In January 2026, Anthropic released the massive ecosystem launch of Claude Cowork. That was followed in February 2026 with the launch of a set of industry specific plugins, which triggered one of the sharpest repricings the software sector has ever seen. Nearly a trillion dollars of market value evaporated from software and services companies in a handful of sessions. It was dubbed the “SaaSpocalypse.”

The logic behind the panic is not irrational. If a general purpose model can read your files, write documents, pull data, and orchestrate teams of sub agents, why pay for a seat in a legacy application? If AI coding tools let a product manager or even someone who isn’t technical at all stand up a working internal tool in an afternoon, why buy vendor software at all? The cost of creating software is collapsing toward zero. For software investors and founders, that is a genuinely destabilizing idea.

We think that framing is directionally right about one thing and wrong about the thing that matters most. The narrative mistakes the easy part of software for the whole of it, and in doing so, it accidentally makes the strongest case for why software is and will continue to be a great place to invest.

“Build vs Buy” Was Never About the Cost of the First Build

The most useful way to see through the panic is to actually run the build vs buy decision the way an operator does. Jamin Ball lays out the argument clearly. The “everyone will just build their own” argument isn’t new; it’s the same calculus enterprise operators have run for decades, just dressed up in AI clothing. The answer has always come back the same way: building is rarely the hard part. Companies have always been able to build; they most often choose not to. What’s hard is everything that comes after. Vendors offering specialized software aren’t just selling a product; they’re selling the accumulated R&D, the security and compliance infrastructure, the customer success apparatus, and the operational accountability that a homegrown tool has to replicate in full before it’s a real substitute. AI makes it faster to write the first version, but it doesn’t make everything else that comes with building software cheaper.

The initial build is the smallest part of the commitment. Depending on the complexity of the system, the first working version might represent a fifth or less of the total effort a company will eventually spend; the rest goes to keeping it running, adapting it as requirements change, patching vulnerabilities, and managing the knowledge transfer problem every time someone who understands the system moves on. AI compresses the first chapter of that story, not the rest. And when hundreds of companies each rebuild the same internal tooling independently, the waste is staggering, not just in engineering hours, but in organizational attention, operational risk, and the compounding cost of maintaining something that generates no competitive advantage.

There’s another cost that’s easy to underestimate: inference. Token costs at low volumes feel negligible in a demo; at production scale, across hundreds of employees running dozens of daily workflows, they compound quickly into a meaningful line item. A software vendor amortizes those infrastructure costs across thousands of customers, continuously optimizes for efficiency as new models ship, and absorbs the operational burden of managing rate limits, latency, fallbacks, and model upgrades. A company running its own internal tooling bears all of that alone, and unlike a SaaS seat, the bill scales directly with usage. Building gets cheaper every month; owning the infrastructure behind what you built does not.

The early data backs this up. Retool’s 2026 Build-vs-Buy survey found that 35% of enterprise teams have already replaced at least one SaaS tool with a custom build, and most plan to build more. But the categories being displaced are telling: internal admin tools, BI dashboards, simple workflow automations, the shallow, generic, horizontal layer. Meanwhile, the failure rate of ambitious AI initiatives remains sobering: Gartner predicts over 40% of agentic AI projects will be canceled by the end of 2027, citing escalating costs, unclear business value, and inadequate risk controls. Industry surveys keep finding that the overwhelming majority of agent pilots never reach production. Most ambitious AI deployments don’t stall because the model or the code wasn’t good enough. They stall because the underlying data was messier than expected, the integrations more fragile, the upkeep more time consuming, the governance questions more fraught, and the domain specific edge cases more numerous than any general purpose tool anticipated. Those are organizational and institutional problems, not technical ones, and they don’t get solved by shipping a better model.

The real risk to incumbent software isn’t the DIY internal build. It’s that when the cost of creating software goes to zero, ten well funded teams can rebuild a category from scratch, AI native, priced for the new world, and commoditize the incumbents. This isn’t to say that all software is safe. Generic, easily replicable software will be exposed, which has always been true. What protects a business is specificity, and that’s a case we’ll make below.

What Actually Stays Defensible

Strip away the code and ask what a software business is really made of. The value of software was never in the lines of code themselves; it was in the accumulated understanding of how a specific set of people do a specific kind of work, translated into something durable and repeatable. The things that make software hard to displace; deep customer relationships, institutional trust, distribution earned over years, and a product shaped by thousands of hours of customer feedback, have nothing to do with how the code was written, and don’t get cheaper to replicate just because the next model shipped.

That reframing points directly at the moats that survive.

• Proprietary, compounding data. Not “we have customer data,” but data that feeds back into the product and improves it in ways a new entrant starting from zero cannot replicate. The key isn’t just accumulating a data lake; it’s inserting your product at the point where data is first created. When you automate a previously manual workflow, you generate an entirely new data exhaust that never existed before: you build the dataset by doing the work. Abridge’s proprietary dataset is derived from more than 80 million medical conversations; Harvey works across legal documents in dozens of countries; EvenUp sits on hundreds of thousands of injury cases. Every time a denied claim is analyzed, every time our portfolio company Freight Hero handles a freight booking call, every time our portfolio company Superpanel processes a legal intake, the product learns. That learning compounds. Over time, the product that has processed 100,000 intake calls in a specific vertical has a structural advantage that cannot be replicated by a new entrant, not because the features can’t be cloned, but because the data cannot.

• Systems of record and workflow embedding. The stickiest software in the SaaS era was the system of record, the place where every other workflow ran through. Replacing it meant touching everything, and that switching cost compounded quietly for years. The equivalent dynamic in the AI era belongs to whoever becomes the governance layer for agents: the system that decides what an agent is allowed to know, access, and do. That’s a fundamentally different kind of lock in than storing data; it’s about controlling the rules of engagement for autonomous action. The broader shift underway is from software that records what happened to software that decides what happens next, actively reasoning over accumulated data and executing under defined guardrails. The companies best positioned for that shift are the ones that inserted themselves early, at the point where work begins, where the first form is filled out, where the first call is logged, where the first document is created. Own that entry point, and every downstream step in the workflow becomes a natural expansion surface.

• Speed and distribution. No moat lasts forever; they have always been fleeting, that has always been true, and AI hasn’t changed the rule, only the pace. What used to be a durable advantage for 12 to 18 months can now be approximated by a capable team in weeks. The implication is that defensibility isn’t a destination you arrive at; it’s something you have to keep earning. The companies with staying power aren’t necessarily the ones with the best technology; they’re the ones that have already embedded themselves in how customers buy, procure, and operate. Enterprise sales cycles in complex verticals routinely stretch from six months to a year. Legal, security, and procurement reviews don’t compress just because a better product exists. A customer who signed two years ago and has since trained their team, integrated their data, and built workflows around your product isn’t switching because a competitor shipped a sharper demo. That accumulated friction, earned through real deployment, not feature development, is one of the few advantages that genuinely doesn’t clone.

Notice that none of these are “we have a better model.” As Harvey’s own trajectory shows, the company orchestrates multiple models under a legal specific layer; the model is increasingly the commodity. The moat is everything wrapped around it.

Why Vertical AI, Specifically, Is Built for This World

Every one of those durable moats is easier to build and deeper to hold in a vertical market than in a horizontal one. That’s not a coincidence; it’s structural, and it’s why the “AI eats software” story is, for vertical AI, closer to a tailwind than a threat.

Depth is the defense against commoditization. Horizontal tools optimize generic workflows that a good model can approximate. Vertical products encode the messy, specific, regulated reality of one industry: the multi step workflows, the payer specific denial rules, the code compliance edge cases, the union specific payroll logic that a general model gets 80% right and 100% wrong in the ways that matter. Vertical AI companies compete on proprietary data, workflow depth, and compliance, precisely where “good enough” DIY software is at its worst.

The prize is labor, not software budgets. This is the point we’ve hammered at Field Ventures, and it’s now the consensus framing: traditional SaaS captures a sliver of an employee’s value; vertical AI can capture a large share of the labor itself. This reframes the commercial conversation entirely. When your customer’s alternative is a full time employee or an offshore BPO contract, the calculation is not abstract: each of my intake coordinators costs $65,000 per year and makes mistakes. Your software does the same work with a lower error rate, at half the cost, and is available 24/7. That is not a productivity pitch. That is a P&L pitch. The ROI is unusually clear, and it unlocks budgets that were never available to SaaS vendors before.

