Reasons SaaS is Evolving With AI

We’re not just in a tech shift. We’re in a work shift. I’ve lived through major technology waves: On-prem. SaaS. Social. Mobile. Each one made us more connected, more efficient, more scalable. But AI? It is different. It’s giving rise to a whole new concept: “work” or “results” as a service. So, what does that mean exactly? Simply put, if software-as-a-service helps people do work, AI will do work to help people. Take a salesperson, for example. Their "job to be done" is to close deals. Every day, they carry out tasks to help them achieve that goal. They spend hours updating records, scoring leads, drafting emails, scheduling calls, researching companies, and following up. Software made it easier to do all this work. But the salesperson was always the one who actually did it. (I was in sales, I know what it’s like 😉) Now, AI is bringing intelligence to software. It can think. It can reason. It can do work – like updating contact records and crafting follow-up emails. The goal hasn’t changed. But the path to get there? Completely transformed. So, how does this change our mental model for thinking about software? When we buy software, we’ll focus less on features and functionality, and more on the specific work and results it can deliver to help our businesses grow. In other words, we will think of software as “work” or “results” as a service. Today, we use passive tools. Soon, we'll use active agents. Today, we buy licenses for our teams. Soon, we’ll buy outcomes for our business. Today, we use software to help us do work. Soon, we'll use software that does work for us. I can’t think of a more exciting shift than that!

I've heard a few times now from enterprise tech leaders that the new agentic AI capabilities coming means they are doing a "strategic reset" on their overall AI strategy. It's not a bad idea, except that if they repeat the mistakes that have led many to minimal results from generative AI these last two years, it will be another failed strategy. The big lessons from Generative AI 1.0 to take into Generative AI 2.0: (1) DREAM BIGGER People are still thinking incremental improvements instead of big transformational ideas that were not possible before agentic AI. Part of the problem is that too few people have seen agentic AI in action. From Davos where I asked 20+ CEOs if they had ever seen agentic AI hands-on (only one yes) to multiple dinners a week since the beginning of the year where only 2-3 hands going up when I ask the same question. If you still haven't seen what's possible, comment below so I can send you some 🤯 REAL demos. (2) IT NEEDS TO UNDERSTAND BUSINESS PROCESSES END TO END Part of people's failure these last 2 years to do anything transformational lies in lack of high-level process understanding. And it's not your fault. The era of SaaS meant that process became nearly synonomous with the software system / system of record for that process. We all got basically generic software with generic UIs on top of generic databases and tens of millions of dollars and years later got that SaaS to be slightly more customized. And that system became our process. Along the way people forgot how to do business process mapping. Yes, even self-orchestrating AI needs some blueprinting! And the biggest time sync I find in our rollouts is how long it takes for execs to agree on re-built AI-first processes — because processes for important things are almost never written down, because no one person owns it, because change requires consensus, because SaaS decimated folks' skills here. Even with awesome self-orchestrating capabilities, we need blueprints for compliance, for auditability, for improving results. You get transformational results from AI when you understand the processes driving the big core workflows that underpin your business. AI can run those workflows only once you understand them. (3) TECH AND BUSINESS LEADERS NEED TO BUILD TOGETHER Many generative AI 1.0 projects failed because it was tech for tech's sake. IT brought projects to the business that the business didn't have a hand in building. And the results were not good. The best ideas will come from people doing the job. Another way to think about it — generative AI 1.0 was driven by the cost centers of the business. You become an AI-first company when AI is driven by the revenue center. It will still be run/enabled by IT, but too many business leaders are hands off, not deeply understanding what is possible. In 5 years there will be AI-first companies, and there will be dead companies. If the leader of your business doesn't get that yet, send them this post.

