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 鈥淎I wrote my memo鈥� motif.

Looking around today, there is . Add to this the recent sell-off in SaaS stocks, and the question is no longer 鈥淎re we using AI?鈥� but rather 鈥淚s this thing working?鈥�

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.

It is becoming clear now that the form that innovation takes will be AI systems trusted with real decisions 鈥� what Peter Drucker would call executives, and what are today referred to as agentic AI.听

As we turn to the question at hand, Is this thing working?, we can look to one of Drucker鈥檚 intellectual disciples for a framework to take us forward. Andy Grove, the legendary former CEO of , turned Drucker鈥檚 writings into a hard-nosed, pragmatic approach to managing knowledge-worker organizations. His book, 鈥淗igh Output Management,鈥� 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.

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 , a mathematician who was using AI to solve real-world problems in a very similar way. He is also co-founder and CTO of 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 鈥淭ime To Production.鈥� 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 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鈥檚 portfolio companies, , 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 鈥渆xceptions鈥� 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鈥檚 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鈥檙e 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鈥� 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鈥檚 not, change your AI until the dial moves. Drawing on the time-tested wisdom of Drucker and Grove in this way, you鈥檒l ensure AI earns its keep at your firm.

co-founded in 2014 and is managing partner. He has served on numerous private and public technology company boards, and currently is a director of , , , , and . Previously, he was a partner and member of the investment committee at . He also worked at and . Morse serves on the board of directors of and as member of the advisory board for the HMTF Center for Private Equity Finance at . He attended , graduating summa cum laude with a B.S.E., and , where he earned his MBA and was an Arjay Miller Scholar. Morse lives in Austin.

contributed to this article. He is chief technology officer at , 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:

• Invention To Innovation: Making Sense Of AI鈥檚 Moment听
• Avoiding The 鈥楩irst Board Meeting Surprise鈥� Problem
• The Rise Of The AI Executive

Illustration:

In an article a few months ago, I assessed AI against the framework of invention vs. innovation. Invention is about creating new capabilities, whereas innovation turns those capabilities into tangible, commercial value.

Tremendous capital is flowing into the 鈥渋nvention鈥� around AI, some $5 trillion at latest count in announced infrastructure to support the compute, largely for frontier models advancing the boundaries of the possible.

This over-investment in invention compares with a relative under-investment in innovation.

How should we even think about the innovation phase, the applications that will be introduced into the real economy with a sustainable long-run profitable model?

I鈥檇 like to propose a framework for the innovation phase of AI 鈥� applying Peter Drucker鈥檚 lessons from the rise of the knowledge worker to the age of AI.

As AI systems become trusted to make some decisions previously taken by human employees, they are best thought of as acting like a knowledge worker. The term 鈥渒nowledge worker鈥� was famously coined by management theorist Peter Drucker in 1959. Previously, corporations had large workforces with primarily manual skills; in the mid-20th century, corporations began building large labor forces with primarily thinking skills. These knowledge workers could not be managed with the same techniques used to manage manual workers. Drucker鈥檚 contribution was to define the knowledge worker and educate a generation about how to manage, measure and compensate folks who think for a living. Those frameworks for managing are still the backbone of organizational design today.

For AI to deliver on its promise as an innovation, it must move beyond chatbots of whom we ask questions. It will become systems which can actually make decisions. In common usage today, the term 鈥渁gentic AI鈥� means an AI system to whom you delegate a certain decision-making authority. (If the system cannot make decisions, it is merely a tool.)

What is an 鈥榚xecutive,鈥� anyway?

To understand AI systems that we trust to make decisions, let鈥檚 look back at how Drucker defined what makes a knowledge worker an 鈥渆xecutive.鈥�

“Every knowledge worker in a modern organization is an 鈥榚xecutive鈥� if, by virtue of his position or knowledge, he is responsible for a contribution that materially affects the capacity of the organization to perform and to obtain results.” 鈥� Peter Drucker, 鈥淭he Effective Executive鈥� (1967)听

Today the language on the future of AI applications uses words like co-pilot, agent or AI assistant, which implies a smart but junior and subservient employee. That is the wrong model. As we delegate decisions that matter to AI, which must happen for AI to earn its keep over the long haul, then those systems are not interns or co-pilots. They are, by definition, AI Executives.