Under digitized industries leapfrog straight to AI. The antiquated verticals we love, construction, logistics, healthcare administration, the trades, and many others, run on slim IT budgets and enormous labor budgets, with workflows still living in spreadsheets, faxes, and phone calls. They were never going to build their own software; they barely bought the last generation of software. And importantly, the incumbent in these markets is often a spreadsheet, a phone call, a manual process, or software built in 2003 that has barely been updated since. There is no sophisticated incumbent to displace. There is only operational drag that the buyer experiences every single day, and they would gladly pay to eliminate it.

Agents accumulate context; that context becomes a moat. An agent, like an employee, gets more valuable the longer it’s deployed and the more institutional context it absorbs, a genuinely new switching cost that compounds over time. Unlike traditional software, which does the same thing on day one as it does on day one thousand, an AI agent deployed in a real production environment gets better over time, absorbing institutional quirks, edge cases, and workflow preferences that no new entrant can replicate from a standing start. The longer it runs, the more it knows about how this billing team handles denials, how this logistics operation manages exceptions, and how this legal practice structures intake. That accumulated context is a switching cost unlike anything the SaaS era produced, because replacing the software means losing the institutional memory embedded in it. That is a moat that accumulates quietly and compounds over time, one that didn’t exist in the SaaS era and can’t be bootstrapped by a new entrant overnight.

Domain experts are now the builders. Domain expertise has become more important, not less. AI has also changed who can build. A solo founder with Cursor, Replit, and Claude can ship in days that would have taken a team of four six months to build two years ago. That democratization matters; it means domain experts (the paralegal who has spent fifteen years in mass tort litigation, the billing specialist who has processed ten thousand Medicare claims) can now build the tools they wished had existed, without needing a technical co-founder to translate their vision into code. But the consequence is that a compelling demo is no longer a signal. When anyone can build an impressive product in a week, it tells you almost nothing about whether you have a compelling business. Defensibility moves upstream: to workflow depth, to trust, to distribution. The founders who win will be the ones who got to the right problem before anyone else, because they were living inside it.

The Counterarguments and Why They Don’t Break the Thesis

“The foundation models will just enter the biggest verticals.” They’re already doing it. OpenAI and Anthropic both launched healthcare offerings in early 2026, and OpenAI has signaled a legal product. This is real, and it caps the upside in the largest, most horizontal adjacent use cases. But it also validates the verticals, and the labs consistently take the low hanging fruit while struggling with exactly the things vertical players own: messy data integration, industry specific compliance, editorially maintained authoritative data, and multi party workflows. The durable answer isn’t “out model the labs”; it’s staying model agnostic and building the data, workflow, and trust layer they won’t.

“The real threat is the next AI native startup, not the incumbent.” This is the version of the risk we take most seriously. When the marginal cost of shipping software collapses, a small, focused team with access to the same frontier models can close what used to be a meaningful feature gap in months rather than years. First mover advantage is a much weaker defense than it used to be, and at the earliest stages, moats are non-existent today. This is precisely why structural advantages are crucial: position over product leads, proprietary data that accumulates and compounds with usage, regulatory credibility that takes years to earn, distribution embedded in how customers already buy, and workflow depth that a new entrant would have to replicate from scratch. Features are a snapshot. Structural advantages compound.

“Foundation models will commoditize every application built on top of them.” This critique is accurate about one category of company and largely irrelevant to another. If your product is a thin interface sitting on top of a foundation model with no proprietary data, no workflow depth, and no switching costs, then yes, the model provider will eventually absorb your use case. The market has already priced that in; capital has moved away from pure interface plays. But that description doesn’t fit companies that own a critical workflow, carry irreplaceable customer data, and have embedded themselves in how an industry actually operates. Those businesses aren’t wrappers; they’re the layer the model runs inside of.

“Incumbents will catch up.” Some will; many are trying. But most legacy application vendors have, so far, shipped little meaningful AI despite having every structural advantage, which is itself an opening for AI native vertical entrants who build the system of work, not just the system of record.

What We Look For

After assessing the founding team, where we believe deep industry insights are more crucial than ever, we look for three things: a convincing wedge, durable differentiation, and operational reliability.

The wedge is the entry point, the first, specific workflow the product solves, narrow enough to demonstrate value clearly, valuable enough to command budget, and positioned close enough to revenue generation or cost elimination that the buyer can see the ROI immediately. Wedges are easier than ever to create. The wedge alone means nothing.

Durable differentiation comes from domain depth and data positioning. The best vertical AI companies are not just automating tasks; they are inserting themselves into the source of data creation. The companies that build and maintain defensibility are the ones that sit at the data creation layer, where the first input is created, where everything downstream flows. Win that entry point, and you earn the right to expand into every subsequent step in the process. We are looking for founders who understand the human workflow deeply enough to automate it appropriately, not just what tasks can be handed to software, but which tasks must remain human in the loop, and why. The best founders in this category can draw a clear line between what AI can handle today and what requires human judgment, and they can articulate exactly where that line moves as the technology improves.

Operational reliability is perhaps the least celebrated but most important evaluation criteria. A demo in front of a sophisticated buyer is table stakes. What separates a pilot from a platform is whether the product performs reliably under actual production conditions, with all the edge cases, regulatory requirements, and operational messiness that never appear in a controlled demo. In verticals like healthcare, legal, and insurance, buyers are not asking whether the AI is capable. They are asking whether they can trust it. Switching costs are not an accident. They are a product design decision.

The Opportunity That History Keeps Confirming

The February 2026 selloff will, we suspect, be remembered the way the dot com and cloud transitions are remembered: a moment when the market correctly sensed disruption and wildly mispriced its shape and timing. The market has already started grading that mispricing. Situational Awareness LP, the AI focused fund built around the thesis that foundation models would gut software incumbents, had constructed exactly the trade this narrative implies: long AI infrastructure, short software names like Adobe. Last week, that position unraveled fast enough that the fund had to offload roughly $16 billion of its long and short book to Citadel in a single block trade just to meet margin calls. Assets that stood near $45 billion at the start of the month finished around $10 billion. The fund that bet hardest on the “software is dead” thesis was also the fund forced into a fire sale by it, a costly real time reminder that being early to the right narrative and being right about the trade are not the same thing.

Every prior platform shift ended with more software, not less. The PC didn’t kill the mainframe; it expanded the total compute market by orders of magnitude. eCommerce didn’t kill retail; it restructured it, creating trillion dollar outcomes on both sides. Streaming didn’t kill media; it created more content, more consumption, and more business models than anyone predicted. In each case, what died was the shallow, generic, undifferentiated middle. What thrived was whatever owned the critical data, orchestrated the critical workflows, and earned the trust to act.

That is a near perfect description of vertical AI.

AI making software cheap to build doesn’t erode the vertical thesis; it turbocharges it. It collapses the cost of automating a workflow while leaving everything hard about a specific industry entirely intact: the data, the regulations, the integrations, the trust, and the domain knowledge locked in experts’ heads. The build vs buy calculation has not been repealed by AI coding agents. It has been sharpened. What gets commoditized is generic, shallow, horizontal software. What gets more valuable is deep, industry specific software that encodes regulatory knowledge, owns proprietary workflow data, sits at the data creation layer, embeds in systems of record, and carries the trust to act autonomously.

And what gets truly valuable, in a way the SaaS era never allowed, is software that stops competing for a line on the software budget and starts competing for a line on the labor or operations budgets, priced against headcount, not seats.

The markets that were too unglamorous to attract the last generation of software builders are, for that exact reason, the most fertile ground for this one. The industries furthest behind are the ones with the most to gain and the fewest entrenched software incumbents to dislodge. The build vs buy head fake is real. But for founders and investors who know where to look, what it reveals is not a threat. It’s a map.

At Field Ventures, we’ve backed domain experts selling outcomes into labor budgets in under digitized verticals since before “generative AI” was a headline. We believe it more today than when we started. If you are building along this thesis, please reach out!

At Field Ventures, we continue to be excited by the evolution of AI in antiquated / real world industries. Unlike tech companies and early adopters, these industries are increasingly AI-curious, but exactly how and where to implement AI remains a challenge. We strongly believe that for these industries, services-as-a-software vs pure AI technology is crucial to adoption and meaningful business impact. Companies that don’t just give you the tools but instead provide you the outcome, enabling actual efficiency, cost cutting, and revenue growth.

