Latest / Best of LinkedIn: Strategy, M&A & Investments / Best of LinkedIn: Venture Capital CW 13/ 14

Transcript

Best of LinkedIn: Venture Capital CW 13/ 14 · Best of LinkedIn: Strategy, M&A & Investments

• 0:00
  Speaker 0: Provided by Thomas Allgeier and Frennus, based on the most relevant LinkedIn posts about venture capital from CWP.
• 0:05
  Speaker 0: thirteen and fourteen.
• 0:07
  Speaker 0: Frenness supports general partners in identifying relevant limited partners across multiple sources researching connection strategies coordinating event attendance And running structured outreach campaigns that turn cold lists into scheduled conversations and committed capital.
• 0:22
  Speaker 0: You can find more info in the description.
• 0:24
  Speaker 1: Imagine raising a hundred twenty two billion dollars.
• 0:28
  Speaker 1: Not not for whole fund you know a single company.
• 0:31
  Speaker 1: Yeah, that is nearly double the entire venture capital output of Today.
• 0:38
  Speaker 1: we are looking at a market that hasn't just shifted.
• 0:40
  Speaker 1: It's been completely distorted by forces, they're rewriting the rules of how capital is deployed.
• 0:46
  Speaker 0: and That is exactly the mission of this deep dive.
• 0:48
  Speaker 0: We are extracting The absolute highest signal trends from the venture capital M&A in private equity conversations that have Been happening across linkedin over ten meter weeks.
• 0:57
  Speaker 0: thirteen and fourteen?
• 0:58
  Speaker 0: We want to go way beyond the headlines you know really unpacked the underlying mechanics Of what is actually happening out there right now because if you are an LP, a GP or even a founder trying to navigate this landscape.
• 1:11
  Speaker 0: The strategies that worked twelve months ago they're effectively obsolete totally
• 1:15
  Speaker 1: obsolete.
• 1:16
  Speaker 1: I mean we are seeing a complete re-architecture of the ecosystem...the data points we pulled from recent industry discussions really highlight four massive shifts.
• 1:25
  Speaker 1: okay
• 1:26
  Speaker 0: let's lay them out.
• 1:27
  Speaker 1: so we've got AI consolidating in the global capital stack Europe's structural fight to basically own deep tech, the professionalization of the fundraising funnel and finally this really rapid evolution of venture capital into a software-driven solar GP model.
• 1:44
  Speaker 0: Yeah Let's dive straight into the numbers for that first one, because they are literally bending the gravity of the entire market.
• 1:50
  Speaker 0: Louis LaHote shared some data on Q One.
• 1:52
  Speaker 0: twenty-twenty six.
• 1:53
  Speaker 0: That just it completely re anchors our understanding.
• 1:56
  Speaker 0: a venture scale.
• 1:57
  Speaker 0: We're talking about three hundred billion dollars deployed globally
• 2:00
  Speaker 1: which breaks all previous venture records by the way.
• 2:02
  Speaker 1: Yeah
• 2:02
  Speaker 0: exactly but the headline number isn't even the real story.
• 2:06
  Speaker 0: The story is the concentration A staggering eighty percent of that, so two hundred and forty-two billion dollars was deployed exclusively into AI startups.
• 2:16
  Speaker 0: Exclusively?
• 2:17
  Speaker 1: Oh!
• 2:17
  Speaker 1: Which is just wild... And it brings us back to the figure highlighted by Andre Krasienescu regarding OpenAI's recent round.
• 2:24
  Speaker 1: Right
• 2:24
  Speaker 0: The one hundred twenty-two billions.
• 2:26
  Speaker 1: Yeah When you have a single entity absorbing twice the capital in the entire European venture market at once swing.
• 2:34
  Speaker 1: You're no longer looking at a sector trend, you are looking at fundamental rerouting of global liquidity.
• 2:40
  Speaker 0: Yes like supermassive black hole and it isn't just the mega cap foundational model sucking up all oxygen either.
• 2:46
  Speaker 0: Peter Walker pointed out that this cascades down to the earliest stages At seed stage.
• 2:51
  Speaker 0: right now, seventy percent of software VC dollars is going strictly to
• 2:55
  Speaker 1: AI.
• 2:55
  Speaker 1: Seventy percent Right
• 2:57
  Speaker 0: but let's actually break why this happening?