The form innovation will take, making good the massive capital put into the invention听phase, will be the rise of the AI Executive.

Thinking about your AI system as an executive (or perhaps many executives) is a bit of a scary concept at first. Let鈥檚 look a bit closer at this analogy and see how it holds up.

One defining characteristic of a knowledge worker is that typically they know much more about how to complete their task than their boss. The international transfer tax pricing expert knows much more about how to do that specific task than the CFO does. The CFO can鈥檛 tell the knowledge worker exactly how to do their job, they instead set the goals and outcomes, provide the support, and appropriately reward the results.

Again, in the words of Drucker (鈥淢anagement: Tasks, Responsibilities, Practices鈥� 1973):

鈥淭he knowledge worker cannot be supervised closely or in detail. He can only be helped. But he must direct himself, and he must direct himself toward performance and contribution, that is, toward effectiveness.鈥�

Paying for outcomes vs. time worked

The analogy to LLMs is strong. No one knows exactly how LLMs reach the decisions they reach. We as the 鈥渂oss鈥� of LLM can only direct them toward effectiveness. Classic SaaS workflow software functions more like the manual skill worker, where the designer of the software knows how each action is taken and defines each branch on a decision tree. So far, so good.

Now let鈥檚 look briefly at how Drucker measured the knowledge worker (from 鈥淭he Effective Executive,鈥� again) and evaluate the analogy to the AI Executive.

“For the manual worker, we need only efficiency, that is, the ability to do things right rather than the ability to get the right things done. The knowledge worker is certainly not defined by quantity. Neither is he defined by his costs. He is defined by his results.”

With knowledge work, we measure, value and pay for the output. If the knowledge work is the design of a new fashionable sneaker, we want the best design. Whether the design was听 produced by one genius in the basement over a long weekend, or a team of 200 in a global product management hierarchy, we care about the best design, not how much labor went into it. In shorthand:

Physical labor –> measure by time worked

Knowledge work –> measure by output

The read-across to the shift from classic SaaS to Agentic AI, then, will be:

SaaS听 –>听pay per user per month

Agentic AI –> pay for outcomes

This change is revolutionary, not evolutionary, for the investment community which has been investing in SaaS software.

Considering just the sphere I know best, the private equity industry is by my count now leveraged long more than $1 trillion in the seat-license model. For a SaaS company with little customer churn and the ability to raise prices, the so-called net retention for Annual recurring revenue typically exceeds 100%. That is for an annual cohort of signed customers, the price rises on renewing customers offset the departing customers, so that annual vintage of customers in total has revenue that is flat to slightly growing over time.

And because it is recurring and has high margins, the lending community and the over time equity investing community began to think of those as bond-like streams of payments, against which loans are advanced and valuations are set.

Shifting from ARR to outcomes-based pricing is terrifying to some in that it pulls away the struts supporting this idea they are financing a bond-like stream of payments. If it is variable, even if in direction it is growing, that is much less popular to lend against or place a valuation multiple on. Moving from ARR to outcomes-based pricing will upend the foundation on which so much of the SaaS investment community has relied.

Delegating decisions to the machines

Let鈥檚 wrap this up with an example. We are indeed now seeing users for some of our AI software platforms beginning to feel comfortable delegating some decisions to the machine.

For instance, one of our portfolio companies, , provides inventory management software to midsized companies. Over the past few years, we have gone from 鈥渏ust鈥� really great inventory management analytics to a conversational AI interface that suggests the order to change or cancel. Users have grown to trust those recommendations.

Indeed, in the Netstock annual survey, 24% of users now say they would be comfortable completely delegating inventory decisions to Netstock AI, and another 50% say they would be willing to partially delegate, with oversight or shared control. This is a recognition that the Netstock AI engine has gotten reliably very good at the game of optimizing inventory, and so here some version of 鈥渉uman + machine鈥� seems a likely path of agentic AI adoption. Those employees can now turn their attention to other initiatives at their companies.