The SaaS Era of 2013-2022 is Over. Welcome to The World of Hyperfunctional SaaS. So is AI taking over everything? Will $1B SaaS companies be run by 1 person teams? I mean … maybe. Across the SaaStr Fund portfolio, I’ve seen multiple portfolio companies lose $1m customers to new, in-house AI initiatives. But I’ve also seen others grow even faster as AI rebooted and enhanced their products. What I do know is many of the billion+ SaaS companies I invested in a decade ago would not be remotely competitive today. And that’s how it should be. Software, innovation and time marches on. And if you can’t keep up, you’ll get left behind. Many today have been unable to keep up, and have fallen behind, and seen growth slow to 20%, 10%, even 0%. What I really think is happening is customer expectations have gone way up. AI is part of it, platforms are part of it. And promises are part of it. We’re promising to do a lot more. Today, customers are expecting: 1⃣ AI to replace 30%-50% of their workforce.  Is this possible? Maybe, maybe not. But it’s already starting in the contact center, where leaders from @Zendesk to @gorgiasio to @intercom to @Talkdesk are automating away 30%-50% of contact center headcount. 2⃣ All unstructured data to be instantly structured and searchable.  Search was sleepy for years. Now, it’s front and center. Products that were complex to extract data before, especially documents, can now be organized in real-time by AI. This is one reason @levie is so excited about AI. It unlocks all your documents. Customers aren’t satisfied anymore with not being able to get answers from every and any dataset. 3⃣ Every product to work elegantly in a plain English prompt.  We’re all getting used to working the OpenAI way. Now customers expect your product to work just as elegantly, or more so. Not crappy interactions that just return crummy results. 4⃣ Even complex onboarding to be fully automated.  Complex enterprise business process change still takes time. But anything below that, customers are expecting to Go … Now. Not in a week or a month. AI has changed expectations here. 5⃣ Core platforms to do the work of 20 add-ons and point solutions. To do a lot more. HubSpot, Rippling, monday.com and others were early here. It’s not just that customers want to buy more products from the same vendor. It’s that they expect their solutions to do more. Why can’t my payroll app handle onboarding and 401k and pension and contractors? Why can’t my CRM also do marketing automation and support, out of the box? Why can’t my CMS also automate clips, video, and social? And so on. Yes, there is point solution fatigue. But it’s more than that. Customers are now demanding every vendor do more of what they need. ⏫ It’s the era of Hyperfunctional SaaS. It’s a lot more work, and harder to build than before. For often the same deal and deal size. Step up — or step aside.

AI Agents have the potential to democratize knowledge work in the same way that SaaS democratized software. And as we've seen in the past couple of decades with software, every time you make a service cheaper and more available, you dramatically increase the size of the total addressable market. Let's take, for instance, what happened in the early days of SaaS. The biggest mistake that most people and investors made was looking at the market sizes of traditional on-prem software to see how big the market could be for this new crop of companies. In fact, some even felt the markets would actually be *smaller* because the software may be cheaper to run for an enterprise. All these theories were wrong, by an order of magnitude. What we actually saw happen was not that SaaS initially replaced or went after traditional incumbent software products for existing customers, but instead, the biggest early customers were actually smaller businesses or teams in large enterprises that previously didn't have access to traditional on-prem enterprise software. Starting with Salesforce and NetSuite, for the first time small businesses had access to effectively the same tech stack that a large enterprise had. AWS ushered in an era where a one person startup could build an app and scale it without ever visiting a datacenter. Box let businesses of all sizes manage documents and content securely. Stripe gave any developer a full payment stack. All of these new services --and thousands more-- led to a 10Xing (or more) the size of traditional markets by serving customers that previously didn't have access to these types of tools. Now, if you extrapolate out what we're seeing in the earliest days of AI, the same dynamic could hold true for AI Agents. While large enterprises have traditionally had access to nearly every specialized form talent or an abundance of labor, the vast majority of businesses don't have this same luxury. For most small startups just getting going, they often don't have the resources to do outbound sales, full customer support, specialized legal work, and so on. And as a startup scales, you're constantly making resource trade-offs that are less driven by what's best for the business, but instead driven by how much capital you have. In the future, by making the barrier to entry to getting knowledge work done as simple as a website signup or API call, we will likely see a massive increase in usage of “services” that previously were near-impossible to access easily. And what's amazing is the vast majority of the usage of these AI Agents will likely come from previous areas of "non-consumption". That is to say, these will be customers that would not have spent anything on similar labor categories in a pre-AI world. We're in only the very beginning of this new era of AI-driven work, but the scale of the opportunity and the market will be massive.

