All Stories
Seed Funding Hasn’t Stalled, But It’s Skewing Larger And Is More Competitive Than Ever, Crunchbase Data Shows
Venture capital news headlines these days are dominated by stories of size: capital concentration into the highest-growth companies, surging valuations, seed rounds totaling tens or even hundreds of millions of dollars, and megafunds raising tens of billions in new capital. Smaller funds and more modest seed rounds are seemingly out of favor. Seed trends bifurcate Crunchbase’s U.S. seed funding numbers confirm that perception. Deal counts and amounts are down roughly 20% year over year for the pre-seed and regular seed funding range bands that include deals of $200,000 to under $5 million. (As always, that proportion will improve a bit over time as smaller seed rounds are added to Crunchbase.) The mid-tier band, from $5 million to under $10 million, was on par year over year. Among U.S. seed funding deals, it’s only the upper bands of larger and outlier seed rounds — those $10 million and above — that grew in 2025, Crunchbase data shows. It’s a bifurcated market, according to Katie Stanton, founder of seed fund Moxxie Ventures. “You’re either an AI elite team that is growing really fast and you’re going to raise a ton of capital at Series A from one of the big firms — or you’re everybody else,” she said. In reaction to the market changing, her fund has shifted its strategy, saving a greater proportion — 60% to 70% for primary capital — compared to 50% in prior funds. “We would rather have more shots on goal,” she said. The second shift has been to find founders even earlier, often not even waiting for product-market fit. Seed deal counts The majority of seed-stage deal counts still occur for rounds $5 million and under. But that percentage has trended down over time, from 93% in 2018 to 75% of deals in 2025. Meanwhile, larger and outlier seed rounds of $10 million and above have climbed from 2% to 9% over that time. That means roughly 1 in 10 seed deals over $200,000 in 2025 were in deals $10 million and over, numbering around 360. Seed amounts U.S. seed funding totaled $19.4 billion in 2025, per Crunchbase data. Large deals drove that increase — 51% in seed deals $10 million and over, compared to a third in 2024. The largest seed round in 2025 was $2 billion for Mira Murati’s Thinking Machines Lab. Between 2018 and 2025, seed rounds of $200,000 to $5 million fell, from 70% of all seed funding amounts to 26%. At the same time, seed rounds of $5 million and above have gained ground since 2021, and remained elevated in contrast to 2020 and earlier. In 2025, the biggest jump in amounts were the outlier seed rounds — those deals $50 million and above — which increased more than 300%. Even those larger seed rounds of $10 million to $50 million gained 20%. Seed reshaped Crunchbase data shows seed funding has by no means stalled. Instead, AI is reshaping seed investment, with multistage venture and mid-tier funds backing hot companies earlier and at higher values due to founder pedigree or company traction. As a result, larger seed rounds increased in 2025 with more than 20 outlier deals of $50 million-plus and over 300 in the $10 million to $50 million range. Seed fund managers are shifting strategies based on a changing funding market. “It has never been so easy to build a product, and it’s never been so hard to build a business,” said Stanton. A small seed round can lead to the next breakthrough company. “There’s still a need for the smaller companies to emerge, and the smaller VCs to emerge to serve those different constituencies,” she said. Related Crunchbase queries: Related reading: Illustration: Dom Guzman
Exclusive: YC Doubles Down On Trayd, A Construction Tech Startup That Just Raised $10M In 3 Weeks
Trayd, a startup that is building a back office operating system for the construction industry, has raised $10 million in Series A funding, it tells Crunchbase News exclusively. White Star Capital led the company’s Series A, which was raised in just three weeks and included participation from repeat backers Y Combinator and Suffolk Technologies. The round also included an investment from new strategic backer RXR Realty, a real estate and technology investment firm. It brings New York-based Trayd’s total funding to $17 million. Co-founder and CEO Anna Berger grew up in a New York construction family, watching her father navigate razor-thin margins and complex compliance requirements. “I saw firsthand the operational strain that comes with juggling union rules, multistate labor laws, and endless manual back-office processes,” she recalls. The experience inspired her to team up with Cara Kessler, the company’s CTO, who spent 10 years as LinkedIn’s web platform lead, to start Trayd in 2021. Specialty trade serviceAnna Berger, CEO, and Cara Kessler, CTO, co-founders of Trayd. (Courtesy photo) For the unacquainted, specialty trade contractors