How do you pay for an AI system that is managing your inventory? It鈥檚 not per user per month, I think we can all agree on that as there are no longer users in the traditional sense. The answer is that the value is defined by the improvements in fill rate and reduced inventory carrying costs, hard data economic measures, which the customer can see and pay for on the basis of value received, not the number of employees using a software package. In considering pay, we must acknowledge some fundamental differences in the risk-bearing ability of human executives, who have mortgages and risk preferences, as compared to AI programs, which do not.

These are the great opportunities of our day:听to not just create but to manage AI Executives so that they in fact produce useful outputs in the world, and to price these AI Executives in a new way, graduating beyond the user-seat-license model.

It is possible. At Netstock, we took a great 鈥渃lassic鈥� SaaS company and introduced an AI agent that users grew to trust and rely upon. Today three quarters of the clients are willing to delegate their inventory management decisions to the Netstock system (and some to delegate completely).

Build software clients trust with their decisions. No longer a system of record, your software will become a system of agency. That is a huge opportunity and the one we are spending all our time and capital pursuing.

co-founded in 2014 and is managing partner. He has served on numerous private and public technology company boards, and currently is a director of , , , , and . Previously, he was a partner and member of the investment committee at . He also worked at and . Morse serves on the board of directors of and as member of the advisory board for the HMTF Center for Private Equity Finance at . He attended , graduating summa cum laude with a B.S.E., and , where he earned his MBA and was an Arjay Miller Scholar. Morse lives in Austin.

Related reading:

• Invention To Innovation: Making Sense Of AI鈥檚 Moment听

Illustration:

Over the years, I have experienced that sinking sensation of the investor-CEO disagreement in that first board meeting.

After the back-and-forth negotiation of the deal process, each side having played its hand, everyone gathers for that first board meeting and 鈥� surprise! One side or the other shares information or intentions not revealed pre-signing, in a way that can鈥檛 be easily reconciled between the private equity partner and the founder or CEO.

Through 25 years in private equity, I am not alone.

The nature of the frustration and emotion with the First Board Meeting Surprise is not that something arose post-investment: It is that something听knowable in advance of the deal听comes to light听after听closing and is a material change to expectations.

The surprises I鈥檓 referring to don鈥檛 arise from bad actors trying to intentionally mislead. In fact, it鈥檚 because they regularly occur among well-intentioned individuals that these surprises are so maddening.

Case in point

For instance, consider the operating partner who shows up at the first board meeting and shares vetted candidates to replace an executive the CEO feels loyal to.

The CEO wonders why he was kept in the dark about his new investor鈥檚 view on this key executive before signing. Perhaps, thinks the CEO, there are other things I wasn鈥檛 told. The new investors wonder why the CEO is reluctant to build an 鈥淎鈥� team. Perhaps, thinks the operating partner, this is not our kind of CEO. Neither side backs down; both feel justified. That is a First Board Meeting Surprise.

It鈥檚 embarrassing for all concerned, and it happens more frequently than we鈥檇 all like to admit. Research from found that roughly three quarters of portfolio company CEOs are replaced during a fund鈥檚 hold period, with a majority of those happening in the first two years.

While the sponsor and the founder/CEO want听to be aligned, the dynamics of the investment process work against them.

The purchase of a company is a one-time, distributive negotiation with large dollars at stake, and it can be adversarial, full of tension and tiring. On the other hand, the relationship between investor and CEO in operating and building a company is a repeat-player game. That game requires a completely different approach.

Stay prepared

Our solution to the First Board Meeting Surprise problem is to adopt repeat-player thinking before signing the deal. What would happen if our underwriting plans 鈥� due diligence findings, best 鈥渟ecret sauce鈥� ideas, and proposed actions 鈥� were shared with the founder and agreed to in writing before signing the deal?

It鈥檚 not without risks. The founder/CEO might see our work and insist on a higher price, shop those ideas to a competing bidder, or simply adopt our best ideas without taking our investment at all. The founder/CEO might disagree with the course of action entirely and simply walk away. We then lose a deal we could otherwise have closed.

