Bernard Hazard

Excited to announce a strategic partnership between Initial Data Offering and Carbon Arc. This collaboration significantly expands the mission at InitialDataOffering: helping data vendors and asset owners reach more buyers — and helping buyers discover differentiated, high-quality data assets – faster. Through this partnership: • IDO and Carbon Arc will jointly offer API access to data assets across our platforms, creating a single point of entry for buyers seeking quality data • IDO data asset announcements and our full historical catalogue, will be integrated directly into the Carbon Arc platform via the newly created API • Data vendors and asset owners who publish to IDO will have the opportunity to monetize their data through Carbon Arc to its network of financial institutions, Fortune 500 companies, hedge funds, high-growth startups, individual investors, and small- & medium-sized businesses For the data ecosystem, this is about distribution and discovery at scale. Carbon Arc continues to build one of the most forward-thinking data platforms in the industry, and integrating IDO’s announcement layer creates a powerful bridge between new data supply and real buyer demand. If you're a data vendor looking for broader distribution — or a buyer who wants earlier visibility into new data sets — send me a DM and I’ll connect you with the team.

114

3 Comments

The most dangerous number in ETA is EBITDA. Not because it is wrong, but because it exists in multiple, conflicting versions. Most searchers underestimate how destructive this is. Every deal ends up with different EBITDA figures circulating among sellers, bankers, lenders, QoE firms, and LPs. That chaos does three things: 1. Slows conviction. 2. Inflates valuation traps. 3. Creates mistrust inside the deal team. Here is the real ETA math most people never talk about: Seller EBITDA. Broker EBITDA. Searcher EBITDA. QoE EBITDA. Lender EBITDA. Post-close EBITDA. Six versions. All defensible. Only one actually matters. And when people operate off different versions, underwriting turns into narrative, not analysis. You think you have a 4.9x deal, the lender is modeling 6.2x, and LPs are trying to guess who is closest to reality. EBITDA confusion is not a technical nuisance. It is valuation risk, trust risk, and closing risk. The most sophisticated buyers maintain one continuously updated EBITDA line from IOI to closing, with full visibility on adjustments, timing, and assumptions. Everyone sees the same number and the same reasoning. It removes noise and forces truth. If you are a searcher or LP working through this pain, and want to see the infrastructure I am building to eliminate this problem, reach out.

4

5 Valuation Takeaways Every VC Firm Should Re-Anchor On After digging into the most common valuation shortcuts in venture capital, a few clear truths emerge: 1️⃣ Cost is not fair value Cost is historical. Fair value is current. IPEV and ASC 820 are explicit: once new information exists, holding at cost stops being conservative and starts being misleading. 2️⃣ Last price per share ≠ portfolio truth LPPS reflects one deal, at one point in time. It ignores preferences, protections, optionality, and changes in performance or market conditions across the cap table. 3️⃣ Waterfalls are not valuations (except at exit) “EV then waterfall” only works when an exit is imminent. Outside of that, it zeroes optionality, overweights later rounds, and fails to reflect market participant assumptions. 4️⃣ Optionality matters Venture equity isn’t binary. Junior and common shares are not worthless just because they’re out of the money today. Fair value must reflect probability-weighted outcomes. 5️⃣ Clarity yields confidence Cap tables are no longer static records. When ownership, rights, and math work together, firms gain defensible valuations, stronger negotiations, and fewer surprises at exit. Fair value is about being accurate, credible, and aligned with how sophisticated market participants actually think. #venturecapital #vc #investing #waterfalls #valuations #ipev