are businesses that place skilled workers on job sites to perform the actual physical building work. These contractors include concrete crews, electricians, plumbers, ironworkers, painters and fireproofing. They’re distinct from general contractors, who manage and coordinate projects overall but don’t typically perform the hands-on trade work themselves. Trayd automates payroll, HR, compliance and labor cost tracking for such contractors. Among the benefits it touts are providing real-time visibility into the costs of labor, equipment and materials. The startup aims to substantially cut the time specialty trade contractors spend on its weekly payroll and compliance process. “What used to take 14 hours of manual work can now be done in under 30 minutes,” Berger told Crunchbase News. Trayd is working to fill what it believes is a unique gap in the market. While there are significantly more specialty trade contractors than general contractors, the majority of construction technology has been built for the latter, Berger believes. The startup’s closest competitors are legacy payroll providers like ADP and Paychex, along with newer companies like Miter and Lumber. “The difference is that most of these systems weren’t built for the complexity of specialty trades,” Berger explains. “Trayd was.” Streamlining payroll In construction, compensation is uniquely complex, Berger said. A single worker might earn four different pay rates in a single day depending on the specific trade task, the project scope and the jurisdiction. “Generic” payroll platforms cannot handle this constant rate variability, contends Berger. For example, payroll admins might receive stacks of paper timesheets or phoned-in hours from various job sites. Then they have to manually key all of that field data into Excel spreadsheets and calculate the pay rates by hand, factoring in union rules, prevailing wage requirements and state-by-state taxes. They might then have to cross-check the spreadsheet math and manually double-enter the finalized numbers into a generic payroll system, and then again into their accounting software. Trayd, according to Berger, dramatically reduces the time to perform all those tasks by capturing the time data directly from the field and automatically calculating the correct variable pay rates, union deductions, and multistate taxes. “Unlike salaried workforces, construction workers can earn multiple different rates in a single day depending on the trade, the project, and whether the work falls under prevailing wage, state or union requirements,” she said. “Trayd was designed from day one to handle that complexity.” National expansion The product seems to be resonating in the industry. Trayd has grown revenue over 600% year over year and moves tens of millions of payroll dollars each week, according to Berger. Several hundred contractors use Trayd weekly. United General Contractors, Wohl Diversified Services and Titan Structural Group are among its customers. The startup operates on a SaaS model, with pricing tied to the number of workers processed through payroll. Trayd started in New York and the broader Northeast, where union density and regulatory complexity are highest. It is now expanding nationally. Presently, it has about two dozen employees. Before Trayd, Berger co-founded Curtn, a consumer social platform backed by Sam Altman that is now defunct. She acknowledges that being female founders in a male-dominated industry has not been easy. “As women building in construction — where we’re outnumbered 9 to 1 — the default assumption is that we’re too far removed or don’t have access to truly understand the problems on the ground. In the early years especially, there’s a ‘prove it twice’ dynamic. Without the benefit of the doubt, we had to earn credibility through repetition —- every meeting, every deal, every product decision. We’ve had to work twice as hard to be taken seriously,” she told Crunchbase News. “But that pressure becomes an advantage. You show up more prepared, you listen more closely, and you build conviction faster. Over time, that compounds into a better product and deeper, more trusted customer relationships.” Eddie Lee, general partner at White Star Capital, said his firm was first impressed by Trayd’s founding team, describing Berger and Kessler as “a rare combination.” “Anna’s background and family ties to the space allow her to understand the unique pain points contractors face from the inside,” he wrote via email. “Cara brings the technical depth to build mission-critical systems without sacrificing product simplicity.” Beyond the caliber of the founders, White Star also believes that Trayd stands out because “it is truly a better product for its customers.” “On a technical level, we were very impressed by how thoughtfully the product has been built,” Lee added. “We see that as a real advantage, because by structuring data cleanly at the system level, Trayd is better positioned to scale reliably and to become a strong foundation for AI in the construction industry over time.” Venture investment in property technology startups has rebounded in recent years after plunging from the pandemic peak. In 2025, startups in the sector pulled in approximately $10.5 billion in seed- through growth-stage financing globally, per Crunchbase data. That’s up about 17% from $9 billion in 2024, with much of the recent investment going to startups that promise greater ROI through the use of automation or AI. Related Crunchbase query: Related reading: Illustration: Dom Guzman