• 2:59
  Speaker 0: because too easy to wave your hands called as hype cycle If you are a GP right now, AI isn't just the technology trend.
• 3:08
  Speaker 0: It operates almost like an index fund for your job
• 3:11
  Speaker 0: security.".
• 3:12
  Speaker 1: Oh that is really critical way to frame it!
• 3:14
  Speaker 1: VCs are pouring money into these rounds at valuations which completely defy traditional mathematical multiples and they're doing this because of its ultimate consensus trade.
• 3:25
  Speaker 0: Explain more about that for our listeners.
• 3:27
  Speaker 1: Well think about if you back up a major AI player.
• 3:30
  Speaker 1: You don't get fired by your OPs, right?
• 3:32
  Speaker 1: Because every top-tier firm made the exact same bet.
• 3:35
  Speaker 0: Right you're insulated by The Herd!
• 3:37
  Speaker 1: Exactly
• 3:37
  Speaker 0: yeah.
• 3:38
  Speaker 1: but on the flip side if you ignore AI and you missed the platform shift of the decade Your career is over.
• 3:44
  Speaker 1: it's essentially a risk transfer mechanism that's disguised as high conviction investing
• 3:49
  Speaker 0: Which raises a massive strategic question for you as a listener, especially if you are on the LP side or managing a corporate M&A strategy.
• 3:56
  Speaker 0: If eighty percent of the capital is chasing consensus deals paying massive premiums and competing against literally everyone else?
• 4:03
  Speaker 0: Yeah
• 4:04
  Speaker 1: Aren't the truly outsized returns hiding in the contrarian twenty-percent like The sectors that Everyone Else Is Ignoring And Are Completely Starved Of Capital Right Now?
• 4:13
  Speaker 0: I mean historically Yes!
• 4:15
  Speaker 0: The Alpha It's Almost Always Found Where Capital Is Scares.
• 4:18
  Speaker 0: And this dynamic explains exactly what is happening across the Atlantic right now.
• 4:22
  Speaker 1: In Europe?
• 4:23
  Speaker 0: Right, while US venture capital is operating like a heat-seeking missile locked onto AI software... ...Europe is aggressively attempting to position itself as the contrarian stronghold for hardware and deep
• 4:36
  Speaker 1: tech.".
• 4:37
  Speaker 1: The numbers back that up!
• 4:38
  Speaker 1: Jan Lozak and Ingrid Kovacic recently broke down data from the new European Deep Tech report….
• 4:44
  Speaker 1: …and the shift in focus undeniable.
• 4:46
  Speaker 0: Oh, it's huge!
• 4:47
  Speaker 1: European deep tech is now valued at six hundred and ninety billion dollars.
• 4:51
  Speaker 1: It is capturing thirty two percent of all VC funding in the region
• 4:55
  Speaker 0: Which just for context?
• 4:56
  Speaker 0: Is a massive jump from just fifteen percent A decade ago.
• 4:59
  Speaker 0: The intellectual property being generated In Europe right Now around robotics advanced compute defense Tech in its world class...it really is.
• 5:07
  Speaker 0: But Lawsack highlights a glaring physical vulnerability that threatens to derail All this momentum.
• 5:12
  Speaker 0: DeepTech doesn't run on code alone.
• 5:14
  Speaker 0: It runs on electricity.
• 5:16
  Speaker 0: Yeah, the
• 5:16
  Speaker 1: physical infrastructure constraint
• 5:17
  Speaker 0: Exactly Whether you are building sovereign AI data centers or scaling automated manufacturing You need gigawatts of power.
• 5:25
  Speaker 0: Yet Europe still imports fifty-seven percent Of its energy
• 5:29
  Speaker 1: And in industrial hubs like Germany That number jumps to sixty seven percent.
• 5:33
  Speaker 1: Right
• 5:34
  Speaker 0: You cannot build the infrastructure for next industrial era if you don't control molecules required to turn lights on.
• 5:41
  Speaker 0: If a geopolitical shark disrupts that energy supply, your deep tech scaling capacity evaporates practically overnight.
• 5:48
  Speaker 1: It's huge blind spot.
• 5:49
  Speaker 1: but power grid isn't even the only structural bottleneck.
• 5:53
  Speaker 1: we have to look at mechanics of capital itself.
• 5:55
  Speaker 1: Oh!