Freight Hero is a clear example of this type of true customer impact in the freight brokerage industry. Field Ventures is excited to back Andre Martins Filho, Founder and CEO, and the Freight Hero team, leading their $5m seed round alongside Flybridge, Tip Top, Dynamo and AI Fund.

The freight brokerage industry is one of exceptions and customization. Each customer and driver has their own way of operating, communicating, and record keeping, meaning a business must tailor each interaction with a customer in a unique capacity.

This makes training an AI to solve the real challenges - tracking, communicating with customers - an uphill battle. AI is terrific at pattern matching, at learning from previous actions, styles and communication. But when those interactions and needs are dynamic, and are high stakes from a business - customer relationship, AI struggles.

Freight Hero takes the best of AI’s capabilities to automate the end-to-end load management process, from rate confirmation through proof of delivery (POD). However, the company has “Heroes”, a 24/7 team of trained freight specialists who are able to handle outliers or unique situations, following each individual broker’s standard operating procedures, without the brokerage ever having to step in.

The need for this is something we heard repeatedly talking to owners of brokerages. AI alone can act like co-pilot today in their workflow right now, trimming time; however, the impact is not enough that they can have their labor focus on more high value tasks. With Freight Hero, brokerages are able to trust the work is getting done, without worrying about hallucination, incorrect communication with customers, or frustrated drivers.

Freight Hero is starting with end-to-end load management, an extremely manual and time consuming process. Current processes can cost north of $20 per load, in an already razor thin margin industry, and consume almost 90 minutes of average dispatcher time per load. Freight Hero and its “Heroes” are able to drastically reduce this to a greater than 25% reduction in cost, 10x faster issue detection, and over 1 hour of time saved per load. This means not only reduced costs and better use of labor, but significant ability to increase throughput. While AI can handle 90% of the tasks, the “Heroes” are able to jump in to manage communication that may be out of the norm such as verifying location with incomplete data and time sensitive driver challenges.

One of Freight Hero’s customers, FortFreight, offers a compelling proof point: before Freight Hero, the broker had to make 30 to 50 update requests per load before 9am daily, requiring constant manual checkins. After onboarding, incoming check call volume dropped to zero “practically overnight,” the team shifted from manually triple checking every shipment to focusing on relationship building and new business development, and the engagement drove a $260K cost impact, an 80% reduction in back office work, and a 75% reduction in outbound calls. As FortFreight’s CEO Eddie Fortenbacher put it: “It’s been an absolute game changer.”

Andre’s expertise and intimate knowledge of the customer is the driver of this differentiated and meaningful approach. He previously co-founded Uello, a logistics tech company in Brazil and understood from the ideation stage of Freight Hero that his customers want the work done, not just another tool they need to manage to do it.

At Field, we are incredibly excited to back Freight Hero and their vision to transform the $20bn brokerage industry that is essential to our economy.

.

At Field Ventures, we’ve written about our excitement around the shift to Service-as-a-Software and our belief that software is evolving from a tool into a digital worker. We are especially excited about this shift when it comes to the most people heavy industries or business functions. Industries and business operations that rely on labor-intensive, repetitive tasks represent a huge opportunity for AI. AI workers can increase revenue by working faster and more efficiently than their human counterparts, expanding productivity without increasing headcount. While today we see them more as working alongside versus replacing current in-house teams, we are already seeing positions not being backfilled and businesses recognizing the opportunity, especially when it comes to outsourced teams. One area we see as a massive opportunity for this type of AI disruption is business process outsourcing (BPO).

The BPO industry is massive, with US businesses spending over $130bn on third party BPO services. This is across a wide array of business functions and industries, including healthcare, HR, finance/accounting, IT, and customer experience. The space is dominated by 3 key leaders, Accenture, Cognizant, and TCS, who are some of the largest employers globally, typically serving Fortune 500s. Yet, they only control about 22% of the total market, highlighting how fragmented this space is and the opportunity even in the long tail.

Across the market, these workforces perform highly manual and repetitive tasks that are often voice or data entry heavy workflows. This spans across customer service calls, scheduling appointments, qualifying leads, processing claims, and managing inventory. However, even the largest providers have challenges. Margins typically run 10-20%, and operations struggle with high turnover and quality inconsistencies.

We believe voice AI can fundamentally reshape this industry: massive margin expansion, fewer employees, greater accuracy, and higher throughput. We’re excited to back companies building software for existing BPOs, enabling a faster and more impactful shift to AI-enabled services. But we also see enormous opportunity for new, AI-native BPOs built to serve different industries and workflows more efficiently, accurately, and at scale.

Why Traditional BPO Models are Straining

Traditional BPOs have always operated on an arbitrage basis. Businesses rely on third parties for services that would be more costly to handle internally in terms of labor, time, or expertise. The BPOs themselves rely on labor arbitrage by staffing operations in lower-cost geographies. Their success depends on handling as many calls or requests as possible at the lowest cost. But this model has fundamental constraints: inconsistent processes, fragmented and manual systems, and heavy reliance on repetition that leaves significant room for errors.

1. Increasing Cost Pressure and Regulatory Shifts: Even with offshore labor, personnel costs represent 60-70% of revenue for most BPOs. US-based BPOs can charge about $10 more per agent per hour, whether for basic call centers or more specialized knowledge providers. However, with similar quality outcomes, that additional cost is hard to justify to customers. Further, rising inflation and strengthening employee protection regulations are forcing BPO businesses to pay more for staff. For example, many countries have placed restrictions on the number of hours worked as well as raising minimum wage and required benefits and the US recently proposed the Keep Call Centers in America Act, all of which increase overall labor costs dramatically.

2. Labor Shortage and Employee Retention Challenges: Employee turnover in the BPO industry averages around 60% across all BPO types. The industry also faces a significant labor shortage, leading to longer hiring cycles and expensive onboarding and training processes that must be repeated constantly.

3. Inconsistent Quality: Given high turnover rates and extensive training needs, maintaining consistent quality is a persistent challenge that requires additional investment in monitoring and QA systems. This is especially critical in regulated industries like healthcare and finance, where errors can have serious consequences.

4. Scaling Challenges and Shifts in Customer Demand: Winning a new client means recruiting, training, and onboarding hundreds or thousands of agents. Meanwhile, clients are demanding better integrated systems, faster response times, and more personalized service. This creates both software and operational scaling challenges for existing BPOs as they try to adapt.

The core operations of BPOs rely on labor, making revenue growth almost as much a cost burden as a benefit. The inability to decouple growth from headcount puts fundamental constraints on the market, worsening as client reliance and demands continue to escalate.

How Voice AI can Transform the BPO Industry

As voice AI continues to improve, the days of frustrating conversations with automated voice response systems will be long gone. Instead, voice AI provides natural language capabilities, able to understand complex requests and handle increasingly complex tasks without human escalation. AI can also enable much more customized responses, tailored to a specific business’ style as well as complying with industry specific regulatory requirements.

The speed at which AI can access data, respond, and execute necessary next steps at 10-20x the rate of a human counterpart. AI is also continuously learning from previous interactions. Where initially, in early, more complex conversations, it may need to loop in a human, as the AI learns from those conversations, it is increasingly able to handle tasks on its own. It can also operate 24/7, allowing for round the clock execution and greater throughput without the additional cost.

Three Opportunities for New Entrants: In-house, AI-Enabled, and AI-Native BPOs

As voice AI continues to proliferate the BPO market, we see three compelling paths for innovation: companies bringing voice AI in-house, software providers selling to existing BPOs, and entirely new AI-native BPO firms.

1. AI Decreasing Need for Outsourcing
   For most companies today, outsourcing is the only cost-efficient way to handle high-volume workflows and customer demand. However, AI is making it exponentially easier, cheaper, and faster to scale operations without adding headcount. Businesses can deploy their own voice AI agents to work alongside current employees, expanding throughput massively while maintaining seamless human intervention for exception cases.

   

   For many companies, adopting voice AI agents will be a simple calculation of cost reduction without sacrificing customer experience. These companies are making that calculation easy by providing the tools to easily set up agents. We are already seeing this across appointment scheduling, insurance verification, customer service, and returns, to name a few categories. Rather than paying for offshore labor at $15-25 per hour per agent, companies can handle the same volume for a fraction of the cost with AI.

   
   The demand could be enormous from small businesses that don’t want to manage service providers and prefer control over operations, to large enterprises looking to scale without additional headcount. However, we do see challenges persisting. Displacing existing BPO relationships takes time, and the setup needs to be streamlined enough that even less tech-savvy businesses can customize the infrastructure themselves.