But perhaps the founder/CEO would be听excited听about the action plan, provide input to improve it, and refocus their energy entirely on where we were going to go together. And, because nothing knowable in advance of the deal on听our听side would remain hidden, and reciprocal engagement from the founder/CEO on the plan would reveal听their听true views on critical items, there would be no First Board Meeting Surprise.

Putting this into practice can feel risky for those used to the traditional information-control approach to closing deals. It was for me. Just before founding , I was trying to recruit a CEO I held in high regard to lead a new platform investment we were evaluating.

He was wary, so I shared our underwriting and diligence. He insisted that we commit to building out a full product suite and adding to the leadership team, so I went to the investment committee and secured an upfront commitment for follow-on capital. Ultimately, we put a set of five actions down on a single page, shook hands, and only then signed the deal.

On the day the deal was announced, he shared that action plan in his all-hands meeting, and it formed the agenda for the first board meeting. grew several-fold in size to become a UCaaS leader. I have that one-pager framed on my desk today.

The five-point plan

We turned that approach into a process we call the Five-Point Plan, which requires us and the CEO to agree听up front听on post-transaction actions. That means we not only agree on the five key actions to take after the deal closes, but the CEO knows he or she has the resources and support to execute them. More importantly, we鈥檝e eliminated the problem of the First Board Meeting Surprise because no one has any surprises to share.

We have accepted that the price for materially improving sponsor-CEO alignment in the deals we听do听close is that we will lose some deals we could have closed but for agreement on the Five Point Plan. In practice, we have closed several dozen founder-led deals versus a handful of walk-aways. Both we and the founder are better off without doing those deals where we would have discovered a fundamental disagreement the day after closing.

Expecting the unexpected

Of course, the world always intervenes, and the moment the deal closes, events unfold: tariffs, interest rate changes, AI breakthroughs and so on. While we all expect the world to change around us, there are some surprises we can avoid.

We can minimize the risk of friction between the CEO and Board due to differing goals. We can begin our repeat-player relationship before signing the deal.

In the founder-led technology buyouts, we believe that a more transparent pre-signing investment process, like our Five-Point Plan, is the most promising way to begin a partnership between a private equity sponsor and a founder/CEO. I am sharing this approach in the interest of encouraging others to experiment with it. Consider showing your hand more openly before closing. Invite the founder/CEO to do the same. See what happens.

co-founded in 2014 and is managing partner. He has served on numerous private and public technology company boards, and currently is a director of , , , , and . Previously, he was a partner and member of the investment committee at . He also worked at and . Morse serves on the board of directors of and as member of the advisory board for the HMTF Center for Private Equity Finance at . He attended , graduating summa cum laude with a B.S.E., and , where he earned his MBA and was an Arjay Miller Scholar. Morse lives in Austin, Texas.

Illustration:

Amidst the claims about artificial intelligence that surround us all today, what can we confidently say at this early moment? The first step is to choose the right frame.

This comes from Joseph Schumpeter, the pioneer of entrepreneurship economics, who wrote in 1942: 鈥淚nnovation is the market introduction of a technical or organizational novelty, not just its invention.鈥�

That is, we can frame the collection of inventions we refer to as 鈥淎I鈥� as distinct from the innovations that result from their introduction into the market as commercial products. Invention is about creating new capabilities, whereas innovation turns those capabilities into tangible value.

Schumpeter developed the concept of 鈥渃reative destruction,鈥� or the process by which new innovations destroy long-established practices. Creative destruction is a driving force of growth in a capitalist economy.

Today, AI is that engine of creative destruction. It is an amazing invention that allows things to be done that were not possible before. AI right now is moving from invention to innovation, following the pattern of classic business transformations.

• The steam engine was an invention, and the steamship was an innovation, as it used that engine to open inshore waterways and transform trade.
• The internal combustion engine was the invention; the Model T assembly line was the innovation.
• Oversimplifying a lot, , who created 鈥檚 original PC hardware, was the inventor; , who created a mass market product through design, marketing and distribution, was the innovator.

Right now, AI is the invention. CEO has . The question is where we can put it to work to solve a problem with some economic value. It can be as simple as something that saves some time on a repeated task, or as novel as a medical breakthrough.