AI is going to provide yet another turboboost to SMB tech. A long time ago, in the days before SaaS, the idea of starting a software company that sells to SMBs was almost unfundable by VCs. There were notable outliers like Intuit. But in general, the consensus was that the cost of acquiring and servicing SMB clients didn’t foot out and the complexity wouldn’t scale. As such, many of the vendors that sold to SMBs were relatively small and regional. SaaS unlocked a goldmine, with massive success stories like Shopify, Hubspot, Toast, Klaviyo and ServiceTitan, along with 100s more. Vendors could cost effectively acquire customers through inside sales and Product-Led Growth (PLG) channels. And they could service them through digital #CustomerSuccess methodologies. Meanwhile, clients no longer had to host and manage the software itself - an arduous task for a small business. As much of a boost as SaaS gave to SMB tech, AI is going to take it many steps further. The “last mile” of delivering value in SaaS is still dependent on the end customer’s ability to leverage the software. Customer Success Managers and digital adoption can help, but ultimately, success often hinges on the skillset of the SMB. This is problematic, because most small businesses are lean and mission-oriented. Running software is often low on their priority list. But what if they don’t have to run the software? With AI agents, vendors can move, as VC Sarah Tavel famously coined, “from selling software to selling work”: * Marketing platforms can give SMBs what they want - leads - and handle the rest automatically (e.g., campaign design, optimization, etc.) * Recruiting tools can move from Applicant Tracking Systems to sourcing and screening hires for small businesses. * Service products can go from tracking customer inquiries to handling inbound calls and emails automatically. In the process, these firms can expand their Average Selling Prices (ASPs) and therefore their Total Addressable Market (TAM) radically, since they are delivering far more value to the end customer. If you’re in SMB tech, the good times are just getting started.

AI is finally making services businesses scalable—and—exciting to VCs. The global services market is in the trillions of💰s, far larger than today’s software market. Yet, services businesses haven’t been the darlings of venture capital, as they were perceived to lack rapid scaling potential. 𝗔𝗜 𝗶𝘀 𝗰𝗵𝗮𝗻𝗴𝗶𝗻𝗴 𝘁𝗵𝗮𝘁. By blending AI seamlessly with human expertise, there is an opportunity to get into much larger markets with models that have the potential to scale in ways services - or even SaaS businesses - can't. For example, instead of offering a marketing SaaS, an AI-powered Service-as-Software business can deliver what the customer really wants: high-quality leads or compelling content. We’ve seen this potential firsthand through Emergent Ventures’ investments in multiple AI-powered companies that leverage humans-in-the-loop. These models resonate with B2B customers because they offer faster, clearer paths to value—reliable outcomes delivered with greater efficiency. For many customers, it’s a significant upgrade over traditional agency or service-provider relationships. While the potential is huge, only a fraction of AI-powered services startups will scale. 𝗦𝘂𝗰𝗰𝗲𝘀𝘀 𝗱𝗲𝗽𝗲𝗻𝗱𝘀 𝗼𝗻 𝘁𝗵𝗼𝘂𝗴𝗵𝘁𝗳𝘂𝗹 𝗲𝗮𝗿𝗹𝘆 𝗰𝗵𝗼𝗶𝗰𝗲𝘀 𝗮𝗻𝗱 𝗲𝘅𝗰𝗲𝗽𝘁𝗶𝗼𝗻𝗮𝗹 𝗲𝘅𝗲𝗰𝘂𝘁𝗶𝗼𝗻. Here’s what we have learned works well: 𝟭. 𝗔𝗜-𝗛𝘂𝗺𝗮𝗻 𝗦𝘆𝗻𝗲𝗿𝗴𝘆: AI and software should do the heavy lifting, with humans involved strategically— e.g. for validating AI output, edge cases, enabling adoption, or acting on AI insights. Over time, reduce human input as the AI learns, and models improve. Target 60%+ initial gross margins, with a path to SaaS-like 75%+ margins over time. 𝟮. 𝗦𝗰𝗮𝗹𝗮𝗯𝗹𝗲 𝗛𝘂𝗺𝗮𝗻 𝗜𝗻𝘃𝗼𝗹𝘃𝗲𝗺𝗲𝗻𝘁: The dependency on hiring & training humans should not constrain scale and economics. Have a path to tapping into freelancers or agency partners. Leverage human experts in a high-talent location such as India. 𝟯. 𝗥𝗲𝗰𝘂𝗿𝗿𝗶𝗻𝗴 𝗥𝗲𝘃𝗲𝗻𝘂𝗲: Focus on high-value, recurring use-cases to ensure subscription-based revenue with strong net revenue retention (NRR). 𝟰. 𝗣𝗿𝗶𝗰𝗶𝗻𝗴 𝗣𝗼𝘄𝗲𝗿: Iterate to a solution that can command higher pricing, and a model that aligns incentives with customers, e.g. based on outcomes. 𝟱. 𝗗𝗮𝘁𝗮 𝗠𝗼𝗮𝘁𝘀: Build solutions that improve with use, creating compounding competitive advantages over time. 𝟲. 𝗠𝗼𝗱𝘂𝗹𝗮𝗿 𝗧𝗲𝗰𝗵: Architect a stack that can evolve with AI advancements. 𝟳. 𝗙𝘂𝗹𝗹-𝗦𝘁𝗮𝗰𝗸 𝗧𝗲𝗮𝗺: A founding team that has the technical expertise to build and rapidly improve complex AI-powered solutions, and deep operational acumen. A rare combination. These are complex businesses to build, and the right playbooks are yet to be perfected. But where this works, 𝗦𝗲𝗿𝘃𝗶𝗰𝗲𝘀-𝗮𝘀-𝗦𝗼𝗳𝘁𝘄𝗮𝗿𝗲 𝗔𝗜 𝗯𝘂𝘀𝗶𝗻𝗲𝘀𝘀𝗲𝘀 𝘄𝗶𝗹𝗹 𝗿𝗲𝗱𝗲𝗳𝗶𝗻𝗲 𝗺𝗮𝗻𝘆 𝗕𝟮𝗕 𝗰𝗮𝘁𝗲𝗴𝗼𝗿𝗶𝗲𝘀 📈 #EnterpriseAI #startups #vc #SaaS