• 5:55
  Speaker 0: The government LP issue?
• 5:57
  Speaker 1: Andreas Schmidt shared a rather sharp critique pointing out that thirty-seven percent of VC funding in Europe last year came from government LPs.
• 6:07
  Speaker 0: Which, you know on paper?
• 6:08
  Speaker 0: Government subsidies sound great.
• 6:10
  Speaker 0: it sounds like a safety net for the ecosystem!
• 6:12
  Speaker 1: On paper sure but in practice they create a highly distorted incentive structure.
• 6:19
  Speaker 1: Schmidt argues that pumping public money into venture doesn't magically stimulate private LP demand.
• 6:25
  Speaker 1: What it actually does is prop up these zombie funds.
• 6:28
  Speaker 0: Zombie fund?
• 6:28
  Speaker 0: Yeah,
• 6:29
  Speaker 1: firms that just wouldn't survive in a purely free market.
• 6:32
  Speaker 0: Let's actually explain why private LPs view heavy government involvement as a massive red flag.
• 6:38
  Speaker 0: When a sovereign wealth fund or national development bank anchors of venture fund That money usually comes with intense geographic deployment mandates.
• 6:46
  Speaker 0: right
• 6:46
  Speaker 1: exactly.
• 6:47
  Speaker 0: the GP Is essentially forced to invest a certain percentage of the fun into startups headquartered In very specific regions Regardless of whether those are genuinely the best companies, it artificially restricts the fund's ability to hunt for pure alpha.
• 7:00
  Speaker 0: Right
• 7:00
  Speaker 1: and furthermore when these government subsidized funds write checks to local startups It inflates early stage valuations without providing any than necessary late-stage growth capital.
• 7:10
  Speaker 0: because The Manday is mostly focused on Early Stage job creation or innovation signaling.
• 7:15
  Speaker 1: Exactly so you end up with a high volume of overvalued seed companies all fighting For A tiny pool Of domestic Series B Capital Down
• 7:23
  Speaker 0: the Road And friction of operating across European borders just exacerbates this whole mess.
• 7:29
  Speaker 0: Patricia Borlivan, Kristoff Nauer and Gad Weiss all weighed in on a proposed regulatory fix called EU Inc..
• 7:36
  Speaker 1: Yeah!
• 7:36
  Speaker 1: This is super interesting...
• 7:37
  Speaker 0: The goal to create digital first pan-European legal entity.
• 7:42
  Speaker 0: they're essentially trying build the Delaware of Europe.
• 7:44
  Speaker 1: ...and the intent there is vital right now.
• 7:47
  Speaker 1: cross border scaling in europe is an operational nightmare.
• 7:50
  Speaker 1: give
• 7:50
  Speaker 0: us example.
• 7:51
  Speaker 1: Well, try recruiting a senior AI engineer in Paris to work for a Berlin-based startup.
• 7:56
  Speaker 1: You have to navigate two entirely different national tax regimes.
• 8:01
  Speaker 1: just issue stock options.
• 8:02
  Speaker 1: Fragmented ESOP structures basically destroy talent mobility.
• 8:06
  Speaker 0: Yeah that's brutal.
• 8:07
  Speaker 1: So EU Inc.
• 8:08
  Speaker 1: aims to standardize share classes and warrant mechanics so the start up can operate seamlessly across the entire continent.
• 8:14
  Speaker 0: Okay but let me push back on whether this actually solves the core problem.
• 8:18
  Speaker 0: The data clearly shows that seventy percent of late-stage deep tech funding in Europe still comes from non European investors.
• 8:25
  Speaker 0: True!
• 8:25
  Speaker 0: If EU Inc successfully streamlines the legal framework but Europe still lacks domestic private growth capital, aren't we just building a much shinier highly efficient runway for American capital to land acquire the best companies and extract value back to US?
• 8:43
  Speaker 1: I mean yeah.
• 8:43
  Speaker 1: Gad Vice made very similar point he noted.
• 8:46
  Speaker 1: While standardizing legacy corporate law is definitely a step forward.
• 8:50
  Speaker 1: It doesn't solve that late stage liquidity vacuum, right?
• 8:53
  Speaker 1: Europe has effectively just subsidized in the early-stage R&D for technologies that will eventually be owned by US private equity and late-stage venture firms.