Kleiner Perkins Raises $3.5B For AI-Focused Funds
Storied venture capital firm Kleiner Perkins announced Tuesday that it has raised $3.5 billion across new funds with a primary focus on artificial intelligence. The fundraise includes $1 billion for KP22, a fund to back early-stage companies, and $2.5 billion targeted for growth-stage investments. It’s a considerable increase in capital commitments compared to the last time the Silicon Valley-based firm raised a flagship fund, back in 2024. In that raise, Kleiner pulled in just over $2 billion for funds to back early- and later-stage startups. This time around, Kleiner believes market fundamentals look particularly attractive for scaling up. “The AI super-cycle is one of the most important company-building moments in our lifetimes, and we are still in the early innings,” its fundraising announcement states. Kleiner also notes that AI is enabling today’s startups to iterate and grow faster than in past cycles. Founded in 1972, Kleiner has long been known as a cross-industry investor, active in virtually every popular sector for venture dealmaking. For its latest fund, the firm also identified a broad array of focus areas, including professional services, healthcare, autonomy, security, financial services and the physical economy. Recent investments Most recently, Kleiner, like most venture heavyweights, has been focused on AI startups. Beyond that, however, its portfolio companies are a highly varied lot. To illustrate, we used Crunchbase data to put together a list of the latest reported rounds in which it served as a lead or co-lead investor. It spans healthcare, accounting and cybersecurity, among other areas. Large lead investments While it’s active in seed- and early-stage dealmaking, Kleiner also leads quite a few larger rounds. Over the past year, it’s been lead investor in at least five valued at $150 million or more, which we list below. Of these, the largest was a $600 million Series F for Applied Intuition, a developer of autonomous vehicle technology. The next-largest include a $356 million Series D for Chainguard, focused on secure open-source software for AI systems, and a $300 million Series E for Harvey, the AI legal tech unicorn. Exits too Kleiner has also seen a few sizable recent exits for portfolio companies that it backed as lead investor. This includes last year’s largest software IPO — Figma — which counted Kleiner as Series B lead investor. The firm was also an early lead investor in business credit card provider Brex, which Capital One agreed to acquire this year for $5.15 billion. Of course, Kleiner also has much more famous portfolio investments in its more distant past, including Google, Uber and Airbnb, to name a few. You don’t last 50 years in the venture business without at least some of those too. Related Crunchbase queries: Related reading: Illustration: Dom Guzman
The Tax Credit Opportunities Startups Often Forget (And Why It Keeps Happening)
By Harrison Garba Founders spend a lot of time thinking about capital. They model burn carefully. They negotiate valuation. They weigh hiring plans against runway. But many startups overlook a source of capital that doesn’t require dilution at all: tax credits. And to be clear, this isn’t typically because a business doesn’t qualify. It’s because no one builds a process to identify and capture these credits consistently. Most startups are aware of at least one major opportunity, and that’s the Research & Development tax credit. But fewer founders take a broader look at business decisions throughout the year and how many of those may lead to tax credit opportunities. Hiring decisions, benefit structures, accessibility upgrades, facility investments and certain energy projects all can carry incentives. So, the issue isn’t eligibility. It’s ownership, timing and consistency. Harrison Garba In early-stage companies, finance teams are lean. Credits often get discussed once a year during tax preparation. However, by that point, it can be too late. The required elections may have been missed, documentation may not support a claim, or deadlines may have passed. When that happens, the opportunity is gone. We see this pattern frequently in examples such as: None of the above decisions are inherently wrong, but they are incomplete. Coordinating