2. Bringing Automation to Existing BPOs
   While we see a huge opportunity for bringing more operations in house efficiently, that doesn’t mean the end for BPOs. In fact, we see a huge opportunity for AI to transform how BPOs operate and scale, making them more sustainable businesses better suited to meet evolving customer needs. A BPO that previously required 1,000 agents to handle 50,000 calls per day can now manage the same volume with a fraction of agents, or handle multiples more volume with the same headcount. These software companies provide a clear ROI to the BPOs, making the sale compelling and easier to get to a “yes”.

   

   
   From a GTM perspective, this is an ideal approach for new companies. By selling directly to the BPO, you get immediate access to all of their customers. This means larger end customer / revenue opportunity at the onset versus going customer to customer. Plus access to large datasets with pre-existing edge cases to train on. Some of the larger, more dominant BPOs are already integrating AI through internal development and acquisitions. While selling to these players presents a higher hurdle given their scale, the massive long tail of mid-sized and smaller BPOs provides ample opportunity to build a venture scale business.
   

   This route also gives new entrants optionality on their future business model. Given access to comprehensive data, these software businesses can learn at a faster, more comprehensive rate, potentially positioning themselves to become more efficient BPOs.

3. AI-Native BPO Services
   Finally, we believe there’s a massive opportunity to build new AI-native BPOs from the ground up. Many customers prefer to outsource but still want the customization and efficiency that typically comes with internal operations. It is hard for an existing company to completely upend theirs for efficiency’s sake. However, starting from scratch, an AI native BPO can maximize the use of AI agents, looping in humans only for complex exceptions while ensuring high quality and lower costs than traditional models.
   

   Starting fresh removes friction around legacy systems and fragmented data. Instead, new BPOs can build unified systems that deeply understand their customers’ needs and the complexity of the industries they operate in. To provide truly top-tier value early on, AI-native BPOs will need to focus on specific verticals or use cases to ensure the best customer service and quality.
   

   AI native BPOs will create significant data advantages over time. A technology-first approach enables a centralized hub for all use cases, calls, and workflows that continues to compound, but, crucially, unlike the systems that exist today, this data is actually structured and usable to systematically improve services.

   

   These new BPOs face the same challenges as the software direct to companies in that GTM is more of a 1:1 motion, and they will likely need to displace existing providers. However, their ability to demonstrate cost savings and superior service quality without requiring deep technical integration can significantly shorten sales cycles.

Key Considerations and Challenges

While we see huge innovation and opportunities across the BPO market, there are still several key hurdles as the technology proliferates:

1. Voice AI Technology Continuing to Improve: While voice AI has improved dramatically in a short time, there is still significant room for improvement. Accuracy, proper communication with humans, and real empathy and judgment are complex challenges for AI models. Currently, voice AI works best with structured, repetitive tasks that meet many of the requirements of BPOs and includes many tasks better suited for AI vs humans. However, to truly disrupt the space at scale, the technology will need to continue advancing.

2. Friction in Changing Existing Operations: BPOs are extremely sticky services. As they work with a customer, they build trust and deep knowledge of the customer’s operations. They also require system integrations that often come with upfront implementation costs, creating both economic and technological sunk cost challenges. New entrants, whether direct to customers or a new BPO, need to quickly demonstrate meaningful value in cost savings and revenue generation to convince customers to switch.

3. Reluctance Reliance on Full Automation: BPOs often handle critical workflows, communicate with clients, and operate in highly regulated industries. Because of this, customers can be wary of removing humans from the loop entirely. However, as businesses continue to implement AI across other parts of their operations, comfort levels will increase.

4. Changing Competitive Dynamics and Economics: The promise that AI can help to lower costs is extremely attractive for end customers and BPOs, but it could also lead to significant pricing pressure. As existing BPOs evolve to integrate more AI, margins could compress before they expand. However, winners who can truly impact growth and efficiency will be able to differentiate and maintain pricing power.

Building a Hundred Billion+ Business

At Field, we’re actively seeking founders building voice AI solutions for the BPO ecosystem.

We believe some of the best companies will have a vertical focus and will be led by founders with deep sector expertise and exceptional talent, both of which are crucial for pricing and regulated workflows. We also see business models shifting from per employee pricing to outcomes based, providing greater alignment between the revenue generation and costs.

We see the BPO industry at an inflection point. Existing players will be forced to adopt AI or fall behind their evolving customer base and new entrants. There is an ability to drastically alter the economics of the industry will improve quality of service businesses and end customers see. Business and BPO providers, both old and new, will be able to significantly increase their capture of revenue.

For founders, this is a generational opportunity in transforming a massive industry at the intersection of labor and technology.

At Field, we are actively seeking founders building in this space. If you’re working on voice AI for BPOs or related opportunities in this space, we’d love to talk.

At Field Ventures, we invest in vertical software businesses that modernize antiquated industries. Today, we’re witnessing one of the most compelling opportunities in healthcare technology: revenue cycle management (RCM). With over $280 billion flowing through the U.S. healthcare billing system annually and administrative costs reaching $300 billion per year, the RCM space represents a massive market ripe for disruption.​

The numbers tell a stark story. Hospital operating margins have been declining for years and now sit at 2.6%, while claim denial rates have now reached 15%. Rising administrative costs, denials, and slower payment cycles are leading to increased costs for patients. This perfect storm of financial pressure creates an urgent need for innovation and a massive opportunity for founders building the next generation of RCM solutions.​

The Broken RCM Status Quo: A $300B Inefficiency

Healthcare revenue cycle management encompasses the entire journey from patient scheduling to final payment collection. The process remains largely manual, involving the collection of patient information, clinical documentation, medical coding, claims submission, and denial appeals. Each step requires multiple human touchpoints, creating countless opportunities for errors and delays.​

Below is an illustrative example of the entire RCM process.

Consider the scale of inefficiency: on average, industry experts believe hospitals leave 1% - 3% of revenue on the table due to coding errors alone. Given that the average hospital in the US generates ~$240 million annually, that’s millions in lost revenue, enough to fund an MRI machine or additional nursing staff. Payment cycles commonly extend beyond 120 days, which leaves an average provider waiting months to be reimbursed by insurance for services already delivered. For many providers, this creates a cash flow crisis that threatens operational stability.

Labor shortages across billing and coding roles have compounded the issue, with 90% of healthcare systems reporting an RCM labor shortage. To keep up, 77% of organizations rely heavily on offshore teams or RCM service firms. These teams are expected to manually navigate evolving payer rules and increasing documentation requirements across fragmented systems, an approach that causes operational deficiencies. The result is a growing administrative burden where both providers and billers spend a disproportionate amount of time fighting denials rather than optimizing the overall revenue cycle.

Software historically struggled to penetrate this space because RCM workflows involve massive amounts of unstructured, unreadable data trapped in faxes, voicemails, and fragmented systems. Payment flows are stitched together by hundreds of thousands of people employed by health systems and outsourced RCM service providers. Companies like R1 RCM employ approximately 30,000 people and operate at roughly 20% gross margins, a business model built on labor arbitrage rather than technology leverage.​

But recent regulatory changes and technical breakthroughs are finally removing these barriers. The 21st Century Cures Act promotes interoperability, pricing transparency rules mandate publication of reimbursement rates, and new CMS proposals would digitize prior authorization processes. Combined with advances in large language models, natural language processing, and optical character recognition, we now have the tools to automate tasks that were previously impossible.​

The Two Common Software Approaches We’ve Seen

Approach #1: Software Businesses Selling to Providers

This approach involves building full-service RCM software platforms that sell directly to providers, hospitals, health systems, and physician groups. These solutions typically address multiple stages of the revenue cycle, from patient eligibility verification through final payment posting. There are ~400,000 physician group practices and over 6,000 hospitals in the US., representing over 100 specialties and sub-specialties with varying medical coding needs and billing complexity. While healthcare is massive and this approach enables providers to focus on care versus billing, it is extremely fragmented, and larger hospital systems have many stakeholders and long sales cycles. While RCM solutions can gain traction in specific segments of the healthcare market, we believe this approach has relatively low barriers to entry, which is why we have seen this segment become increasingly crowded.

Approach #2: Software Selling to Payers

The second approach targets the opposite side of the equation: insurance payers. These companies develop AI-powered platforms that enable payers to process claims more efficiently, detect fraud, identify underpayments, and make more accurate adjudication decisions. This approach can reduce the administrative burden of retrospective audits that payers are currently facing. However, this is a customer base with extremely long sales cycles and not fragmented enough to capture leverage via automation.