4

1 Comment

Every VC want to invest in the next Airbnb, but would your fund have actually said yes? This was Airbnb when Paul Graham said yes: 1. 𝗔𝗹𝗺𝗼𝘀𝘁 𝗻𝗼 𝘁𝗿𝗮𝗰𝘁𝗶𝗼𝗻: After 2 years, it was making $200/week. 2. 𝗠𝗮𝗿𝗸𝗲𝘁 𝗹𝗼𝗼𝗸𝗲𝗱 𝘁𝗶𝗻𝘆: Renting a stranger's spare room didn't seem large or scalable. 3. 𝗧𝗿𝘂𝘀𝘁 𝗰𝗼𝗻𝗰𝗲𝗿𝗻𝘀: Most people didn't believe travellers would trust strangers with their safety. 4. 𝗣𝗮𝘆𝗺𝗲𝗻𝘁𝘀 𝗼𝘂𝘁𝘀𝗶𝗱𝗲 𝘁𝗵𝗲 𝗽𝗹𝗮𝘁𝗳𝗼𝗿𝗺: Early Airbnb hosts often handled payments manually or off-platform, making the model look unprofessional and hard to scale. 5. 𝗥𝗲𝗴𝘂𝗹𝗮𝘁𝗼𝗿𝘆 𝘂𝗻𝗰𝗲𝗿𝘁𝗮𝗶𝗻𝘁𝘆: Short-term rentals sat in a legal grey zone in many cities. 6. 𝗡𝗶𝗰𝗵𝗲 𝗲𝗮𝗿𝗹𝘆 𝘁𝗿𝗮𝗰𝘁𝗶𝗼𝗻: Low revenues coming from specific events (like conferences), making it look like a temporary solution. 7. 𝗡𝗼𝗻-𝗼𝗯𝘃𝗶𝗼𝘂𝘀 𝗳𝗼𝘂𝗻𝗱𝗲𝗿𝘀: Young, non-hospitality founders with no traditional industry credentials. 8. 𝗟𝗼𝗼𝗸𝗲𝗱 𝘁𝗼𝗼 𝗰𝘆𝗰𝗹𝗶𝗰𝗮𝗹: Early demand was seasonal and event-driven, with no recurring revenue to give confidence in long-term growth. Would your VC fund pass on this opportunity? If yes, what in your analysis needs to change?

7

Three Series A founders asked me the same thing this week: what actually drives value right now? Here’s the short CFO answer. 1) Growth efficiency. Focus spend on your top two channels and turn the rest off. Aim CAC payback ≤ 12 months. Reallocate every two weeks on cohort LTV/CAC, not clicks. 2) Unit economics. Cut COGS before you raise prices. Lock annual cloud/model commits, kill idle workloads, consolidate tools (target COGS −20–30% and Burn Multiple < 1.5) 3) Cash & working capital. Default to annual prepay and milestone billing, keep AR < 30 days, and hold 12–18 months runway. Prepayment shortens payback and buys negotiating power. If a project doesn’t move one of these three needles, pause it. What’s your #1 lever to improve one of these metrics this quarter? #cac #burnmultiple #runway #uniteconomics #financeforfounders #valuecreation #gtmstrategy #scaleefficiently

6

AI inside the fund: hype vs leverage Half of all VC dollars now go to AI — but inside most funds, the real transformation is just starting. While the world chases the next foundation model, CFOs and back offices are quietly turning reporting, portfolio tracking, and LP updates into prompt-driven workflows. And yet — how many funds are really using it today? The irony is that the many of the firms fueling the AI boom are still running on spreadsheets. Automation is here, but adoption is not. True leverage won’t come from chasing the next model, but from rethinking how a fund operates — how data moves, reports close, insights scale, and how the ivestments teams scrap the market. Because when the hype fades, only leverage remains.