Let me share one recent example of turning AI inventions into innovative services from our portfolio at . provides supply chain planning software. Companies use Netstock to forecast demand, avoid stock-outs and reduce excess inventory. It is a SaaS application that an analyst uses to optimize and re-order inventory in real time.

A challenge for customers is that performing inventory analysis requires significant training for the user to understand the trade-offs and to adapt to changing conditions. In the past, only experienced staff knew enough to correctly resolve the choices between too much and too little inventory. But AI is well-suited to analyze data on successful and unsuccessful choices. Netstock built a customer AI tool to do this work based on its trove of data, both past and constantly updating.

Netstock applied AI to expand the set of users beyond those with deep supply chain expertise. It created a conversational AI experience, , that makes a specific recommendation to the user in natural language. The user receives a proactive message such as: 鈥淵ou鈥檝e ordered too much of item #7971, you should cancel that order and save $7,800.鈥� The user can still verify the analysis directly, and that has built trust in the recommendations over time.

Customers love this AI functionality. Users appreciate the time it saves and the improved outcomes it delivers. It has changed how they work with the software and how they manage their inventory. Recently, Netstock received top recognition for this tool from an independent, tell-it-like-it-is .

The Opportunity Engine is included in the standard offering, introducing the invention to customers in an easy-to-adopt format. Netstock has delivered more than 825,000 opportunities to customers since its release. In late January, Netstock unveiled its new , additional AI capabilities that allow users to interpret dashboards and analyze and troubleshoot individual items. Netstock is now charging for these capabilities and generating incremental revenue 鈥� moving from invention to innovation.

The framework of innovation vs. invention can also help to explain why we hear such a wide range of opinions about AI, from effusive and idealistic to skeptical, and at different points on the spectrum from inventor to innovator.

Many are writing about AI as an invention, including in his book 鈥�.鈥� In contrast, the startup economy has produced some coverage of AI as an innovation, as in an by of that describes changing the market pricing for AI solutions from a seat-license model to an outcome-based model.

Early days

But AI as innovation has not yet arrived as the main event in private equity, where capital values proven business models, not just the promise of innovation.

I have asked each of the technology bankers I have met over the past several months, 鈥淗ave you seen anybody in private equity make money on AI yet?鈥� Without exception, the answer is 鈥淣o.鈥�

One banker said: 鈥淎I is not even in our valuation matrix, because it is not showing up in the income statement.鈥�

When we explained how one of our companies is in fact selling their new AI offering with real uptake by customers and that it has accelerated revenue growth, the banker replied, 鈥淲ell then, you are in the 1% of companies.鈥�

What this suggests is that the market introduction of AI 鈥� the innovation 鈥� is still so early that the capital markets aren鈥檛 seeing it show up in financial statements in any broad or consistent way. Instead, markets are tracking the trend and watching to see when innovations mature. It is still early days.

The entrepreneur鈥檚 moment

So, this is where we find ourselves at this moment. AI is by any measure the major invention of the day. For those of us leading, investing in or developing businesses, the opportunity of the moment is to translate that invention into innovation.

The fact that AI is not 鈥渞eady for prime time with our customers鈥� is not the problem. It is the opportunity. AI inventions are broadly available now to us all, and the insight about how to apply it to a business problem is the entrepreneurial moment for each of us and our organizations.

Where can you use it to solve a problem you understand?

An entrepreneur is not someone who starts a company or gets angel financing. It is someone who puts an invention to work to produce a market outcome.

Or, as Schumpeter writes: 鈥淭he function of entrepreneurs is to reform or revolutionize the pattern of production by exploiting an invention.鈥�

Have you put AI to work on some activity that makes a difference to your revenue or cash flow? Well, if you have, you belong to that group Schumpeter so admired. In other words, congratulations 鈥� you are an entrepreneur.

co-founded in 2014 and is managing partner. He has served on numerous private and public technology company boards, and currently is a director of , , , , and . Previously, he was a partner and member of the investment committee at . He also worked at and at . Morse serves on the board of directors of and as member of the advisory board for the HMTF Center for Private Equity Finance at . He attended , graduating summa cum laude with a B.S.E., and , where he earned his MBA and was an Arjay Miller Scholar. Morse lives in Austin, Texas.

Illustration: