The Innovius Edge - Edition #3
Discipline Over Hype: Building Enduring Value in a Frenzied Market
It’s been another pivotal month at Innovius as our momentum continues to build. While the broader venture market is shaped by rising valuations and AI-driven frenzy, we remain focused on building something enduring. Our strategy—centered on go-to-market excellence, disciplined entry valuations, and post-investment impact—continues to set us apart.
This month, we led the Series A (functionally a B) in RightRev, a company applying AI to the increasingly complex world of revenue recognition—a problem made even more challenging by the proliferation of AI-driven business models. RightRev embodies the asymmetric upside and managed downside we seek.
We also advanced our internal capabilities: launching a new machine learning model in Athena (our proprietary platform) to better predict high-fit fundraising companies, rolling out the beta of our AI-powered memo generator, and progressing several other initiatives to increase sourcing velocity and investment precision
In this edition of The Innovius Edge, we examine the growing gap between early-stage markups and long-term fundamentals, the incentives driving short-term paper gains, and the risks of FOMO-fueled venture strategies. We also spotlight emerging opportunities in supply chain technology, share updates from GTM University on fit-engagement strategies, and unveil our prototype of an MCP-powered Innovius AI Agent designed to unify data and accelerate our workflows.
As always, our goal is to provide a transparent lens into how we think, what we’re seeing, and where we believe alpha is being mispriced—or missed entirely. Thanks for being on the journey with us.
The State of Venture
Valuation and other trends in venture capital
Last month, we noted that Series B valuations were becoming increasingly polarized. A small group of perceived AI category leaders is commanding large premiums, while valuations across the broader market continue to soften. The result is a “barbell market”—a few clear winners at the top, and many companies languishing in a crowded, struggling middle.
Beneath that surface, a more structural shift is emerging. The market’s AI obsession has sparked a frenzy at earlier stages. At the 75th percentile, step-ups from Seed to Series A are roughly 3.5x for both AI and non-AI companies. But from Series A to B, non-AI companies see ~3.0x step-ups, while AI-native companies are now commanding 5.3x.
This gap helps explain why many VCs—some of whom had never previously focused on AI—are now abruptly pivoting toward early-stage AI investments. They’re chasing markups, not fundamentals. The playbook is simple: get into a "high-hype" AI company with ~$1M ARR at Series A, mark it up aggressively at Series B, and lock in interim gains ahead of the next fundraise.
In effect, firms are manufacturing unrealized paper gains to boost short-term fund performance and meet LP expectations for visible marks. We know of multiple funds shifting their strategy explicitly for this reason, driven by LP pressure to show interim marks.
This behavior creates a self-reinforcing cycle:
More capital flows into early-stage AI, inflating valuations even further
Fundamentals matter less, since the goal is short-term markups—not long-term durability
Risk concentrates as fragile companies are pushed into unrealistic expectations without the underlying operating metrics to justify them
In short, the “AI boom” is happening on paper, and the paper boom is increasingly concentrated among a few anointed companies.
At Innovius, we stay grounded in discipline. We back companies built on durable fundamentals—not hype cycles. Our focus is on maximizing outcomes at exit, not chasing interim marks. We’d rather own a capital-efficient business that never needs to raise again than one propped up by inflated valuations, weak unit economics, and a towering preference stack.
*All data taken from Pitchbook.
Deal Intelligence
Transactions of note
Consensus High-fliers – “Consensus” deals we avoid entirely due to pricing we view as irrational. These are often strong companies—but at valuations that feel like lottery tickets, not investments.
[Cloud Security Company] (Series B):
Operates in the crowded and well-funded cloud security space
Forecasting growth from $5M to $25M ARR in 2025
Raised $180M to date ($80M Seed + $100M Series A)
Targeting a $1.5B post-money valuation
Strategic Passes – Deals we don’t engage with or pass on quickly given high valuations that we believe could be rational in hindsight given the outsized promise of a new technology or solution.
[Care Ecosystem AI Company] (Series B):
Backed by top-tier Seed and Series A investors
Provides software for care planning, staff management, and resident safety in senior living communities
Projecting growth from $4M to $14M ARR this year
Large CARR ($10M+) vs. live ARR ($4M) gap signals implementation friction with non-technical users
Raising at a $200M+ valuation
High-conviction Passes – Deals in our sweet spot in terms of stage and traction - but we passed due to fundamental concerns.
[Government Contracting AI Company] (Series B):
AI platform for streamlining government contracting (RFP process)
ARR grew from $1.5M to $4M in 2024; targeting $12M by end of 2025
Held first-mover advantage by aggressively underpricing competitors
Now faces a hypercompetitive landscape with minimal technical differentiation
Pricing pressure and commoditization risks impair downside protection and upside asymmetry.
Sector Spotlight
A space we’re sizing up
Supply chain technology is at an inflection point. The pandemic, geopolitical tensions, and rising tariffs have exposed the fragility of global supply chains—and accelerated the need for modernization. In response, we’re seeing a wave of startups aiming to become the system of record for critical supply chain workflows, embedding themselves deeper into customer operations.
We’re closely tracking innovation in the following areas:
Visibility & Transparency: Platforms leveraging AI, advanced analytics, and IoT to provide real-time tracking of goods, inventory, and shipments.
By aggregating and surfacing operational data—often through digital twins—these platforms enable faster, smarter decisions during disruption.
Notable incumbents: FourKites, Project44, Tive
Startups we’re watching: Terminal49, Vizion, PAXAFE
Forecasting & Planning: AI/ML-driven solutions that improve demand forecasting, scenario modeling, and dynamic inventory optimization.
The most compelling tools unify planning data across the enterprise, helping companies minimize stockouts, optimize working capital, and respond to volatility.