credits Tax credits don’t appear automatically because money was spent. Taking advantage requires planning, including specific documentation, elections and coordination between departments. Without that coordination, even well-managed startups leave savings unclaimed. More-mature companies approach this differently. Instead of waiting until year-end to ask, “Did we qualify for anything?” established organizations build periodic reviews into their operating cadence. This doesn’t mean turning every department into tax specialists. It requires clarity around who’s responsible for asking the question early enough, and it ideally includes expert guidance and support to get it right. It’s helpful to think about this as an evolution. At a reactive stage — which is most startups — credits are evaluated only when the tax return is being prepared. At a more structured stage, the company reviews credit opportunities quarterly and aligns documentation throughout the year. And in a strategic stage, leadership fully understands how certain business decisions may create incentives and ensures the right processes are in place before those decisions are implemented. Multiple credits add up The accumulated financial impact can be meaningful. While a single credit isn’t likely to transform a business, multiple credits across hiring, development and benefits can offset real costs. For companies focused on extending runway without raising additional capital, those offsets matter. There’s also a governance component. Investors and buyers increasingly review operational controls during diligence. A startup that has evaluated available credits and maintained documentation signals discipline. A company that hasn’t considered them at all may invite additional questions (especially if elections were missed or filings need to be amended). None of this is to suggest credits should drive a founder’s core strategy. Product development, revenue growth and customer demand remain the priority. But when companies are already investing in innovation, hiring and infrastructure, it makes sense to evaluate whether part of that investment can be recovered. The first step is simple: Get the full picture before making any decisions. In many cases, that includes working with an adviser who understands how credits apply to growing businesses. Then, assign ownership. Determine who is responsible for reviewing credit opportunities throughout the year. Coordinate among departments like finance, HR and operations before major decisions are finalized. Make documentation part of the process rather than a reconstruction exercise at the end of the year. Being proactive Again, tax credits are not automatic. They’re for those who plan the entire year. Startups looking to be more proactive should keep credits like the R&D in mind for its potentially meaningful offsets when investing in product or technical improvements. But don’t stop there. If considering structured paid leave, review the Paid Family and Medical Leave Credit, which can apply when policies meet specific requirements. Businesses reviewing facility improvements may qualify for the Disabled Access Credit. While credits such as these don’t apply to every company, they’re common enough to demand attention before decisions are finalized — even seemingly unrelated ones. Startups focused on capital efficiency will see this planning make a measurable difference over time. Harrison Garba is a tax supervisor specializing in research and development tax credits at Burkland Associates. He holds a master of science in accounting from The University of Texas at Dallas and has experience across both public and private sectors. Garba has spent several years advising companies on R&D tax credits, helping startups and growth-stage businesses navigate complex tax regulations and maximize available incentives. Related reading: Illustration: Dom Guzman
The Week’s 10 Biggest Funding Rounds: Investment Slows, But Security And AI Remain Top Picks
Want to keep track of the largest startup funding deals in 2025 with our curated list of $100 million-plus venture deals to U.S.-based companies? Check out The Crunchbase Megadeals Board. This is a weekly feature that runs down the week’s top 10 announced funding rounds in the U.S. Check out last week’s biggest funding deal roundup here. In insecure times, security looks like an appealing sector for investment. That’s one interpretation of this week’s tally of the largest startup funding rounds. The size of the largest U.S. deals was smaller than in recent weeks, and heavily featured cybersecurity- and privacy-focused startups. This includes the week’s biggest round — a $375 million Series B for consumer privacy and security platform Cloaked. Other areas that attracted good-sized financings included AI infrastructure, biotech, healthcare, and robotics. 