A Third, More Compelling Approach

While both of the above strategies can have positive impacts on the healthcare market, we at Field are most excited about a third approach: AI software designed specifically for RCM service firms.

This aligns with our Service-as-a-Software investment thesis: software that acts as a digital worker.

Traditional RCM service providers employ tens of thousands of people performing repetitive, rules-based tasks. By building software that dramatically increases the productivity of these service firms, companies can capture enormous value while helping the entire healthcare ecosystem operate more efficiently.​

Why This Approach Wins

1. Large Total Addressable Market

Rather than selling to individual providers or payers, software that powers RCM service firms can reach the entire ecosystem. The top RCM services companies, like R1 RCM, serve over 750 healthcare organizations, Access Healthcare manages over $120 billion in claims annually with 27,000 revenue cycle professionals, and Optum360 employs 7,700 performance experts. Optum acquired Change for ~$13 billion, with its RCM capabilities being a key driver of the acquisition. Improving the efficiency of service firms of this scale represents a multibillion-dollar opportunity.

2. One-to-Many GTM with Clear ROI

Unlike individual hospitals or physician groups, which may resist workflow changes, RCM service providers have direct economic incentives to adopt tools that reduce their cost structure. They also have centralized decision-making, reducing lengthy sales cycles and minimizing change management friction associated with selling into fragmented healthcare delivery organizations. Partnering with an RCM service firm gives a business access to hundreds or thousands of providers at once rather than selling to each provider individually. That said, these RCM service firms can be harder to find, as they tend not to have a strong online presence.

3. Deeper Data Moats

Software that sits at the core of RCM service operations process millions of claims across hundreds of payers and thousands of providers. This aggregated, cross-cutting view of the entire healthcare billing ecosystem creates proprietary insights and defensible network effects.

The data advantage here is particularly powerful: pattern matching at scale is the key to solving medical billing automation. The majority of denials can be avoided by addressing problems earlier in the process, before the claim is even submitted. As the platform processes more data, it becomes increasingly aware of payer-specific idiosyncrasies, coding nuances, and denial patterns, creating a virtuous cycle of improved performance and decreasing need for human intervention.​ This creates a strong data advantage and a path towards a fully AI-native RCM service firm.

4. Revenue Model Alignment

RCM service firms charge based on outcomes, recovered revenue, increased collections, or reduced denials, rather than seat-based SaaS pricing. Billers typically charge providers a percentage of net collections, which means they’re directly incentivized to collect more money and prevent denials. Software that helps service firms win more business while reducing costs makes the purchase decision straightforward.

The convergence of market tailwinds, labor constraints, and regulatory changes creates a unique window of opportunity:

For entrepreneurs considering opportunities in healthcare RCM, several factors will determine success:

Integration Complexity is Real: Success often hinges on deep integration with EHR systems like Epic, Cerner, and Athenahealth. Companies with strong partnerships or native integrations have significant advantages.​

Human-in-the-Loop Beats Full Automation: Despite AI advances, healthcare providers remain skeptical of fully autonomous solutions. The most successful companies combine AI with human expertise to handle edge cases and build trust.​ Billers need systems that augment their intelligence, not replace them entirely.

Data is the Moat: The most defensible companies in this space are those that process high volumes of claims across multiple payers and providers, building proprietary datasets that continuously improve their models.​ This creates network effects that competitors can’t easily replicate.

Go Multi-Product Eventually: While point solutions can gain initial traction, the real opportunity lies in connecting multiple workflows, from documentation to coding to denial appeals, into comprehensive platforms. Companies that can expand from a wedge product to full-stack solutions will capture the most value.​

Conclusion: The Decacorn+ Opportunity

Healthcare accounts for nearly 20% of the US GDP. The scale of the industry creates potential for category-defining outcomes. But the game-changing opportunities will be captured by companies that think beyond point solutions to build platforms that deliver measurable outcomes: reduced denials, faster collections, and lower labor costs.

We believe the Service-as-a-Software model, focused on empowering RCM service firms, represents the most compelling opportunity because it combines a large market, an efficient go-to-market motion, outcome-aligned pricing, and data-driven defensibility.

At Field, we’re actively seeking founders building in this space. The healthcare RCM market is massive, the pain points are acute, and the technology is finally ready. The next leaders in RCM will be those who apply automation at the service layer, where AI-powered software performs the work directly.

If you are working on an RCM solution, would like to discuss the space, or have viewpoints on different approaches, please reach out!

At Field Ventures, we believe some of the most enduring companies will be built in antiquated, under-digitized industries through Service-as-a-Software solutions, a thesis we referenced in this blog post: products that don’t just replace a workflow with software, but that function like a digital teammate (rather than just a tool), capable of delivering outcomes with the reliability, scale, and trustworthiness of a service.

That’s why we’re thrilled to announce our investment in Superpanel, which just emerged out of stealth with $5.3M raised to transform how law firms evaluate and qualify legal cases. Field Ventures co-led the round alongside Outlander Ventures, who share our conviction that legal intake is overdue for reinvention.

Unblocking the Intake Bottleneck in Legal Services

For plaintiff law firms, case intake is everything: it determines which cases get pursued, how resources are allocated, and ultimately, the economics of the business. It is also the first touchpoint with potential customers, helping law firms grow their top of funnel while also verifying potential cases. Yet this process has historically been handled by large teams of intake specialists working across phone calls, text, web forms, and email, an expensive, error-prone, and inconsistent function that caps growth and drives up the cost per signed case.

Superpanel changes this dynamic. Its AI-powered digital teammate, custom-built for legal intake, manages intake end-to-end, just as a human staff member would, from initial contact to evidence collection. Superpanel never gets tired, never misses a follow-up, and always works the way your firm does. Even as consumers switch communication channels (across phone calls, text messages, and email), it gathers details, qualifies claims, and surfaces the most viable cases. Superpanel has already evaluated over 150,000 potential cases and manages thousands of inquiries every week for clients. Private beta customers report remarkable ROI: firms have been able to shift over 50% of time-consuming admin (manual follow-ups, document collection, and intake triage) to the digital teammate, doubled the number of signed cases per month without sacrificing quality, and are no longer limited by intake constraints. Beta customers are also seeing large increases in revenue with Superpanel, enabling them to see and qualify more cases.

Why We Invested

When we met Julien Emery (CEO) and Dingyu Zhang (CTO), it was clear they embodied the rare founder-market fit we look for. Julien helped to scale Hootsuite in its early days, while Dingyu has been building decision engines via AI for years. Together, they bring the enterprise AI expertise and consumer-grade product sensibility that this market demands.

Superpanel exemplifies our Service-as-a-Software thesis: a product that doesn’t just digitize a process, but replaces a labor function with software that performs at scale, with consistency, accuracy, and empathy. We believe this model will define the next wave of vertical software winners, and Superpanel is leading that charge in legal tech.

Looking Ahead

The legal services industry is massive, growing, and increasingly receptive to AI-driven transformation. With this raise, Superpanel will expand its team, accelerate go-to-market, and continue building the infrastructure that can make case evaluation more efficient for law firms and more accessible for consumers.

We’re proud to back Julien, Dingyu, and the entire Superpanel team as they build what we believe can become a category-defining company at the intersection of legal tech and AI.

Follow the discussion on LinkedIn

At Field Ventures, we've built our investment thesis around a fundamental belief: the most transformative businesses emerge when domain experts apply innovative technology to modernize antiquated industries. Today, we are in the very early innings of a significant paradigm shift in enterprise software, the transition from Software-as-a-Service to Service-as-a-Software. This isn’t just a change in how software is delivered; it’s a change in what software is expected to do.

Software is no longer a tool. It’s becoming a teammate.

The Fundamental Shift: From Tools to Workers

The traditional Software-as-a-Service model has served us well for over two decades, enabling organizations to digitize workflows, boost productivity, and create new revenue streams. However, SaaS has inherent limitations that become particularly apparent when attempting to penetrate traditional markets. Traditional software acts as a tool that assists humans with workflows, requiring significant manual input and configuration to achieve desired outcomes. Software-as-a-Service (SaaS) traditionally provides customers with a tool or platform, but the customer is responsible for utilizing it to achieve desired outcomes. For example, Salesforce helps manage leads, but your sales reps still do the selling. In a services business, by contrast, the provider delivers an outcome for the customer (e.g., an accounting firm preparing your taxes for you).