Notable incumbents: Cin7, Blue Yonder, Katana
Startups we’re watching: Mandrel, Gaia Dynamics
We’re focused on platforms that can become the operational backbone of supply chains—driving measurable ROI, automating key workflows, and supporting sustainability and compliance goals.
Winners will integrate deeply, automate intelligently, and become indispensable to day-to-day customer operation
For an interesting lens into just how complex and regulated parts of the supply chain can be, see this WSJ Article about the challenges of becoming a U.S.-licensed customs broker—a great example of where AI could drive major transformation.
GTM University
Outbound precision in an AI-driven world
In our last issue, we discussed how AI-powered search is reshaping inbound demand—reducing website traffic and pushing companies to rely more heavily on outbound tactics. But outbound, done poorly, can be just as inefficient.
The challenge is simple: most companies scale outbound by hiring more SDRs and sales leaders, hoping that more "activity" (emails, dials, touches) will drive more pipeline. Activity helps—to a point—but without systematic prioritization, conversion rates stay low, CAC climbs, and efficiency breaks down.
The root problem: outbound motions often lack precision infrastructure. SDRs operate blindly without real visibility into prospect "fit" or "engagement." Without these signals, teams resort to volume over quality—an expensive and unsustainable approach.
We’re working with our portfolio companies to fix this. Using tools like Clay and other enrichment platforms, we’re dynamically scoring "fit" based on real-world signals (e.g., recent job changes, headcount growth in buying teams, buyer-relevant keywords in job postings). In parallel, we’re scoring "engagement" based on weighted actions (e.g., form fills, webinar attendance, site visits).
When fit and engagement are mapped together, they create a fit-engagement matrix that allows outbound teams to prioritize efforts with far greater precision. Different tactics can then be deployed across quadrants, dramatically improving pipeline scalability and conversion efficiency. Early results show a 2.5x improvement in top-of-funnel conversion rates.
To build on this, we are experimenting with 11x to automate list curation based on fit-engagement methodology, conduct account research, draft personalized messaging, and execute multi-channel outreach across email, LinkedIn, and other platforms. These tools don’t just improve productivity—they create true alpha by:
Increasing TOFU (top-of-funnel) conversion rates
Automating repetitive, time-consuming tasks like research and messaging
Importantly, we aren’t just advising our portfolio—we’re also adopting this outbound philosophy internally for Innovius. More on that in the next issue.
Applied AI
Tools for speed, leverage, and clarity
Over the past few months, Model Context Protocol (MCP) has emerged as one of the most exciting developments in the AI ecosystem. By enabling agents to reason across tools, APIs, and memory, MCP marks a shift away from monolithic models toward modular, composable systems. It’s one of the most exciting developments in the AI ecosystem—and we’re building around it.
At Innovius, we’re prototyping the Innovius AI Agent, a natural-language assistant powered by MCP and integrated directly with our internal systems (PitchBook, Harmonic, Affinity, OneDrive, Athena. etc.). The goal: unify workflows, accelerate decision-making, and eliminate tedious context-switching.
You’ll soon be able to ask:
“Which Series A vertical SaaS companies in our CRM have grown FTE headcount by 30% in the last 6 months, haven’t raised in over a year, and operate in the US, UK, Canada, or Israel?””
Rather than jumping between tools and setting filters manually, the Innovius AI Agent can bridge systems, apply logic, and return results in seconds—delivering the exact answer, not just a spreadsheet.
We’re also developing write-back capabilities. For example, an investor could say:
“Mark all companies we spoke with last week that are raising a Series B in Q3 as ‘High Priority’ in the CRM, and set a reminder for the Deal Team to follow up in June.”
The Agent interprets the command, identifies relevant companies, updates records, and triggers reminders—turning natural language into structured workflow automation.
This is just the beginning. We’re focused on making MCP practical and powerful for everyday investing workflows—so our team can spend less time navigating tools and more time applying sharper judgment.
It’s another step toward our broader goal of using AI to drive speed, clarity, and conviction in every investment decision we make.
Brain Food
What we’re reading (and, sometimes, listening to)
AI could make apps irrelevant, says Meta’s tech chief (Business Insider) – Meta CTO Andrew Bosworth envisions a future where AI replaces traditional apps as the primary digital interface. Instead of opening Spotify or Netflix, users will simply instruct AI agents to handle tasks—streamlining UX, but challenging app-based monetization models.
Secure ‘quantum messages’ sent over telecoms network in breakthrough (The Financial Times) – Scientists in Germany successfully transmitted quantum-encrypted messages across a commercial network, demonstrating that quantum encryption could integrate with today’s telecom infrastructure—and safeguard data against future quantum computing threats.
Within six years, building the leading AI data center may cost $200B (Tech Crunch) - Building a leading AI data center could soon cost $200B and require 9 gigawatts of power—equivalent to nine nuclear reactors. Despite efficiency gains, compute demand is outpacing energy savings, raising serious economic and environmental challenges.
Mira Murati doubled the fundraising target for her new AI startup to $2 billion. It could be the largest seed round in history. (Business Insider) - Former OpenAI CTO Mira Murati launched Thinking Machines Lab, aiming to raise a record-setting $2B seed round. Despite no product yet, the startup has attracted top talent from OpenAI, Meta, and Anthropic—highlighting the premium on elite technical teams in AI’s next wave.
Chinese Manufactures Make Appeals to Americans: Buy Direct. (NY Times) -Facing tariffs and rising inflation, Chinese factories are using TikTok to sell low-cost replicas of luxury goods directly to U.S. consumers—sidestepping traditional retail channels and reshaping cross-border commerce dynamics.