1. Cloaked, $375M, privacy: Cloaked, a provider of consumer privacy and security tools, raised $375 million in Series B funding led by General Catalyst and Liberty City Ventures. Founded in 2020, the Massachusetts-based company sells monthly subscriptions for individuals and families. 2. Frore Systems, $143M, AI infrastructure: Frore Systems, a developer of integrated cooling architecture for AI computing and networking hardware, announced that it closed on $143 million in Series D funding. MVP Ventures led the financing, which set a $1.64 billion valuation for the 8-year-old, San Jose-based company. 3. (tied) XBow, $120M, cybersecurity: Seattle-based XBow, a provider of autonomous security testing technology, picked up $120 million in Series C funding. DFJ Growth and Northzone led the round, which values the 2-year-old company at over $1 billion. 3. (tied) Oasis Security, $120M, cybersecurity: Oasis Security, a developer of identify security tools with a focus on AI agents, secured $120 million in a funding round backed by Craft Ventures, Cyberstarts, Sequoia Capital and Accel. The 4-year-old company, which is headquartered in New York and has a presence in Israel, has raised $195 million to date, per Crunchbase data. 5. (tied) Imperative Care, $100M, medical devices: Imperative Care, a medical device company focused on treatment for stroke and vascular diseases caused by blood clot formation, secured $100 million in convertible note financing. Elevage Medical Technologies and Perceptive Advisors led the investment for the Campbell, California-based company. 5. (tied) Bluesky, $100M, social media: Seattle-based social network Bluesky disclosed this week that it raised a previously unannounced $100 million Series B round that closed last spring, led by Bain Capital Crypto. 5. (tied) Cape, $100M, privacy and security: Cape, a recently launched privacy-focused mobile network, landed $100 million in Series C funding. Bain Capital Ventures and IVP led the financing, which set a $900 million valuation for the Arlington, Virginia-based company. 8. Latent, $80M, healthcare AI: Latent, an AI platform aimed at helping move patients from clinical decision to therapy, picked up $80 million in a Series A round. Spark Capital and Transformation Capital led the financing for the San Francisco-based company. 9. Crossbow Therapeutics, $77M, biotech: Cambridge, Massachusetts-based Crossbow Therapeutics, a biotech startup focused on developing new antibody therapies to treat a broad range of cancers, raised $77 million in Series B funding. Taiho Ventures and Arkin Bio Ventures led the round, which will support a Phase 1 clinical trial of the company’s lead program. 10. RoboForce, $52M, robotics: RoboForce, a startup focused on developing AI-enabled robot labor for industrial environments, said it picked up $52 million in fresh funding, bringing its total raise to $67 million. YZi Labs led the financing for the Milpitas, California-based company. Methodology We tracked the largest announced rounds in the Crunchbase database that were raised by U.S.-based companies for the period of March 14-20. Although most announced rounds are represented in the database, there could be a small time lag as some rounds are reported late in the week. Illustration: Dom Guzman
The Most Active Startup Acquirers Of The Past 3 Years Aren’t Always Who You’d Expect
Companies that buy a lot of startups don’t always have a lot in common. Some are longstanding blue chip tech and pharmaceutical companies. Others are fast-growing venture-backed unicorns. And still others are more recent public market entrants looking to stay competitive in the age of AI. To get a sense of who’s buying in bulk, we used Crunchbase data to put together a list of 79 companies that acquired three or more seed- or venture-backed startups in the past three years. From there, we picked the most acquisitive names. The most prolific startup acquirers of the past 3 years Per Crunchbase data, the most prolific acquirers of seed- and venture-backed startups in recent years are Salesforce 1, OpenAI and Snowflake. Overall, our query showed six companies with six or more known purchases, charted below. For top-ranked Salesforce, high-volume M&A is nothing new. The San Francisco software giant has purchased at least 91 companies in the past 20 years, per Crunchbase data. Its most recent startup purchases include Momentum, a revenue orchestration platform, and Cimulate AI, which focuses on agentic AI for e-commerce. OpenAI, by contrast, has a shorter track record of M&A shopping sprees. The pioneering generative AI company has bought 16 companies in the past three years. Among the most recent was an acqui-hire deal involving open-source AI agent OpenClaw and its creator, Peter Steinberger. This month, it also snapped up Astral, a creator of open source tools for software developers, and Promptfoo, an open-source tool for testing AI applications. Snowflake, meanwhile, has 19 acquisitions to date. Most recently, it acquired Observe, a developer of AI observability tools that previously raised more than $460 million in venture funding. Notably, recent the active acquirers list for recent years looks quite a bit different that the ranking of all-time top M&A dealmakers in the Crunchbase dataset, shown below: Highest-spending acquirers The most prolific startup buyers also aren’t always the biggest check-writers. By the latter metric, the far-and-away leader is Google, and its $32 billion acquisition of Wiz. For a broader picture view, we used Crunchbase data to put together a list of six companies that made the biggest-ticket funded startup acquisitions of the past three years. 