Whereas, Service-as-a-Software represents a fundamental reimagining of this relationship. Service-as-Software flips SaaS on its head by having the software itself deliver the outcome, blurring the line between software product and service provider. Instead of providing software that helps users complete tasks, Service-as-a-Software delivers autonomous AI agents that actually perform the work itself. In other words, the software becomes the worker. Instead of just a tool, you’re buying an “employee” that can execute tasks autonomously. This isn't merely an incremental improvement. It's a complete transformation from software that organizes human work to software that becomes the worker.

The Evolution from Systems of Record to Systems of Work

Over the past several decades, enterprise software has undergone a profound transformation, mirroring the broader digitization and automation of business processes. At Field, we view this evolution as a multi-stage journey, with each phase unlocking new value for organizations and creating fresh opportunities for investors.

Phase 1: Systems of Record emerged as foundational solutions for organizing critical business data. Think of early construction project management tools or legacy healthcare record systems. These platforms served as digital filing cabinets—essential for documentation but requiring manual data entry and offering limited actionable insights. While valuable for compliance, they often created siloed information that couldn't dynamically inform operations.

Phase 2: Systems of Engagement transformed static data into interactive workflows. These solutions, such as real-time restaurant inventory trackers or field service dispatch platforms, enabled collaboration and mobile access. They bridged the gap between data and action, allowing teams to coordinate more effectively. However, they still relied on human judgment for decision-making and execution.

Phase 3: Systems of Work represent our current frontier, where AI agents autonomously execute entire workflows. This is a paradigm shift driven by advances in artificial intelligence and automation. Unlike previous generations of software, which required users to initiate and oversee every step of a process, systems of work empower AI agents to take on entire workflows autonomously. In vertical markets such as construction or logistics, this means platforms that not only track equipment maintenance but also automatically schedule repairs, order parts, and optimize routes based on real-time conditions. These platforms don’t just assist users or store data; they actively perform tasks, make decisions, and deliver outcomes, often with little or no human intervention. Moreover, these systems seamlessly digitize previously offline or siloed data, generating rich, actionable datasets that empower organizations with deeper industry insights.

For an illustrative example of these three phases, we can look at the evolution of customer support.

Shifting from SaaS tools to AI-run services isn’t just a technology upgrade; it’s a game-changer for productivity and business models. Mundane and repetitive tasks get handled 24/7 by tireless AI agents, freeing human team members to focus on high-impact, creative, or complex problems. Organizations can achieve new levels of efficiency, consistency, and scale: AI agents don’t take breaks, don’t make arithmetic errors, and can be replicated instantly. And if you need more capacity, you can spin up more agents via software, without the recruiting, salaries, and management overhead of hiring an equivalent army of people. Just as important, Service-as-Software can deliver more consistent quality and oversight. And whereas humans might vary in their adherence to processes, an AI agent can be updated instantly across the organization when improvements are found. In essence, software services offer the dream of a perfectly trained, always-available team member who never forgets a policy or skips a step.

Why Vertical Industries Are Poised to Benefit

This shift from software-as-tool to software-as-worker is exciting everywhere, but we believe that it will be especially transformative in traditional / antiquated industries, the kinds of verticals that have been “under-digitized” and overlooked by many modern software providers. Fields like construction, logistics, healthcare, and financial services are rife with manual processes, paper-based workflows, and labor-intensive operations. These sectors often operate on slim IT budgets but have huge labor and operations budgets, making them prime candidates for AI-driven services that can be easily integrated to perform tasks.

Vertical markets offer several distinct advantages for this new paradigm:

1. Higher Willingness to Pay for Outcomes

Vertical industries often have specific, measurable pain points that directly impact their bottom line. Unlike horizontal solutions that compete on features, Service-as-a-Software in vertical markets can charge based on tangible outcomes: qualified leads generated, compliance reports completed, or medical records processed, for example. This outcome-based pricing model aligns perfectly with how industry operators think about value creation.

2. Data Moats and Industry Expertise

Successful Service-as-a-Software businesses in vertical markets can build substantial data moats by aggregating industry-specific information. As these platforms process more industry data and tailor their services to the nuanced workflows of their target market, they become increasingly effective, creating a virtuous cycle that’s difficult for competitors to replicate. Importantly, these industries often require highly specific integrations and configurations during implementation, which means that once a solution is in place, switching costs are significantly higher and the product becomes much stickier. This stickiness not only strengthens customer retention but also amplifies the platform’s ability to capture and leverage proprietary data, further reinforcing its competitive advantage in the market.

3. Lower Technology Adoption Barriers

Many traditional industries have been resistant to software adoption due to the complexity, high implementation costs, and challenges associated with change management. Service-as-a-Software eliminates many of these barriers by delivering immediate value without requiring extensive process changes or employee training. Instead of asking users to learn new software, these solutions simply deliver better outcomes through familiar interfaces, which is critical for antiquated markets, which typically have high turnover amongst their employees.

Below are a few real-world examples of pain points in antiquated industries and how Service-as-a-Software solutions could shine:

• Construction: Many construction firms still coordinate projects via spreadsheets, whiteboards, and phone calls. Tasks such as scheduling subcontractors, reviewing building plans for code compliance, or processing permits require a significant amount of human hours. An AI “project manager” service could automatically adjust schedules when a delay occurs, or an AI permitting assistant could review plans for code issues and generate permit applications in minutes (startups have already emerged to do exactly this). By acting as a tireless project coordinator, software could eliminate costly delays and rework in construction projects.

• Logistics: In freight and supply chain operations, booking shipments and managing exceptions is often a 24/7 human job (think of freight brokers and dispatchers). A Service-as-a-Software freight broker could automatically match loads to carriers, handle communications, and optimize routes using real-time data, effectively behaving like a digital logistics coordinator. This is not far-fetched, digital freight platforms already automate pricing and matching, and adding generative AI could make them even more autonomous.

• Healthcare: Healthcare administration is notorious for its bureaucracy, processing insurance claims, scheduling patients, and transcribing medical notes. We now see AI services tackling these tasks: for instance, an AI medical coding assistant can read doctors' notes and assign billing codes, or an AI agent for prior authorizations can call insurance companies to obtain approvals, saving countless hours for clinic staff. Companies have already begun deploying AI agents in healthcare revenue cycle management (RCM) that handle repetitive billing tasks, yielding significant cost savings. In a heavily regulated field like healthcare, these AI services can operate under human oversight to ensure compliance while massively improving throughput.

• Financial Services: In lending or insurance, processing an application involves gathering documents, verifying data, assessing risk, and other tasks, traditionally performed by back-office analysts. AI underwriting assistants can automate large portions of this, evaluating documents and even interacting with customers for missing information via natural language. Meanwhile, in wealth management or accounting, we see the beginnings of AI co-pilots that prepare financial plans or tax filings, which a human then reviews. Instead of selling QuickBooks software to a small business and letting them do the bookkeeping, an AI-driven service might be the bookkeeper and tax preparer; you “hire” the software to do the work, not just provide a tool.

Why are these verticals so attractive for Service-as-a-Software? Because many are operating on outdated software (or no software at all) and relying on heavy manual labor, a combination that is ripe for leapfrogging directly to AI automation. Traditional software often failed to penetrate these industries deeply, in part because simply offering a tool wasn’t enough; the internal capacity or willingness to adopt new tech was limited. But offering an AI-powered service that delivers outcomes (e.g., “we will handle your compliance paperwork for you via our software”) is far more compelling. It reduces the need for in-house tech expertise and addresses business owners' key pain points.

Moreover, vertical markets have unique, domain-specific workflows that AI can learn, providing defensibility for startups that develop expertise in a specific industry. At Field, we look to partner with domain experts for this reason: a founder who deeply understands, say, trucking logistics or hospital administration is best positioned to build an AI solution that truly automates those tasks. We specifically look for underserved markets where modern AI can unlock massive value from outdated infrastructure. For example, Wrapbook targets the entertainment production industry. Production payroll was largely manual, think paper time cards and faxed documents, which led to inefficiency and errors. Wrapbook introduced a vertical SaaS solution that not only digitizes the workflow but also acts as the Employer-of-Record, essentially taking on the service of payroll processing for production companies via software. This kind of vertically integrated approach, combining software and service, can be incredibly sticky and valuable in antiquated sectors. We believe that investing in companies that are building systems of work, not just systems of record or engagement, will be the next wave of enterprise value creation.

