WHAT YOU GIVE AIRTIME TO WILL EXPAND
Communication is attention management—mention constraints briefly, then immediately pivot to solutions.
Full-text search across article titles and summaries. Refine with sector, year, content type, and reading-time filters.
Communication is attention management—mention constraints briefly, then immediately pivot to solutions.
Forerunner, known for investments in Oura and Chime, anticipates future growth in the AI era.
Flexport CEO Ryan Petersen discusses the impact of new U.S. tariffs on global trade.
Listen Labs revolutionizes market research with an AI-powered platform, allowing businesses to understand customers in depth and scale. Sequoia supports Listen's mission to automate and streamline customer research processes, significantly reducing time and costs. Listen has already conducted over 300,000 interviews for major companies like Microsoft and Canva, demonstrating its potential to transform product development across various industries.
OpenAI announced its restructuring as a public benefit corporation with the nonprofit that controls it remaining its largest shareholder. This reverses earlier plans to shift control to investors, which had drawn criticism from Elon Musk (who sued the company) and regulatory scrutiny from California and Delaware attorneys general. While Sam Altman called the change a "more understandable structure," Musk's lawyer dismissed it as a "transparent dodge" that still betrays the company's mission.
DoorDash just splurged $5 billion in five hours, buying SevenRooms ($1.2B) and Deliveroo ($3.9B). This massive spending spree transforms it from US delivery leader to global hospitality tech giant with access to 50 million monthly users across 40+ countries.
Most founders admit their early pricing is essentially a guess. Focus on learning rather than maximizing revenue before finding product-market fit. Freemium converts only 2-5% of users but can drive massive adoption, while paid pilots convert an impressive 60-90% of customers. Your pricing sends strong signals about your positioning - Superhuman launched at $30/month to appear premium, while Atlassian used rock-bottom pricing to spread virally through organizations. Treat early pricing as an experiment, run small tests, and be ready to adjust as you learn what your market actually values.
Mach CEO Ethan Thornton discussed his company's recent U.S. military contract and plans for cross-functional factories at the StrictlyVC 2025 event. Thornton aims to compete with China in weapons systems development.
1984 Ventures has launched an incubator program for Rippling alumni with checks between $500K-$1.2M.
Build any mobile app fast.
BASE44 is an all-in-one no-code platform for building software, from personal apps to complete online businesses, with optional coding flexibility.
Counterintuitively, focusing on a narrow audience creates more value than chasing mass appeal. By diving deep into tech CFO topics that seem "too niche," the author built a podcast that just hit “stadium level” capacity in downloads. Your staying power is the key – consistently delivering insider knowledge that resonates so deeply with your core audience that adjacent groups want in too. When you speak to the core loudly enough, everyone else starts listening.
Secondary sales give employees and investors liquidity in today's slow IPO market, but they come with a significant hidden cost. These transactions raise your company's 409A valuation by providing evidence of common stock value, which increases the exercise price for all future option grants. One executive found secondaries added $0.50 to his option price, costing him an extra $1M on his 2M share grant. When you want to sell shares, secondaries seem great. When you join a company and get options, they reduce your upside. Most employees never know these transactions happened, but they might be silently erasing millions in potential gains.
The seed ecosystem has rapidly expanded, with more funds and faster deal closing, but standing out as a VC firm remains challenging. Many successful seed funds are increasing in size, leading to larger checks and fewer diverse perspectives in funding rounds. Liquidity opportunities have grown with more secondary market activity, while some funds are adjusting their reserve strategies due to abundant capital.
"Safe Hands" people transform how founders scale their companies – these are the team members you can delegate to without worry knowing they'll execute with quality, communicate proactively, and solve problems independently regardless of their title or experience.
Most venture capital funds can't be judged on performance for nearly a decade, leaving reputation to fill the gap.
Google Cloud, under CEO Thomas Kurian, attributes its recent 30% growth in quarterly revenue partly to the adoption of generative AI, leveraging in-house capabilities like DeepMind and over 200 models.
As thousands of "tweener" startups find themselves stranded in today's funding environment, a new class of hybrid funds is emerging to fill the gap. These funds target companies too mature for venture capital but not profitable enough for private equity, buying them at reduced valuations and implementing operational discipline. The strategy resembles distressed-asset arbitrage with modern branding. These funds often launch separate vehicles rather than repurposing existing funds to maintain clean track records with investors.
With $300M ARR and a rumored $10B valuation, Cursor's growth raises the question: how defensible is its position? The AI-powered coding tool faces significant long-term challenges despite early traction. The underlying LLM technology is rapidly commoditizing, while giants like Microsoft and GitHub control both distribution channels and vastly superior data for training models. Cursor's high valuation assumes it will expand beyond an IDE into a full development platform, but its data advantage remains tiny compared to GitHub's global repository corpus.
Why do customers praise your product but never purchase? This frustrating gap between enthusiasm and action signals "product purgatory"—your solution may be valuable but lacks urgency. The insightful "Magic-Wand Test" reveals whether your product fails due to perceived value or activation hurdles. If prospects would take it fully implemented for free yet still won't buy, your challenge isn't the product but its priority level (usually #7 on their list). The solution isn't to add features, but to narrow your focus to customers facing immediate catalysts like regulatory deadlines or leadership mandates where urgency drives decisions.
Cofounders often split before their company succeeds. To prevent this, share equity equally no matter who has more experience, write down your good and bad feelings weekly, share rooms on business trips to talk more, and work on what each of you cares about rather than sticking to rigid job titles. Start meetings with opportunities, not just problems, and when you disagree, solve it by writing down principles you both believe in.