2026 off to a promising start So far this year, it looks like the pace of startup M&A dealmaking remains fairly robust. This includes two deals in the multiple billions: Capital One’s $5.15 billion purchase of Brex and Eli Lilly’s $2.4 billion acquisition of Orna Therapeutics. The AI sector’s appetite for acqui-hires and smaller purchases of earlier-stage startups also continues to boost momentum. We’ll see if it keeps up. Related Crunchbase list: Related reading: Illustration: Dom Guzman
Why You Haven’t Raised Startup Funding (Yet)
By Julia Sabitova If you’re the CEO of Lovable or Higgsfeld and reached $100 million in ARR in under a year, this article isn’t for you — enjoy being a unicorn as thousands of investors beg to fund you. But if you’re not, then let’s be honest: raising in 2026 is tough. Although global venture funding is growing, raising capital isn’t any easier for the average startup. According to Crunchbase data, more than a third of global funding in 2025 went to just 629 companies, compared to 24% of funding in 2024. This highlights a growing concentration of capital, making most of that funding effectively inaccessible to early-stage startups. So what can founders do to fix that? We don’t invite strangers to our houses, and we don’t hire them for important jobs either. For thousands of years, trust and credibility were the most important factors in forming relationships, both business and personal. In 2025, Silicon Valley companies attracted nearly 50% of the entire U.S. venture funding. Silicon Valley is also home to 312 unicorns, over half of all U.S.-based unicorns. Julia Sabitova It’s not because San Francisco Bay Area founders are inherently smarter — it’s largely about being close to capital and networks. When you’re in constant proximity of MAG7 companies and hundreds of VCs, connections happen organically, through social gatherings, meetups and referrals. This is how credibility is formed: through connections and exposure. So, is networking the secret to raising capital? Partially, but it doesn’t scale. You can’t just meet the whole industry and invite them all to a 1-1. So instead, you have to build your reputation. Here are my top four pieces of advice on how to build it right. Be visible Make your growth visible. Whenever you reach a significant milestone — raising a round, hitting a user target, or achieving revenue growth — the market should hear about it. We’ve seen countless companies reach a huge target and then fail to spread the word about it. Make sure to plan all media coverage in advance, keep exclusive news up your sleeve, and have an extensive media strategy. Once the word is out through your company’s social media, pitching to journalists becomes significantly harder. Everybody wants exclusives, and no one wants to write about old news. Global media outlets are all about relationships. Make sure to form a meaningful connection with journalists covering your particular niche. Focus on customers The second priority when raising funds is your company’s place in the overall market landscape. Be where your customers are. Many founders make the same mistake: chasing investors instead of customers. Remember that investors will always find good investment opportunities. Your job is to make sure that your company is one of them. Investors have to see that your company has a sustainable customer acquisition approach and is able to continuously grow its user base. Chasing investors can even damage your public picture. If VCs see you spending heavily to attract investors rather than customers, it may signal misaligned priorities. Be a thought leader Important thing No. 3: thought leadership. You have to prove your credibility through actively participating in conferences and meetups. Speaking at industry events signals credibility at scale. Conferences are highly selective. Being on stage implies that organizers have already vetted your expertise. Getting on the stage and delivering your core message will help your credibility more than any degree or a title. Raise symbolic capital The fourth significant factor is symbolic capital — the way your company is perceived by the market. A great way to acquire symbolic capital is through various ratings and features. They’re