• 9:01
  Speaker 0: which means if European founders can't rely on domestic growth capital And us founders are competing in a market where eighty percent of dollars or being hoovered up by AI The actual mechanics of fundraising they have to fundamentally change.
• 9:13
  Speaker 1: Oh, absolutely.
• 9:14
  Speaker 1: The days of prestige hunting are completely over.
• 9:17
  Speaker 1: Alejandro Cremades summarize this shift perfectly.
• 9:20
  Speaker 1: founders must prioritize active capital Over brand names.
• 9:24
  Speaker 0: Yeah a tier one logo on your cap table is totally meaningless if that partner's spending all their time managing emergency bridge rounds for Their legacy portfolio.
• 9:33
  Speaker 1: exactly speed to term sheet and actual deployment velocity Are really the only metrics that matter right now
• 9:39
  Speaker 0: And this requires a massive psychological reset for founders.
• 9:42
  Speaker 0: By month four or five of the fundraise, The Operational Drag is just exhausting!
• 9:47
  Speaker 0: You want to get back building your product?
• 9:48
  Speaker 1: It's grueling and Jorian Hoover shared brilliant anecdote about danger of fatigue.
• 9:54
  Speaker 1: He advised founder who received an unsolicited eight million dollar term sheet.
• 9:59
  Speaker 0: Okay...
• 9:59
  Speaker 1: The VC wasn't great fit.
• 10:02
  Speaker 1: The terms were frankly mediocre, but the founder was so tired of the quote-unquote fundraising game They were ready to sign it just to make the pain stop.
• 10:11
  Speaker 1: Just
• 10:11
  Speaker 0: get over with
• 10:12
  Speaker 1: exactly.
• 10:13
  Speaker 1: But Hoover intervened and basically forced the founder to hit pause.
• 10:17
  Speaker 1: instead of taking a path of least resistance they mapped out the active market and systematically set up Forty-three targeted investor meetings.
• 10:25
  Speaker 0: All right, let's break down the mechanics of why taking those forty three meetings was so crucial.
• 10:30
  Speaker 0: It wasn't just about finding a better partner.
• 10:32
  Speaker 0: it was about manufacturing competitive tension.
• 10:35
  Speaker 0: bingo.
• 10:36
  Speaker 0: by having dozens of active conversations Simultaneously The founder signaled to the market that the deal was moving and that mediocre eight million dollar offer It suddenly turned into a ten million dollar term sheet at significantly better terms without breaking their original twenty percent dilution threshold.
• 10:54
  Speaker 0: They
• 10:54
  Speaker 1: literally engineered a bidding war, and understanding how to trigger that fear of missing out requires understanding investor psychology which Idemarnovic highlighted with the great real-world example... Oh!
• 11:04
  Speaker 0: The traction excuse?
• 11:05
  Speaker 0: Yes
• 11:06
  Speaker 1: so founder pitched a VC.
• 11:07
  Speaker 1: while it roughly ten K in monthly recurring revenue The VC passed, giving the absolute classic feedback.
• 11:14
  Speaker 1: Come back when you have more
• 11:15
  Speaker 1: traction.".
• 11:16
  Speaker 0: The dreaded More Traction Line?
• 11:18
  Speaker 0: Let's translate what that actually means.
• 11:19
  Speaker 0: in todays market It rarely means your metrics are fundamentally broken.
• 11:23
  Speaker 0: No!
• 11:24
  Speaker 1: Almost never...
• 11:24
  Speaker 0: It means the associate or partner lacks internal political capital to pound a table for their deal.
• 11:30
  Speaker 0: They're waiting.
• 11:31
  Speaker 0: risk transfer
• 11:33
  Speaker 1: Precisely the point Novik made.
• 11:35
  Speaker 1: So Novik s own firm looked at exact same data built conviction and issued a term sheet that week.
• 11:42
  Speaker 0: And what happened?
• 11:43
  Speaker 1: Less than fourteen days later, the original VC who had demanded more traction suddenly called the founder aggressively trying to squeeze into the round.
• 11:51
  Speaker 0: Unbelievable!
• 11:52
  Speaker 1: Right... The MRR hadn't changed…the products haven't change….The only thing they've changed was another institutional investor has validated this asset.
• 12:03
  Speaker 0: So if you are a founder or an operator listening right now, the takeaway is entirely operational.