Design gets easier when you follow simple rules instead of starting from scratch every time. Some useful ones: stick with patterns users already know, keep choices to seven or fewer, make response times under 400ms, and focus effort on what makes your product special while using standard designs for everything else. These shortcuts help you make good decisions faster without getting stuck in endless debates.
Epiphany turns voice notes into instant actions in tools like Notion, Asana, and Todoist.
Intelswift automates customer support with AI chatbots, Live Chat, and AI Copilot for faster, smarter service.
Credal deploys secure AI agents that collaborate, automate workflows, and act across enterprise tools while maintaining data permissions.
A disturbing University of Zurich study, which used year-old models that weren't cutting-edge, found AI responses on Reddit were 3-6 times more persuasive than human posts. While AI companies focus on technical "alignment," the real danger may be systems that evolve to subtly convince us of almost anything through rhetorical techniques we can't detect, similar to how social media platforms initially promised connection but ultimately optimized for engagement that polarized society.
The structure of venture capital compensation creates potential misalignment when early winners mask later failures. Clawback provisions require VCs to return excess carried interest if the final fund performance falls short, aligning long-term interests between investors and managers. The strength of these provisions—including timing tests, personal guarantees, and tax considerations—signals how seriously GPs take partnership with their LPs. Beyond legal mechanics, robust clawbacks reflect a cultural commitment to shared outcomes rather than opportunistic profit-taking.
Online, the loudest people often seem right even when they're wrong. We see big follower counts and think they must know what they're talking about. However, being loud doesn't make someone right - it just makes them heard. While we can't fix this system where volume beats accuracy, we can at least try to stay honest ourselves while playing this game.
What we perceive as "luck" often boils down to simple mathematical consistency.
Stripping away venture capital jargon reveals a simpler truth: companies exist in either a "risk" stage (seeking breakthrough outcomes) or "scale" stage (predictable growth).
Klarna and StubHub are pausing their anticipated IPOs due to market volatility from recent tariff announcements.
When Amazon briefly considered showing tariff costs on product pages then quickly reversed course after a presidential phone call, it revealed our new economic reality: a stage where perception trumps market principles. This post unpacks this "economic surrealism," where corporate power bows to government intervention, AI flatters rather than informs us, and traditional economic relationships break down. For founders, this means preparing for inflation over recession, investing in companies tied to physical assets, and watching narratives closely—because in this environment, the story often matters more than the fundamentals.
The protective moats that once shielded startups have vanished. Cloud infrastructure costs pennies, no-code tools eliminate technical barriers, and global talent pools make team-building accessible to anyone. Today we face instant competition the moment we show traction. Remember, difficult is defensible. Try your best to invest in building proprietary technology, creating unique distribution channels, or serving overlooked markets.
Mark Zuckerberg discussed Meta's ongoing evolution, highlighting how AI and open-source models like Llama fit into Meta's broader business strategy. He emphasized Meta's focus on integrating AI to enhance advertising, grow consumer engagement, and support business messaging, asserting the long-term value it brings in spite of substantial training costs. Zuckerberg also touched on the future potential of AR glasses and the vision of combining digital and physical realms under the Meta brand.
We all fall prey to building elaborate product roadmaps, only to watch reality tear them apart. Stop assuming you know exactly what to build and adopt traction roadmaps instead. The GO-LEAN framework creates immediate focus: identify your funnel's biggest bottleneck, direct 80% of your effort there, run experiments, analyze what worked, and then move to the next constraint.
The void between your past and future holds power, whether you chose this transition or it crashed into you uninvited. Most founders rush to fill this space, but intentional reflection during transitions helps you reconsider identity and reimagine possibilities. The Revolve framework (Reset + Evolve) turns these moments from crises into opportunities.
Stripe now lets iOS developers redirect users to a secure Stripe-hosted payment page outside the app for purchasing digital goods. The integration supports creating products, managing customers, and using universal links to return users to the app after payment. Developers handle order fulfillment via Stripe webhooks, enabling a seamless payment experience without native in-app purchases.
timeOS 3.0 provides AI-powered summaries of calls and meetings across platforms like Zoom, Google Meet, and WhatsApp. It now has a macOS app and team workspace.
Superlines tracks your brand's visibility across AI chats and helps improve it.
Codegen helps engineering teams analyze and improve codebases 24/7 via Slack.
B2B marketing doesn't have to be complicated. Start with the basics: define your market, craft a unique message, create content mechanisms that drive demand, and develop processes to nurture leads before passing them to sales. However, beware of the ABM trap—targeting specific accounts doesn't mean abandoning broader marketing efforts. The buying journey inevitably includes research and peer recommendations, so you still need people "bumping into you all over the place." Success requires balancing account-targeted activities with building a broader presence that establishes credibility across your entire market.
Product leaders often split into two camps—pragmatic pirates who focus on delivering customer value versus those who invest in broader organizational health. Pirates drive innovation with customer focus, but someone must eventually treat the company itself as a product worth improving. Your startup needs both energies: the push to build great products and the systems where multiple teams can thrive.
South Park Commons has announced its $275M third fund. It previously backed companies like Render, Baseten, Luma AI, Gamma, and Goodfire.
Conventional wisdom warns founders against mixing business with personal relationships, but Patrick Collison found success at Stripe with his brother John and at Arc Institute with his wife.
ChatGPT's sudden switch to flattering users reveals a truth about AI persuasion: small personality tweaks dramatically reshape conversations and influence behavior.
Kalshi, the largest U.S. prediction market, is viewed by some as illegal gambling and by others as a financial opportunity.
Sponsored
An MCP should make it easy to give your AI agent tools.