In summary, the evolution from systems of record to systems of work reflects a broader trend toward automation, intelligence, and outcome-driven software. For us, at Field, that means backing founders who understand how to harness these technologies to transform traditional industries and deliver measurable results for their customers.

The Economic Imperative: Accessing Labor Budgets vs. IT Budgets

One of the most compelling aspects of the Service-as-a-Software model and a key reason why we are particularly excited about this opportunity relates to budget allocation within antiquated industries. When selling traditional B2B software into antiquated markets, you're competing for limited IT or departmental budget dollars, often requiring extensive justification for technology investments that may not show immediate returns or a quick time to value.

In contrast, a Service-as-a-Software company is effectively selling into labor budgets, the money a company would otherwise spend on salaries or outsourcing. These budgets are vastly larger than pure software budgets. As a result, startups in this space can unlock much higher contract values and growth potential by replacing or augmenting labor rather than just offering a tool.

Beyond just market size, selling an AI service often means delivering clear ROI in the language executives care about: outcomes and dollars. Instead of asking a client for a $100K software subscription from their IT budget, you might ask for a $1M contract out of the operations budget because you’ll save them $5M in headcount costs, oftentimes an easier sell. In many cases, Service-as-a-Software companies price their product in line with the value of the work they’re doing (e.g., per transaction processed, per sale closed, per report produced), which directly ties to business outcomes. This outcome-based pricing approach creates strong alignment between the business and their customers’ success. It also tends to scale naturally: if the AI service delivers more work or better results, the customer earns more, and the customer is happy to pay, as it’s tied to the value delivered.

Of course, this doesn’t mean SaaS is going away or that every software product should turn into a service. In many domains, a hybrid approach will remain. Some customers will still want to perform certain functions in-house with their own personnel and tools, especially when they believe they have a competitive advantage or require strict control. We’re likely to see multiple models coexist: human-in-the-loop systems, fully automated services, and traditional software can all thrive side by side. But the takeaway is that for startup founders, Service-as-a-Software opens a new avenue to build massive businesses quickly by capturing value that was previously paid to humans or outsourcing firms. It’s an evolution of SaaS that expands the pie rather than cannibalizing it.

The Pattern: From Tools to Autonomous Services

Across the companies we have evaluated and those within our portfolio, we're seeing consistent patterns emerge:

Key Investment Themes in Service-as-a-Software

Based on our research, several investment themes are emerging in the Service-as-a-Software landscape:

1. Agentic AI Systems

The most promising Service-as-a-Software companies are building systems of AI agents that work collaboratively to handle complex, multi-step workflows. These systems mimic human teams, with specialized agents assigned to specific tasks and continually learning from one another. Unlike single-purpose automation tools, these multi-agent systems can adapt to changing requirements and handle exceptions autonomously.

2. Data-at-Source Positioning

The most defensible Service-as-a-Software companies position themselves at the source of data creation within their target industries. By capturing and acting on information in its raw, unfiltered state, these platforms can initiate actions across downstream systems and build comprehensive workflow orchestration capabilities. This positioning creates significant switching costs and network effects.

3. 24/7 Operational Capabilities

Service-as-a-Software enables entirely new categories of work that were previously impossible due to human limitations. AI agents can operate continuously, providing reliable support and round-the-clock operations in industries like healthcare, security, and customer service. This creates opportunities to deliver services that humans never provided due to time constraints or cost considerations.

Business Model Innovation: From Seats to Outcomes

Traditional SaaS pricing models based on user seats or feature tiers are often misaligned with how Service-as-a-Software creates value. Instead, we're seeing the emergence of outcome-based pricing models that tie costs directly to results achieved.

This shift in business model creates several advantages, including the following:

What’s Ahead: Challenges to Navigate

Before we declare the end of human work and fully hand over the reins to our AI coworkers, it’s important to level-set on challenges. Startup founders in the Service-as-a-Software space must navigate several key issues:

1. Technology Readiness & Scope Limits

Despite the hype, today’s AI still has limitations. Generative AI can sometimes produce errors or require a large amount of data to perform effectively. Not every task is yet ripe for full automation. Founders must carefully choose use cases where AI currently excels (or is rapidly improving) and be cautious about over-promising. It’s also important to define the scope of the AI service narrowly at first. Many successful startups are focusing on automating a specific workflow deeply, rather than claiming they’ll be a fully autonomous generalist from the outset. This reduces complexity and allows for better accuracy. There are still workflows and decision points that require human oversight, particularly in high-risk industries such as healthcare and financial services, where sensitive data and critical outcomes demand a cautious approach. Robust human-in-the-loop mechanisms may be necessary to provide validation, manage exceptions, and ensure compliance in these environments. This hybrid approach not only mitigates risk but also builds trust with users and regulators, making it a vital component of any solution operating in high-stakes sectors. This is analogous to managing a human team, except the “training and coaching” involves machine learning and prompt engineering. Field’s view is that domain-expert founders have an edge here: their intimate understanding of the work lets them spot where AI is falling short and quickly iterate to improve it. Nonetheless, technology risks (from model inaccuracies to integration hiccups) will need continual management.

2. Trust and Change Management

Convincing customers (and end-users) to trust an autonomous software service is a non-trivial challenge. Companies may be hesitant to hand over mission-critical tasks to AI without proof of reliability. Early AI solutions will likely need to start with a hybrid approach – for instance, AI generates the output and a human reviews it, or AI handles simpler cases and escalates edge cases to humans. This builds trust over time as the AI demonstrates its capabilities. Moving too fast and losing trust is a quick way to set a business back in a material way or kill the business entirely. Transparency is key: Service-as-Software providers should allow clients to see what the AI is doing, provide audit logs, and make it easy to intervene when needed. Culturally, organizations will also face internal resistance; workers might fear being displaced or feel uneasy working with AI agents. Founders should frame AI services as tools that elevate human workers to more strategic roles (rather than replace them entirely) – at least in the interim. Building customer confidence through pilot projects, reference cases, and small wins is crucial to drive adoption.

3. Competition and Differentiation

As Service-as-a-Software gains momentum, we’ll see incumbents and new entrants vying to automate the same services. There’s a question of who builds the winning AI workers: the SaaS incumbents, the BPO/service firms, or startups? In some cases, incumbents will add more AI automation to their products (as, say, Salesforce is doing with Einstein GPT). In others, entirely new companies will offer a better end-to-end service. Startups should be prepared for competition from both established companies and new entrants. The best defense will be a head start in creating unique datasets and learning. Each task your AI performs teaches your system, creating a data moat. Strong customer relationships and a focus on outcomes (not just technology) also create stickiness. We may also see partnerships (for example, a SaaS firm might partner with an AI service startup to offer their customers a managed service option). Founders need to stay agile and keep an eye on both traditional competitors and other AI players in their niche.

Despite these challenges, the trajectory seems clear: the benefits of AI-driven services are too compelling for businesses to ignore. We are still in early innings, and many companies are just experimenting around the edges with AI copilots and assistants. The transformation of core workflows will take time, perhaps years, and involve plenty of trial and error. But the momentum is there. Those who start adapting to Service-as-a-Software now (with the appropriate safeguards and phased approach) will be ahead of the curve when the technology and market readiness hit full stride.

Investment Implications for Field Ventures

As early-stage investors focused on vertical markets, the Service-as-a-Software revolution aligns perfectly with our investment thesis of backing domain experts who understand industry-specific challenges and can apply innovative technology solutions.

When evaluating founders building within a Service as a Software approach, we seek founders who have the following traits:

Conclusion: A New Era of Work

The rise of Service-as-a-Software signals a new era where the lines between software, services, and labor are blurring. Software is no longer confined to being a tool in the hands of employees; it can be the employee. This shift will undoubtedly reshape business strategies, IT budgets, and career paths in the years and decades to come. It holds the promise of unparalleled efficiency and scalability, as well as the challenge of ensuring that humans and AI work together in harmony and trust. As with any technological revolution, there will be success stories and cautionary tales, as well as wild optimism and justified skepticism.

For startup founders and operators, the takeaway is this: the opportunities at this intersection of AI and traditional services are enormous. By delivering outcomes (not just software) and by targeting the huge pools of spend on legacy labor, you can build category-defining companies while genuinely helping industries evolve. But it requires empathy for the customer’s workflow, respect for the complexity of real-world operations, and a willingness to tackle hard problems with both AI and elbow grease. We believe that the most successful companies will be those that start with a deep understanding of their industry, embrace AI-native approaches, and relentlessly focus on delivering measurable outcomes for their customers. By focusing on outcome-based solutions that tap into labor budgets rather than constrained IT budgets, these businesses can achieve rapid growth, massive revenue, and sustainable competitive advantages.