usually put together by the larger media outlets and include programs such as Forbes’ 30 Under 30, TechCrunch Startup Battlefield and Slush100. Similar to conferences, participating in different features shows potential investors that a credible player with a good reputation has already done a background check on you and is ready to endorse you. One well-known logo in your endorsements list can go a long way in securing the next round of funding for your startup. A somewhat unexpected benefit of getting into the biggest ratings and roundups is your AI visibility. Your company being featured in one of these lists will significantly improve the odds that AI will highlight your company in relevant conversations. AI visibility is increasingly important for user acquisition, considering that according to Feedonomics, 39% of users already use AI instead of traditional search engines for shopping. Reputation: You can’t buy it Reputation is one of the rare things in the business world that you can’t just buy. One of our longstanding partners received an invitation to a dinner with the Royal Family of the United Kingdom, which is something that no amount of marketing budget will give you. It takes a lot of coordinated work and effort that won’t result in exact KPIs on day one, which is why many startups just don’t have the patience and strategy it takes to build credibility. As development and compute costs fall, the number of startups continues to grow. In that environment, reputation becomes the key differentiator between companies that attract capital — and those that don’t. Julia Sabitova is a communications strategist and serial entrepreneur with more than 10 years of experience. She co-founded CloEE, an AI adviser for smart manufacturing, and leads BeGlobe, a PR agency for tech startups and VCs. She is a graduate of UC Berkeley’s SkyDeck Accelerator. Illustration: Dom Guzman
Don’t Just Talk About AI. Measure Business Outputs. Here’s How.
By Bob Morse and Dario Fanucchi Last year felt like the Year of the AI Pilot. Companies bought LLM subscriptions, managers checked on employee usage, and coffee chats abounded with the “AI wrote my memo” motif. Looking around today, there is widespread disappointment with the impact of these AI pilots. Add to this the recent sell-off in SaaS stocks, and the question is no longer “Are we using AI?” but rather “Is this thing working?” AI is an invention that is in the process of becoming an innovation. An invention is a new capability; it is not an innovation until it has a business model. In that light, experimentation last year was the sensible move. Bob Morse It is becoming clear now that the form that innovation takes will be AI systems trusted with real decisions — what Peter Drucker would call executives, and what are today referred to as agentic AI. As we turn to the question at hand, Is this thing working?, we can look to one of Drucker’s intellectual disciples for a framework to take us forward. Andy Grove, the legendary former CEO of Intel, turned Drucker’s writings into a hard-nosed, pragmatic approach to managing knowledge-worker organizations. His book, “High Output Management,” provides the classic framework for measuring the outputs of middle managers. This is not an easy thing to measure. But Grove is relentless in insisting it can and must be measured. Dario Fanucchi As we address the question of whether AI agents are delivering tangible value, we have to shift our focus away from activities, anecdotes and initiatives. These are inputs. Grove argues that organizations must instead focus on outputs. If we try to think like Grove, we would first define the business outcome we wanted to achieve, and then measure our agentic AI only by whether this performance metric is better. A mathematical approach As we began working on this several years ago across our software portfolio, I had the great good fortune to meet Dario Fanucchi, a mathematician who was using AI to solve real-world problems in a very similar way. He is also co-founder and CTO of Isazi 1, a decade-old, 70-plus-person team of mathematicians and engineers who have completed hundreds of projects for leading companies around the world. His approach to these has a singular focus: improving core business metrics. Isazi came to the same idea of measuring outputs, although starting from the field of mathematics rather than organizational behavior. The idea is to approach AI projects as though they are mathematical optimization problems: Define a target measure (such as throughput or working capital), ask what variables influence that metric, and model the mechanism by which the target measure is moved. Then all initiatives are aligned to this target measure, and success is measured by its improvement. This aligns well with how AI models are built and improved: benchmarks and evals are always the core measure of success. Here, these evals are directly