11

1 Comment

AI is reshaping every stage of the M&A lifecycle, from unlocking deeper & faster diligence to screening targets at scale. Firms using AI meaningfully in their M&A processes report ~20% cost reductions and 30–50% faster deal cycles. (McKinsey 2026) Here is a holistic view of where AI is being used & the software I have found to do this: 1. Deal sourcing (Grata, Blueflame AI) AI scans private markets, filings, news, and signals to surface targets at scale. Deal teams can now screen hundreds of companies in hours rather than weeks. 2. Financial analysis (Pigment) AI accelerates modeling, scenario testing, and memo drafting. The value is not replacing judgment. It removes the mechanical work so analysts can focus on decision-making. 3. Due diligence This is where I believe AI is having the biggest impact. – Contract AI extracts clauses and flags legal risks across thousands of documents. (Litera) – Data room AI indexes files, answers Q&A, and compresses diligence timelines. (DealRoom) – Behavioral AI measures execution risk by analyzing leadership patterns and decision behavior. (Humanaq) 4. Integration planning (Ansarada, Midaxo, BowMerge) AI centralizes integration workstreams, automates milestone tracking and models operational scenarios. 5. Post-close monitoring (AlphaSense, Humanaq) AI tracks performance against the deal model and surfaces early warning signals before value leakage becomes visible in financial reporting. 6. Strategic orchestration (Tiger Team M&A) Strategic orchestration tools help anchor sourcing, diligence, integration, and monitoring to the core value creation thesis of the deal, ensuring that faster analysis still serves consistent strategic decision making across the entire M&A lifecycle. AI does not replace dealmakers. It removes friction from the process and makes execution visibility possible earlier. As deals become more operationally complex, this visibility is becoming a real competitive advantage in M&A. How do you use AI in the deal lifecycle?

34

21 Comments

Most venture capital frameworks are calibrated to measure first-order effects—the immediate, visible pulses of a business. While these metrics signal activity, they often mask hidden structural risk and value leakage. The First-Order Trap Investors typically optimize for short-term efficiency and "growth-at-all-costs." They ask: "How much does this reduce Customer Acquisition Cost (CAC)?" "What is the immediate ROI on this isolated software tool?" "How quickly can this reach $10M ARR?" This approach leads to a portfolio of fragmented vendors. While these companies may scale revenue, they fail to scale system coherence. As these disconnected systems grow, unowned coordination risk compounds, eventually compressing margins and capping multiple expansion. The Second-Order Framework: Engineering Durability At Second Order Ventures, we reject "narrative momentum" in favor of Infrastructure Control. We ask the questions that determine long-term capital efficiency and market dominance: Infrastructure Ownership: Does this technology convert industry coordination from an unpriced risk into a controlled, compounding asset? Margin Compounding: How do portfolio-level shared data and governance improve unit economics over time? Governance as a Return Driver: How does structural compliance reduce volatility and preserve exit optionality in regulated markets? The Shift from Products to Ecosystems Durable advantage is not found in experimental products, but in the foundational layer that industries depend on. When you own the infrastructure—Communication, Data, and Governance—you own the "Data Gravity." Second-order thinking isn’t just about being smarter; it’s about owning the layer where consequences compound. By the time a first-order gain plateaus, a second-order advantage has already built a moat that isolated vendors cannot replicate. The Mandatory Check: Does your portfolio look like a collection of fragmented vendor solutions, or a unified infrastructure play? What first-order metrics are currently blinding investors to the structural risks in your industry? #VentureCapital #CapitalEfficiency #InfrastructureControl #SecondOrderThinking #StructuralAdvantage #PrivateEquity #UnitEconomics

5

Meta's 2026 capex outlook also could provide a starting point for framing 2027 expectations. Even if capex growth moderates in 2027, D&A expense is likely to grow faster than revenues due to a lagged impact of prior-year investments, creating less near-term flexibility in the cost structure. For 2027, we estimate capex of $129bn (+$16bn y/y or 14% y/y) and D&A to increase to ~$46bn (up 53% y/y), well ahead of expected revenue growth of ~19% y/y. We estimate D&A as a percentage of total expenses to increase from ~16% in 2024 to ~24% by 2027. This mix shift could create a sustained margin headwind, particularly if AI-driven revenue monetization ramps more gradually than infrastructure-related depreciation.

1