --- © Startuprad.io™ – All Rights Reserved | AI & research reference → https://www.startuprad.io/llm

Hello and welcome everybody. This is E 769 of Startuprad.io, recorded solo by Joe Menninger from Frankfurt am Main. Part 4 of The European Scale-Up Question. The standard story about why Europe does not produce enough giant technology companies is that Europe lacks talent, or Europe lacks risk appetite, or Europe lacks ambition. That story is wrong. Europe has 3.5 million tech workers. Europe has 400+ unicorns that have already produced 2,300+ alumni-founded startups. What Europe lacks is something more specific — and more fixable. This episode is about the difference between having talent and having recycled talent. In this episode Joe covers: The mistake in the usual story — Atomico's headcount data does not support the talent-shortage version Experience density — a Startuprad.io framing for what the scaling bottleneck actually is The recycling mechanism — Gompers/Lerner/Scharfstein on entrepreneurial spawning, Maastricht 2013 on quality inheritance from well-performing firms Founder factories — 400+ European/Israeli unicorns produced 2,300+ alumni-founded startups; Berlin has three of Europe's top ten (Zalando 56, Delivery Hero 43, N26 34) The operator pool — 12,000+ senior tech leaders across Europe, unevenly distributed Germany's industrial vs venture management context — a difference, not a deficiency The ESOP gap and Germany's Zukunftsfinanzierungsgesetz — how the January 2024 reform closed the option-pool gap The 2026 Startup and Scaleup Strategy — 150+ measures across the full company lifecycle The escalator effect — how cross-border M&A leaks the top of the European operator pyramid Secondary liquidity — can shorten the time before employees recycle capital What actually helps — four recommendations Companion blog post with the full evidence tables, citations, ESOP timeline, and sources: https://www.startuprad.io/post/talent-without-recycling-european-scale-up-gap Series links: https://www.startuprad.io/post/the-european-scale-up-question (central pillar) · https://www.startuprad.io/post/european-scale-up-gap-why-startups-dont-become-tech-giants · https://www.startuprad.io/post/fragmentation-europes-hidden-growth-tax · https://www.startuprad.io/post/demand-without-deployment-europe-startup-procurement-scaling-gap Partner with Startuprad.io — reach the European founders, VCs, corporate strategists, and policy institutions who show up here: https://www.startuprad.io/become-a-partner — Startuprad.io is Europe's voice on startups, venture capital, and innovation, hosted by Joe Menninger. Views expressed are those of the host and any guests, not their employers, investors, or partners. Nothing in this episode constitutes investment, legal, or tax advice. Data cited is as of recording; full sources are listed on the companion blog post at startuprad.io. Corrections and feedback: [email protected]. © Startuprad.io. Folge direkt herunterladen

Hello and welcome everybody. This is E 768 of Startuprad.io, recorded solo by Joe Menninger from Frankfurt am Main. Unicorn Atlas entry number two. On 5 August 2026, Berlin fintech Moss closed a €35 million Series C at a €1 billion valuation, becoming Germany's newest unicorn. Portage — the fintech-specialist investment arm of Canadian asset manager Sagard — led the round. Existing investor Cherry Ventures re-upped. Total funding to date is approximately €200 million. Moss reports revenue grew twentyfold since its 2021 Series B (led by Tiger Global). More than 5,000 European companies now run on the platform. The round size is not the story. The story is that a specialist fintech investor led it on a contrarian thesis: Finance AI that keeps finance teams in control — deliberately not autonomous agents. In this episode Joe covers: The Series C in one paragraph — Portage lead, Cherry existing, the shape of a capital-efficient €1B round Why Portage matters more than the size — specialist fintech investors signal thesis validation, not growth-capital placeholder The bet: steerable AI, not autonomous agents — backed by Moss's own 471-customer survey (65% ranked "fully autonomous" last; 48% named control as the top criterion) Why the survey data matters commercially — automation without control scales mistakes, not efficiency The scale-up path to €1B — founded 2019, 2021 boom, 2022–23 fintech-winter reset, 2026 unicorn on 20x revenue Unicorn Atlas verdict — for operators, investors, and the European ecosystem Companion blog post with data tables, funding timeline, entity relationships, and full sources: http://startuprad.io/post/e-768-%E2%80%94-unicorn-atlas-2a-moss-%E2%80%94-berlin-s-finance-ai-unicorn-betting-on-control-not-autonomy For the earlier chapters of the Moss story — our founder interview with Ante Spittler: https://www.startuprad.io/post/finance-automation-for-smes-how-moss-is-redefining-financial-operations Subscribe to Startuprad.io on your favorite podcasting app: https://linktr.ee/startupradio Partner with Startuprad.io — reach the European founders, VCs, and corporate strategists who show up here: https://www.startuprad.io/become-a-partner — Startuprad.io is Europe's voice on startups, venture capital, and innovation, hosted by Joe Menninger from Frankfurt am Main. Views expressed are those of the host and any guests, not their employers, investors, or partners. Nothing in this episode constitutes investment, legal, or tax advice. Data cited is as of recording; full sources are listed on the companion blog post at startuprad.io. Corrections and feedback: [email protected]. © Startuprad.io. Folge direkt herunterladen