AI has flipped the SaaS business model on its head. Because what costs nothing to build costs everything to integrate. Last weekend, I built a functional version of a Series B company's core product using AI tools. The build took 14 hours. The integration costs? Still ongoing after two weeks. This isn't unusual anymore. AI has dramatically compressed product development timelines while the complexity of fitting new tools into existing tech stacks remains unchanged. The implications for SaaS cannot be understated. When I examined where my weekend project hit roadblocks, it wasn't in creating features. It was in designing APIs, building connectors, ensuring compliance, and developing migration paths. The nature of value creation has shifted dramatically and few realize it. The majority of development hours weren't spent building core functionality, but rather on making it play nicely with everything else. This pattern repeats is a well-known problem already across the industry. Companies are discovering they can build sophisticated products and are fit for purpose, only to face the unchanged reality of enterprise integration challenges. Now this well known problem is taking a different face. The emerging reality: 1️⃣ Products that were once differentiators are becoming commodities 2️⃣ Integration capabilities now determine competitive advantage 3️⃣ Customer success teams matter more than development teams 4️⃣ Professional services revenue grows while license revenue shrinks As AI commoditizes building, integration becomes the new competitive moat. For founders, this means rethinking resource allocation. When product development costs approach zero, the relative value of integration expertise approaches infinity. The most successful SaaS companies of the coming era won't necessarily have the best products. They'll have the most seamless integrations. Technology value is fundamentally about what it connects, not what it contains. #startups #founders #growth #ai

Yesterday, Klarna’s CEO announced they were firing several SaaS providers (including Salesforce and Workday). This is just the start of a long, painful trend. Many SaaS companies will soon miss their targets and lose top accounts. Here’s why and how SaaS teams can prepare to avoid this cold winter: BACKGROUND: For those unfamiliar, Klarna offers flexible payment plans for online and offline purchases. With $2B in revenue and a $7.8B market cap, Klarna is a profitable company—a dream client for any enterprise SaaS vendor. Yet, Klarna does not see the need for many SaaS tools. Why? They’re not replacing Salesforce or Workday, but instead reimagining their entire workflows using AI. They’re realizing that many SaaS tools designed for old workflows are now obsolete. They may end up developing some in house AI tools to serve their proprietary workflows. Klarna is setting new benchmarks for AI-driven organizations, and their approach is likely to become mainstream within 18 months. As more companies adopt this model, the TAM for SaaS will shrink dramatically. SaaS vendors that quickly adapt to AI use cases will survive and thrive, but many will fade away as the market contracts. Additionally, Klarna recently announced it would cut over 50% of its workforce, reducing staff from 5,000 to under 2,000, driven largely by AI integration in marketing and CS. This will naturally lead to a reduction in SaaS spend, but the layoffs aren’t the cause—it’s Klarna’s AI-centric organizational design that’s causing the decrease in spend. Mark my words. SaaS vendors must evolve to stay alive. If you are a SaaS vendor, how do you cope with this? 1. Cannibalize your own business: Just like Netflix shifted from DVD rentals to streaming, SaaS companies must embrace AI and be willing to disrupt their own offerings. Those who don’t evolve will face the fate of Blockbuster. 2. Identify your AI use case: Look for adjacent areas where AI can bring quantum leaps and major productivity gains, not incremental improvements. Don’t just build another system of record—focus on real innovation that pushes boundaries and transforms your space. 3. Reduce Costs: The era of zero-interest rates led to bloated processes and structures. Now is the time for SaaS companies to learn to do business with lower costs and operate more efficiently. 4. Prepare for Price Pressures: Expect downward pricing pressure as AI-driven enterprises like Klarna redefine efficiency. Expensive SaaS solutions will struggle, while lower-priced alternatives will thrive in this new wave of innovation.   5. Go Global: In the past, selling in the U.S. alone was enough for a series A/B companies. But as competition in the US intensifies, expanding internationally with competitive pricing will offer much-needed room for expansion. SaaS vendors do have choices, but they ain’t pretty. They can disrupt themselves in the next 12 months, or... They can wait to be disrupted by others later. 

