Low-hanging fruits are gone.

Where are the opportunities to build generational SaaS, infrastructure, and marketplace companies? The truth is most low-hanging fruit has already been picked, and the majority of new product ideas can be easily integrated as a feature by competition. Better products are not The remaining verticals have inherent low adoption. Incumbents with distribution are on a bundling spree. Sales channels are saturated, and customer acquisition costs take a bigger chunk of revenue. Enterprise customers are consolidating software tools, leading to the emergence of SaaS pricing and cost-saving products (e.g., Orb and Vendr).

The secret behind recent successful SaaS and marketplace companies.

But what about Rippling, Faire, Newfront, Figma, Service Titan, and Flexport? Rippling was the antithesis of conventional startup advice that startups should solve a small niche problem. From the beginning, Rippling aimed to build a constellation of software tools around employee information. Faire solved the trust problem in wholesale markets by providing free returns and net 60-day inventory financing. There is a graveyard of startups that have tried to cut out the insurance middleman, however, Newfront took the opposite path with tools for brokers. Figma built a community and marketplace for designers to unlock the growth required to outcompete Adobe’s gigantic distribution. Off the radar of Silicon Valley, Service Titan doubled revenue for more than five years with a massive software platform for all home services companies, which inherently has low adoption. However, the initial focus was solely on scheduling and dispatching software that was specific to just one trade, only plumbing companies. And probably the most prominent example is Flexport. Equipped with lessons from building a data platform for global trade, Ryan Peterson decided to build a vertically integrated global shipping juggernaut.

One might argue that these companies are all over the map with different markets, products, business models, and go-to-market strategies. Perhaps, that is the truth. **Aside from having ambitious founders, they all share a deep understanding of the problem maze, which is evident in founders’ familiarity with the technology and its unique applications. The next successful company will not be a Stripe for X. Most likely, it will be a unique, differentiated solution that solves a problem in a neglected market.

Where to build generation SaaS and marketplaces?

1. Business in-a-box solutions solving all major problems for nascent segments (e.g., AirBnB and Turo hosts, influencers on social media, and YouTubers) and/or existing ones (e.g., SMBs, and loan brokers) with integrated software solutions.
2. Platforms that enable new value networks of entrepreneurs and businesses making money by using or building complementary products and services on top of it. These platforms solve key technology and business problems with a simple solution and ****enjoy competition among their users and complimenters. They can have a wide range of use cases, from Figma and Notion, where designers and template builders sell digital products, to Shopify and AirBnB, enabling transactions on physical products.
3. Horizontal platforms for new markets such as bioinformatics, manufacturing digitization, and robotics. These platforms foster the growth of the whole sector and, from an investment standpoint, act like an ETF.
4. Products with simple user interfaces and fewer features powered by language models that flatten the learning curve of existing tools and target non-consumption and new markets. Design, CAD, and productivity software are prime examples in this category. In some industries like 3D printing and manufacturing, this abstraction on the user interface can allow modularizing of the value chain and aggregating manufacturers.
5. Low-end disruption of existing workflows with a simplified, automated, and cheaper product. A prime example is electronic design automation tools like Cadence and Synopsis, where AI can automate most steps, eliminate support functions in the process of designing a chip, and disrupt the existing business models that sell expensive annual licenses and computing packages.
6. AI native products with novel applications of intelligence functions and new user interfaces beyond the skeuomorphic design of LLM and chatbots (e.g., new mechanics that allow learning by AI). Intelligence functions are abstraction, logic, problem-solving, reasoning, critical thinking, and creativity. Some emerging patterns are:
   1. Productivity software for knowledge workers, including lawyers, accountants, investors, scientific researchers, procurement specialists, engineering and business consultants. Core benefits are automation, faster access to information, and better user experience.
   2. AI products replacing outsourcing services and processes. As AI automates repeating and outsourced tasks, the conventional model of productivity software will be challenged by startups delivering the final product. To name a few, outsourcing services in (a) accounting, financial, and insurance, (b) sourcing, procurement, and supply chain management, and (c) IT and consulting are ripe to be automated with AI.
   3. Search and information retrieval platforms. Some prime examples are (a) multimodal search, metadata tagging, and visual Q&A for videos and images (b) enterprise knowledge portal and help desk trained on documents and policies for HR, IT, and administration (c) sourcing and procurement assistant searching trade documents and product specs.
   4. Content and information generation platforms. Beyond obvious use cases like image, code, text, and presentation, building an end-to-end workflow with generated content can create and capture more value. For example, training and coaching platforms powered by language models and generative AI to help HR in creating courses, training material, and evaluation modules. Another example is similar to what Audible did to books, selling an augmented copy of books with short videos explaining key concepts.
   5. Data transformation, generation, and enrichment platforms. Language models can enable structuring unstructured data into databases and APIs, transforming data to other modalities, and creating synthetic data. Platforms providing these services can create moats by developing best-in-class data transformation models. In addition, they can be used to significantly improve business processes such as healthcare payment systems and freight forwarding and trade documentation.
7. Virtual reality infrastructure and application for consumers and businesses. The VR headset market is currently a duopoly between Apple - a closed platform - and Facebook, with an alternative similar to Android. This creates a huge whitespace for startups to fill Google’s shoes in search, maps, and streaming. In addition, the infrastructure layer for VR is up for grabs, as there is no analog for Figma, DataDog, or Shopify in the VR space. Finally, generational companies will be created in applications and platforms layers around entertainment, shopping, education, and social.
8. Vertical software. The largest vertical SaaS companies (e.g., Toast) have a full suite of offerings to maximize revenue and profit, shorten payback, and retain customers. Some avenues to achieve this goal are:
   1. Targeting niche markets: There are still niche markets (e.g., catering ($68B) and energy management) and growing subsegments in markets with large TAM (e.g., social media influencers) to build an operating system-level software with additional offerings, including payments, procurement, and marketplaces.
   2. Hardware-software integration. Pure software can’t create enough value for large untapped markets like manufacturing and food processing. However, integrating technologies like visual inspection with software can generate enormous value.
   3. Constellation Software playbook. Rippling has applied Constellation Software’s playbook in HR by owning a core system of record. There will be startups doing the same in agriculture, healthcare, real estate, and financial services industries by finding the common denominator across all transactions.
9. Marketplaces. Particularly three areas:
   1. Simplifying user experience and atomizing services with language models. Services are still not searchable, and users need to fill out a form and wait for a quote. There is a huge opportunity to make services instantly available. ****
   2. Markets with complex sales processes and need for customization that requires sales agents. A huge portion of B2B products, including parts & equipment, machinery, and raw material, is sold by sales agents and has detailed spec and information. Language models can finally solve discoverability, augment sales agents, and eliminate long response times in B2B marketplaces.
   3. Workforce marketplaces disrupting staffing agencies. Categories like construction subcontractors and retailers are looking for on-demand reliable workforces to replace job boards, paper timesheets, and long lead times.