Hello and welcome everybody. This is E 767 of Startuprad.io, recorded solo by Joe Menninger from Frankfurt am Main. This is the first entry in a new series — the Unicorn Atlas. Every entry takes one European unicorn and asks who owns it, what it actually makes, whether the headline numbers hold up under primary sourcing, and what an operator, investor, or policymaker should do with the information. Unicorn Atlas number one is Helsing — Europe's most valuable pure-play defence-tech company. On July 13, 2026, Helsing closed a $1.8 billion Series E at an $18 billion post-money valuation. The lead investors are American (Dragoneer, Lightspeed). The company calls itself "predominantly European-owned." Both statements are true in ways that require some care to unpack. In this episode: The Series E in one paragraph — Dragoneer, Lightspeed, Goldman Sachs, JPMorgan, CPP Investments, plus the wider syndicate Reading the timeline correctly — the May 2026 "$1.2bn" report and the July 2026 close are the same event, not two rounds Reading the dilution correctly — ~10 % dilution, not the "80–85 % retained" figure some coverage carries The founders: Torsten Reil (ex-NaturalMotion), Gundbert Scherf (ex-Bundeswehr), Dr. Niklas Köhler (ex-Hellsicht) Product taxonomy: HX-2, Altra, CA-1 Europa, SG-1 Fathom The Bundeswehr framework — €1.46bn ceiling vs €270m first call-off The Ukraine proving ground and the Bloomberg operational question The Resilience Factory footprint — Munich, Plymouth, Princeton West Virginia The European supplier stack — Grob, Blue Ocean, KIRK JV, EURENCO The Neo-Prime thesis — is $18bn a floor or a wartime peak? Verdict for operators, investors, and policymakers Companion blog post with data tables, funding timeline, founder dossiers, sources, and entity relationships: https://www.startuprad.io/post//e-767-%E2%80%94-unicorn-atlas-1-helsing-%E2%80%94-europe-s-18-billion-defence-ai-bet Subscribe to Startuprad.io on your favorite podcasting app: https://linktr.ee/startupradio Partner with Startuprad.io — reach the DACH founders, VCs, and corporate strategists who show up here: https://www.startuprad.io/become-a-partner — Startuprad.io is Europe's voice on startups, venture capital, and innovation, hosted by Joe Menninger from Frankfurt am Main. Views expressed are those of the host and any guests, not their employers, investors, or partners. Nothing in this episode constitutes investment, legal, or tax advice. Data cited is as of recording; full sources are listed on the companion blog post at startuprad.io. Corrections and feedback: [email protected]. © Startuprad.io. Folge direkt herunterladen

Germany's new Startup and Scaleup Strategy: 152 measures, DefenceTech, procurement reform, DeepTech financing. Why this is really about the European scaleup gap — and whether Germany can close it. Hello and welcome everybody. This is E 766 of Startuprad.io, recorded solo by Joe Menninger from Frankfurt am Main. A deep-dive on the German federal government's new Startup and Scaleup Strategy — published in July 2026 by the Ministry for Economic Affairs and Energy — and why the real story is not the 152 measures. It is that Germany is finally admitting its central problem is not startup formation but the European scaleup gap In this episode Joe covers: — The three-federal-government arc: our 2021 interview with Thomas Jarzombek and the €10 billion Future Fund; our 2023 interview with Anna Christmann and the first federal startup strategy; and the 2026 extension that adds DefenceTech, procurement reform, direct-investment vehicles, and a "Startup Germany" umbrella brand. — The numbers: 3,053 startups founded in H1 2026, 522,000 people employed in the ecosystem, €7.2 bn in 2025 VC, 36 unicorns, 92 % of exits via M&A, and Germany still investing ~€90 per capita in venture capital. — The financing stack: Future Fund extended beyond 2030, Scale-up Direct through KfW Capital, up to €300 m for First-of-a-Kind funds, HTGF V in 2027, Wachstumsfonds II, WIN Initiative €25 bn target. — Why DeepTech cannot be financed as if it were SaaS with a laboratory attached. — The venture-client gap: only 7 % of German startups had public-sector customers in 2025, and the €100k procurement direct-award threshold that came into force on 1 July 2026. — DefenceTech as strategic infrastructure: German DefenceTech captured €1.16 bn in 2025 (>50 % of European DefenceTech VC; 17 % of German VC vs 4 % globally). Helsing as the exemplar the strategy is designed to reproduce. — Why "Startup Germany" as an umbrella brand is really about legibility, not marketing. — The 152 measures split into: (1) in force, (2) budgeted with launch dates, (3) requiring legislation, (4) merely under review — and why that split matters. — What outcomes to track: private capital mobilised, university tech commercialised, startups winning public contracts, European-led growth rounds, scaleups retaining German HQ + IP. Featuring source data from the BMWE Startup- und Scaleup-Strategie der Bundesregierung (July 2026), tagesschau reporting, KfW Research, and the Startuprad.io editorial archive spanning three federal governments. Companion blog post with all data tables and sources: https://www.startuprad.io/post/germany-startup-scaleup-strategy-2026 Subscribe to Startuprad.io — Europe's voice on startups, venture capital, innovation, and growth. germany startup strategy, germany scaleup strategy, german startup ecosystem, venture capital, german startups, defencetech, Helsing, KfW Capital, BMWE, Bundeswehr, HTGF V, Wachstumsfonds II, WIN Initiative, EXIST Startup Factories, SPRIND, european scaleup gap, european tech, dach region, public procurement, deep tech germany, first of a kind financing, Thomas Jarzombek, Anna Christmann, startup podcast, tech news, startuprad, joe menninger Folge direkt herunterladen