B2B SaaS founders will hate me for saying this. But the traditional B2B SaaS playbook is dead. Here's why: The smartest founders I know have stopped building traditional SaaS. Instead of selling software for a mere $5/seat/month, they're selling AI-enabled services that solve real business problems. These services are generating millions in ARR with skeleton crews and 50%-70% margins that make traditional SaaS businesses look weaksauce. Their philosophy is: Stop forcing customers to learn your tool and start solving their actual problems. When you charge based on outcomes (instead of seats), customers happily pay multiples of traditional SaaS pricing, because you are directly impacting their bottom line. I'm see smart tech founders implement this exact model and quietly banking seven-figure revenues in months: • An AI-enabled financial services company hit $4MM ARR in 18 months with 15 people while maintaining 70% margins • An AI-enabled IT services company hit $20MM ARR with 15 FTE and 50% margins • Creme Digital is a 7 figure agency and made $170k/m building low-code services on Lovable with under 10 people • Hundreds of AI Automation Agencies have exceeded 6 figures in ARR by building chatbots and voicebots for SMB workflows These companies are an entirely new species that combine the scalability and margins of software with the precision of human expertise. The economics are insane compared to both traditional SaaS and services: - 50%+ margins - Higher ACVs than SaaS (customers pay for outcomes, not licenses) - Faster sales cycles (solving painful problems = urgent purchases) But picking the right niche is everything. After analyzing countless businesses, I have identified a pattern: Look for businesses where highly-paid people waste time on spreadsheets and manual processes: - The VP who manually pulls data from 5 systems every week to create the same dashboard  - The law firms where associates bill $400/hour to review documents  - The financial team making million-dollar decisions based on Excel formulas There’s a reason these industries haven't adopted software yet: their workflows are too nuanced and specialized for one-size-fits-all SaaS. AI changes this completely. You can now build custom workflows for niche industries with a fraction of the engineering that traditional SaaS required. The key is solving very specific, high-value problems with AI plus human expertise. While everyone else chases the same crowded markets selling CRUD SaaS tools for $5/month, the real money is in specialized workflows that very few are addressing. The new playbook is: 1) identify a niche process 2) apply AI to automate 80% 3) add human oversight for the 20% that matters 4) charge based on outcomes This is one of the fastest paths to building a wildly profitable AI-Native business in 2025. ------------ If you liked this, follow me, Henry Shi as I decode what actually works for founders building in the AI era.

18 months in, AI isn’t eating SaaS—it’s eating the $4.6 TRILLION services budget. Our latest blog (link in comments) tracks the first year of Services‑as‑Software companies and distills lessons for founders: 1. Forward-deployed engineers evolved from configurators to product architects  Pre-AI era: FDEs configured systems and ran training. Now they're strategic assets who shadow users for weeks, mapping every undocumented workflow and tribal knowledge that keeps operations running. They convert edge-case business rules into runtime-editable parameters rather than hard-coded logic, instrument every decision point with telemetry for automated model retraining, and abstract successful patterns into reusable deployment modules. Each implementation feeds back into core product - turning one-off customization into scalable product leverage. 2. POCs became the sales process, breaking traditional economics AI performance depends on messy customer data, not demo environments. Startups now invest weeks of engineering pre-revenue - data ingestion, orchestration logic, prompt tuning. The cost of a failed POC includes headcount hours plus non-trivial token expenses. 3. Pricing shifts from enabling work to doing work Seats → usage → workflows → outcomes. But outcome pricing fails when customer variability is high. Harvey charges per lawyer but ties renewals to hours saved. AI SDRs retreat to usage pricing because sales outcomes depend on customer PMF. The trend is clear but messy. $4.6T in wages and services, not software budgets. The opportunity is vast, and the only currency that matters is the speed with which you can turn promises into provable results.