For entrepreneurs building in this space, the opportunity has never been greater. For investors willing to back domain experts with transformative visions, the potential returns could dwarf even the most successful software investments of the past two decades. If you are building along this theme, please reach out!

Field Ventures Perspectives

Perspectives from Field Ventures

In our previous post, we discussed our thoughts on vertical fintech, and in this post, we will dive deeper into one specific model within vertical fintech: verticalized payroll. In that previous post, we discussed the challenges of leading with fintech rather than workflow because of the commoditized nature of many fintech products. Payroll, often seen as a commodity offering in a horizontal sense, becomes a unique and indispensable service when tailored to the specific needs of individual verticals. This transformation from a commoditized offering to a unique solution set occurs because payroll, deeply embedded within the HR workflow stack when targeted to a single market, must account for the nuanced demands of any given market.

A vertically integrated payroll offering offers industry-specific features and is superior compared to horizontal players while also providing a strategic entry point for additional monetization offerings. With that said most markets are adequately served by horizontal payroll providers like ADP, Paychex, Paylocity, etc. So, what traits does a market need to make it a good fit for a verticalized payroll opportunity? Let’s look at the entertainment industry as a case study.

Case Study: Entertainment Industry

The entertainment industry relies heavily on a transient workforce, including actors, makeup artists, and other on-set professionals. In addition, this is an industry that is project-based, which means that large swaths of workers are hired, onboarded, insured, and terminated routinely on a daily, weekly, and monthly basis. Additionally, each project often necessitates the creation of new corporate entities, each of which requires its own insurance policies, further increasing the complexity of providing payroll for the industry. Finally, this workforce is largely unionized, which creates very esoteric wage rules.

Historically, most of this work has been handled manually via paper and pen. Workers manually fill out paper timecards that are reviewed and scanned by production managers, and finally sent to a payroll provider for processing. A portfolio company of mine, Wrapbook, addresses the challenges of manual and laborious payroll processing by digitizing the entire payroll experience for the entertainment industry while also serving as the Employer of Record (EoR) for the industry. An Employer of Record (EoR) is an organization that serves as the employer for tax purposes while the employee performs work at a different company. Being an EoR can be a critical component of building a vertical payroll provider.

The EoR takes on the responsibility of traditional employment tasks and liabilities by offering a slew of services including, but not limited to:

Beyond Entertainment: Real Estate and Insurance

Another take on vertical payroll could be around businesses that manage the lion’s share of a worker’s earnings for a given market. These could be industries where employees make most of their income on commissions. Two sectors, in particular, come to mind that could be potential fits for this approach: real estate and insurance.

In both real estate and insurance brokerages, the primary employee within the industry is an agent/broker. In both markets, the agent/broker receives the bulk of their compensation via commissions rather than traditional W2 or 1099 income streams. Within real estate, brokerages hire real estate agents as their employees. Each real estate office has a designated broker, and all commissions are paid directly to the brokerage, who splits the commission with agents that were involved in the transaction.

Like the real estate industry, insurance agents also earn the majority of their income via commissions. However, unlike real estate agents, insurance agents can also earn residual commissions over the policy’s life. They receive a commission at the time a policy is binded and continue to receive (lower) commissions annually as long as their client keeps the policy in place or renews with them. Commission overrides can make tracking the payout even more complex, as producers may be paid on behalf of other agents. Residual commissions and overrides create additional complexity when thinking about payroll-adjacent businesses, and we believe that increased complexity makes a market more attractive for a vertical payroll opportunity.

There isn’t a single definitive checklist of traits that make a market ripe for a vertical payroll business because of the nuances of each given market. However, we do think that there are a few traits that could make a market ripe for this kind of opportunity. This is not a comprehensive list, nor are all of these traits required for a vertical payroll opportunity. Instead, we think any one of these traits being present to be attractive enough to explore vertical payroll.

Vertical Payroll Market Traits

Conclusion

Vertical payroll solutions do more than just handle payroll; they embed themselves deeply into the operational workflows of their target industries. This deep integration allows for the creation of additional value-added services, such as health benefits, earned wage access, and banking services, which can further enhance customer loyalty and generate new revenue streams.

We believe there are several markets (many more than we have discussed in this post) where a vertical payroll and/or vertical commission-like model make sense, and we are excited to continue spending time with founders, operators, and investors who are working along these lines. Whether you are working on building a vertical payroll provider or building a “payroll” like business focused on commission-like revenue, we would love to hear from you.

Before generative AI became all the rage, vertical software, and embedded/vertical fintech were all the rage (perhaps it still is for many others as it is for me). At Field, we love investing in vertical software businesses. When properly executed, vertical software leverages the nuances of a given market to improve efficiencies around critical workflows and to aggregate and surface key business insights. The best-in-class examples create a system of record for businesses within the category, while creating substantial enterprise value.

We also love investing in fintech focused businesses; with that said, there isn’t anything inherently defensible about financial services for the sake of financial services. Most pure-play financial services offerings are commoditized offerings. It is extremely difficult to create stickiness with your end customer when the customer can always find a better rate, lower payment costs, or better rewards programs somewhere else.

However, marrying vertical software and financial services can be extremely powerful because of the significant increase in market size that occurs from combining these two product offerings and business models. Workflow software can be quite sticky, but there is always only so much that a business will be willing to pay for software. Conversely, financial services (payments, take rates, insurance, lending, etc.) can be lucrative ways to monetize but tends to have challenges regarding retention.

When executed correctly, the marrying of workflow optimization and financial services create the following attractive business dynamics.

Before generative AI became all the rage, vertical software, and embedded/vertical fintech were all the rage (perhaps it still is for many others as it is for me). At Field, we love investing in vertical software businesses. When properly executed, vertical software leverages the nuances of a given market to improve efficiencies around critical workflows and to aggregate and surface key business insights. The best-in-class examples create a system of record for businesses within the category, while creating substantial enterprise value.

We also love investing in fintech focused businesses; with that said, there isn’t anything inherently defensible about financial services for the sake of financial services. Most pure-play financial services offerings are commoditized offerings. It is extremely difficult to create stickiness with your end customer when the customer can always find a better rate, lower payment costs, or better rewards programs somewhere else.

However, marrying vertical software and financial services can be extremely powerful because of the significant increase in market size that occurs from combining these two product offerings and business models. Workflow software can be quite sticky, but there is always only so much that a business will be willing to pay for software. Conversely, financial services (payments, take rates, insurance, lending, etc.) can be lucrative ways to monetize but tends to have challenges regarding retention.

When executed correctly, the marrying of workflow optimization and financial services create the following attractive business dynamics

Market Traits for Vertical Fintech Success

An example of a market that we think fits this model well is the construction market. The construction market is a massive industry with many stakeholders such as general contractors, subcontractors, material suppliers, developers, etc., all of which have tremendous amounts of spend (hundreds of billions) that flow through them. In addition, general contractors, subcontractors, and material suppliers are all fragmented categories. However, GCs and subs might actually be too fragmented of a segment to focus on. The average GC or sub are very small businesses, which means that one can’t afford to spend any meaningful amount of money to acquire GCs or subs given the low level of monetization that you can capture from them.

Like GCs and subs, material suppliers are also a fragmented group, but not as fragmented, and more importantly, the average material suppliers are much larger businesses and transact with hundreds if not thousands of GCs and subs. This is why I chose the material supplier segment as a sub-category within construction via my investment in Suppli, a payments platform for material suppliers.

Another vertical that we found attractive from a verticalized fintech play was the entertainment industry. There is well over $100B spent annually by production companies across payroll, insurance, and B2B payments within the entertainment industry across film, tv, commercial, and live events. Like the material suppliers within construction, there is fragmentation amongst production companies, many of which are sizable businesses with lots of spend to capture. Lastly, the entertainment industry is unique in that the workers in the space are all contractors and are primarily unionized. This makes it ripe for a vertical fintech business focused on payroll, which is why I led the seed round in Wrapbook, a vertical payroll provider for the entertainment industry. We will share more of our thoughts on vertical payroll opportunities in a future post.

We are certainly not unique in our excitement for vertical SaaS + fintech, but hopefully, this post provides context into how we think the playbook can work across markets. If you are working on a vertical fintech opportunity or want to jam on the space, feel free to reach out!