• 12:08
  Speaker 0: You cannot treat fundraising as a bespoke art project anymore.
• 12:12
  Speaker 0: Raviv Sapir argues that you must treat it exactly like a B-to-B sauce enterprise sales funnel.
• 12:17
  Speaker 1: Yep...you don't guess why an Enterprise client churns and shouldn't guess what VC passes?
• 12:24
  Speaker 1: You have to track conversion bottlenecks at every single stage!
• 12:28
  Speaker 0: Like how many first meetings convert into a partner
• 12:30
  Speaker 0: pitch?!
• 12:31
  Speaker 1: Exactly.
• 12:32
  Speaker 1: How many partner pitches make it to the investment committee?
• 12:35
  Speaker 1: If there's a massive drop-off at the data room stage, well your narrative is probably falling apart when they look at you're cohort retention
• 12:42
  Speaker 0: Right.
• 12:43
  Speaker 0: Discipline in top of funnel process drives predictability and term sheet outcomes.
• 12:48
  Speaker 0: Hope isn't strategy here.
• 12:50
  Speaker 0: Competitive tension is ultimate leverage.
• 12:52
  Speaker 0: It's
• 12:53
  Speaker 1: only leverage that you have.
• 12:54
  Speaker 0: And Here Is The Fascinating Twist.
• 12:56
  Speaker 0: Founders are professionalizing their fundraising pipelines, but the VC sitting across the table or being forced to undergo an even more radical operational evolution just keep pace.
• 13:06
  Speaker 1: Yeah Venture Capital itself is rapidly turning into a software business.
• 13:10
  Speaker 0: The legacy model of Venture capital relied heavily on human capital to source and filter deals.
• 13:15
  Speaker 0: But Simon Betier recently mapped out the twenty-twenty six VC tech stack And the sheer volume of infrastructure is staggering.
• 13:24
  Speaker 0: We are looking at forty-eight distinct tools spread across sixteen categories.
• 13:28
  Speaker 1: A completely different job now!
• 13:29
  Speaker 0: We're talking about inbound deal matching platforms like Deckmatch that automatically ingest and score pitch decks, CRM systems like Affinity that passively track every email & calendar invite to map relationship strength, Deep Research Tools leveraging clod and perplexity... ...to instantly synthesize competitive landscapes
• 13:47
  Speaker 1: as a full enterprise software stack.
• 13:50
  Speaker 1: And it's designed to replace the traditional army of junior associates which is driving a massive shift in firm architecture.
• 13:56
  Speaker 0: The Solo GPs?
• 13:57
  Speaker 0: Yeah,
• 13:58
  Speaker 1: Michael Schneider shared data from VC Lab that illustrates this perfectly over sixty-five percent of the new venture firms they are helping launch our structured as Solo GP's.
• 14:07
  Speaker 1: The solo capitalist has no longer a niche boutique strategy.
• 14:10
  Speaker 1: it is becoming the default operating model for emerging managers.
• 14:13
  Speaker 0: let's explore the mechanics of why.
• 14:15
  Speaker 0: that is because historically you needed to ten person firm just to manage Sourcing, initial screening drafting investment memos doing the LP reporting.
• 14:27
  Speaker 0: Yeah it
• 14:27
  Speaker 1: was very manual.
• 14:28
  Speaker 1: Today
• 14:28
  Speaker 0: a single GP armed with an optimized AI stack can cover the exact same market surface area.
• 14:35
  Speaker 0: They have the leverage of an institution but the agility of an angel investor.
• 14:40
  Speaker 1: That's a great way to put it.
• 14:41
  Speaker 1: But there is even deeper mathematical reason why large partnerships are becoming structurally disadvantaged.
• 14:48
  Speaker 1: Pavel Prada unpacked this through what he calls The Rainmaker Paradox.
• 14:52
  Speaker 0: Okay, tell us about the Rainmaker paradox.
• 14:54
  Speaker 0: So
• 14:54
  Speaker 1: the traditional assumption is that if you add more partners and associates to a venture fund The firm's collective network grows proportionally right?
• 15:01
  Speaker 1: Sure!
• 15:02
  Speaker 0: More bodies...more handshakes Right
• 15:04
  Speaker 1: But Prada analyzed the CRM data across more than fifty venture funds And discovered the exact opposite as true.
• 15:10
  Speaker 1: Yeah As a firm scaled its headcount Network concentration actually worsens.