aligned to business metrics. You must begin with the output you want to measure. And then you watch that output measurement, as a gauge, and see how long it takes until that gauge is reading changes, how much it changes, in what direction, and whether it sustains. The time it takes to see (and sustain) a material movement is called “Time To Production.” Our theory on why so many pilots fail is that companies tend to pick an AI tool and a pilot duration and qualitatively check in with users at the end of that time. While we at Strattam and Isazi appreciate experiments and pilots, we have found that results are best when that process is reversed. We choose the output we want to see improved, vary the AI tools until one moves the dial, and measure the time it takes to change the output positively and in a sustainable way. The shorter the Time To Production, the better. A real-world example Let me share an example. One of Strattam’s portfolio companies, Trax Technologies, is in the business of helping very large multinationals manage their global shipping. A key part of the offering is ensuring that freight bills are complete, match the contract, are approved for payment, and are properly accounted for. Trax works across all geographies and all shipping modes, with thousands of carriers. Discrepancies between the bill and the shipper contract are common. Handling those “exceptions” at scale is a key part of the service, and historically, Trax has had a large in-house team that resolves those. In 2024, it identified AI’s ability to resolve some of those exceptions as a key opportunity and developed the AI Audit Optimizer in-house. The output goal was clear: the fraction of exceptions resolved without human intervention. The first quarter after its release, the Trax AI Audit Optimizer resolved some 826,000 exceptions that otherwise would have required human intervention. That was a good start, but not worth writing home about just yet. In Q2, however, the system remained stuck at that same level, rather than improving. So Trax rapidly experimented to see what would improve outcomes. In Q3, the company discovered that a human prompt engineer interacting with the system made a big difference. As a result, in Q4, resolved exceptions tripled to 2.5 million. Now we’re talking. With the output gauge firmly in mind, Trax is moving forward by adjusting interaction points of the prompt engineer and the system. It used data from successful and unsuccessful resolutions to retrain the system. The company also set quarterly goals; next quarter, it will aim for the Trax AI Audit Optimizer to resolve more than any previous quarter. This story shows how studying an output gauge allowed the company to tune and adapt the AI tooling to deliver the outcomes that actually matter. Trax is intent on fixing its customers’ problems so it can earn market share. Its use of AI helped it do that, and its output measurements prove the real-world value of the AI innovation. Measure what matters Amidst all the hype, we all care that our companies actually adapt, actually deliver customer value, and actually succeed. We know that we cannot keep doing what we are doing as we have been doing it, that our futures may well depend on our ability to adapt. But this is different from actually adapting. To adapt successfully, resist the urge to buy tools and run pilots and tell anecdotes and report on activities. Those are just inputs. Instead, determine the outcome measurement that matters, and watch it like a hawk to see if AI is delivering cold hard business results. If it’s not, change your AI until the dial moves. Drawing on the time-tested wisdom of Drucker and Grove in this way, you’ll ensure AI earns its keep at your firm. Bob Morse co-founded Strattam Capital in 2014 and is managing partner. He has served on numerous private and public technology company boards, and currently is a director of CloudHesive, Contegix, Daxtra Technologies, Green Security, Resource Navigation and Trax Group. Previously, he was a partner and member of the investment committee at Oak Hill Capital Partners. He also worked at GCC Investments and Morgan Stanley. Morse serves on the board of directors of Austin PBS and as member of the advisory board for the HMTF Center for Private Equity Finance at The University of Texas at Austin McCombs School of Business. He attended Princeton University, graduating summa cum laude with a B.S.E., and Stanford Graduate School of Business, where he earned his MBA and was an Arjay Miller Scholar. Morse lives in Austin. Dario Fanucchi contributed to this article. He is chief technology officer at Isazi, a Johannesburg-based applied artificial intelligence firm purpose-built to deliver production-grade AI software solutions for clients. Fanucchi has excelled academically in the fields of computer science, mathematics and physics throughout his career. Related reading: Illustration: Dom Guzman