• 15:16
  Speaker 1: In larger funds, specifically those with eleven or more employees there is a structural gap of over twelve thousand professional contacts between the top-performing partner and the median employee.
• 15:26
  Speaker 0: Wow!
• 15:27
  Speaker 0: Let's walk through.
• 15:29
  Speaker 0: AventureFirm's.
• 15:30
  Speaker 0: CRM is supposed to be this shared brain.
• 15:33
  Speaker 0: But what actually happens as the film scales, Is that junior staff get bogged down in administrative tasks and portfolio support while senior partners spend a hundred percent of their time out on the market hoarding relationships?
• 15:43
  Speaker 1: Exactly!
• 15:44
  Speaker 1: The firm doesn't operate as cohesive network.
• 15:47
  Speaker 1: it operates as single massive star orbited by operational satellites.
• 15:51
  Speaker 0: So if you are an LP listening right now allocating capital to funds, this should completely reframe your risk models.
• 15:57
  Speaker 0: Why Are You paying two percent management fees on a five hundred million dollar fund?
• 16:01
  Speaker 0: To support a fifteen person firm where all the proprietary deal float?
• 16:04
  Speaker 0: The actual alpha generating intelligence lives entirely On one specific partner's iPhone.
• 16:09
  Speaker 1: Right If that rainmaker leaves the firms network evaporates overnight.
• 16:14
  Speaker 1: Emerging, smaller managers actually have a genuine structural advantage because their concentration is intentional not just byproduct of organizational failure.
• 16:22
  Speaker 0: And this shift toward precision and concentration impacting every layer in the market including how capital was distributed demographically.
• 16:30
  Speaker 0: Nell Daly and Zabreen Khan shared some crucial inclusion data that really highlights the complexity of this new landscape.
• 16:36
  Speaker 1: Yeah, his date is really revealing!
• 16:38
  Speaker 0: In twenty-twenty five female founders raised a record seventy three point six billion dollars representing twenty seven point seven percent.
• 16:49
  Speaker 0: Now, on the surface that is a fantastic headline number.
• 16:52
  Speaker 0: It's a significant jump from historical averages!
• 16:59
  Speaker 1: Oh,
• 17:10
  Speaker 0: wow.
• 17:11
  Speaker 0: So this illustrates the barbell effect happening in venture funding right now.
• 17:15
  Speaker 0: The ceiling has absolutely been shattered for a select few founders operating in the AI consensus trade.
• 17:20
  Speaker 0: Yes But if you strip out those Ai mega rounds... ...the middle of market remains incredibly tight.
• 17:26
  Speaker 0: Capital is concentrating into fewer vastly larger rounds while overall deal count actually declined.
• 17:35
  Speaker 0: AI, Series A software startup is higher today than it has been in a decade.
• 17:39
  Speaker 1: It's completely a market of extremes.
• 17:41
  Speaker 1: we have record amounts of dry powder sitting on the sidelines but unlocking it requires an entirely new level of operational rigor.
• 17:49
  Speaker 0: right whether you are a solo GP leveraging language models to compete with legacy institutions or founder treating your fundraise like enterprise sales funnel engineer competitive tension.
• 18:00
  Speaker 0: The margin for error has basically evaporated.
• 18:02
  Speaker 1: Yeah all these insights as efficiency.
• 18:06
  Speaker 1: Capital is no longer patient, it demands immediate velocity.
• 18:09
  Speaker 1: whether that's chasing the AI wave in
• 18:21
  Speaker 0: I do
• 18:30
  Speaker 1: want to leave you with one final lingering thought from Mike Schatzman before your next board meeting.
• 18:35
  Speaker 1: We've spent this entire deep dive talking about the billions flooding into AI capabilities, but Schatzmann highlighted a quietly profound piece of math.
• 18:45
  Speaker 1: right now for every dollar company saved by reducing human headcount they only spend three cents on AI as replacement.
• 18:51
  Speaker 0: just three cents
• 18:52
  Speaker 1: three cents.
• 18:53
  Speaker 1: The question isn't just about that technology itself.
• 18:56
  Speaker 1: ask yourself where are those other ninety seven cents going?
• 18:59
  Speaker 1: And how does that massive phantom margin completely rewrite your expectations for capital efficiency in a market still orbiting the AI black hole?