The full-stack startup, as previously written about by our partner Chris Dixon, is an approach taken by some startups to bypass incumbents by building a complete, end-to-end product or service. This approach is in contrast to the traditional approach, in which startups aim to sell or license their technology to existing companies in their industry. We believe that there are tremendous opportunities in American Dynamism categories for these full-stack companies to be built; moreover, we think these companies can enable a thriving ecosystem of new technology companies to grow in industries that were previously more difficult for new entrants.
The prominent technology-first entrants in industries that contribute heavily to American resilience and dynamism often emerge first in the form of full-stack startups. Generally speaking, these industries include sectors where companies sell to government (like aerospace and defense) and provide citizen services in government-adjacent markets (like education or transportation), and in core domestic industries where government is a stakeholder (like logistics and industrials). In general, these companies own their distribution and relationship with the end customer, as well as vertically integrate more (though by no means all) of what goes into their product. In some ways, they run counter to the conventional wisdom that advises startups to narrow in on a core competency early, use it as a wedge with customers, and then expand from there.
Full-stack startups represent the beginning of a new chapter of technological acceleration in these industries. They function as the enablers of new startups in the industry, by creating new markets for suppliers and other services around them. These new startups do not have to be full stack, and can grow into massive businesses in their own right, because the full-stack companies and the markets they operate in grow so large that even adjacent markets represent massive opportunities.
Some prominent examples of full stack startups in these industries include:
The goal of this post is to outline a framework for thinking about full-stack startups — in a technology-agnostic way — as enablers of innovation in government-adjacent sectors, and how this might translate to other important yet technologically stagnant industries. To do this, we explore: (1) why markets in certain industries lend themselves well to full-stack startups; (2) how full-stack startups accelerate technology development and adoption in these industries; (3) some common characteristics of full-stack startups in these markets; (4) areas of opportunity; and (5) why these companies are still anomalies.
The full-stack startup can be seen as a reaction against the challenges of reform in stagnant industries. Marc Andreessen has spoken about the limits of institutional reform, where organizations — particularly large bureaucracies — are incapable of improvement via reform for a variety of reasons. Instead, Marc argues that progress in the domains occupied by these incumbents is achieved by building new institutions from scratch, and that the challengers starting these new things have outsized opportunity because incumbents won’t change.
This dynamic is particularly visible in a number of verticals where technology contributes heavily to American Dynamism, including but not limited to aerospace, defense, public safety, education, housing, supply chain, industrials, and manufacturing. While these sectors are different across a range of dimensions, they often have at least one of a few qualities that make a full-stack startup approach compelling.
Some of these markets are largely dominated by a few extremely large players. Examples include automotive, which, though less consolidated than in the past, is largely an industry of large automakers with long histories; defense, which has consolidated since the 1990s and is dominated by the defense primes; and space launch, which prior to SpaceX was largely the domain of government.
This concentration creates a market dynamic that makes a vendor approach (selling technology to incumbents) more difficult for a venture-backed startup. First, the vendor approach faces an oligopsony given the high customer concentration among incumbents — meaning customers hold the power in this market, and moving downmarket becomes more difficult (increasing the likelihood of a vendor growing into a services provider for the largest incumbents). Secondly, the less competitive nature of this market dynamic can create less urgency for technology adoption, as well as a tendency to build technology in-house over buying from a new vendor.
A number of these markets are supply constrained due to some type of regulatory barrier or licensure requirements. These barriers mean that even if the market is incredibly large, value is likely largely captured by those incumbents able to participate in it; new technology vendors, in contrast, are only able to capture a smaller sliver of the market.
An example is higher education: While revenues from postsecondary institutions in the United States were $695 billion (2019-20), the players that are able to meaningfully participate in this higher education market are accredited, Title IV-eligible institutions. Most startups that try to serve this market are therefore only participating in a smaller sliver, such as college IT budgets or consumer educational spending. Similar dynamics exist in markets like transportation and housing. Though it may be a much more arduous and time-consuming path, startups that seek to overcome these barriers and compete with incumbents may see greater value accrual in the long run.
There are also often processes and markets within these industries that are characterized by high levels of fragmentation. While some markets within the industry are highly consolidated, as described above, other parts of the industry — particularly at the vendor, supplier, or intermediary levels — may be very fragmented. An example is the number of machine shops that make up the space, defense, and advanced manufacturing industrial base. The effects of this fragmentation can include long lead times, limited visibility, and complex coordination problems — and present an opportunity and incentive for the full-stack startup to own more of its value chain by making gains on speed, cost, and quality.
As they grow, full-stack startups often enable other technology companies to enter these categories. They act as a catalyst for many more startups, especially those that do not take the full-stack approach, to form (e.g., SpaceX and the NewSpace industry). This enablement typically happens via a few mechanisms:
New full-stack startups that compete with incumbents can increase competitive pressure on incumbents to procure and adopt similar technologies. If the technology leverage of these startups delivers faster growth or better results, it can catalyze interest in and budgets for comparable solutions from incumbents, creating an opportunity for a new generation of startups taking the vendor approach. For example, across multiple markets in supply chain management, including freight brokerage, freight forwarding, and B2B marketplaces, there is a recurring pattern where one wave of startups looks to disrupt and displace incumbents, and a second wave of companies emerge to sell to incumbents.
In some cases, an early full-stack startup serves as the infrastructure that enables a new set of markets and companies pursuing them. A prominent example of this dynamic is in the commercial space industry, where SpaceX serves as the launch infrastructure for a number of other space companies, products, and services, making these businesses economically viable by de-risking getting to space. In these cases, the difficult work done by the full-stack startup enables new companies and markets by lowering technological and economic hurdles.
Some of the full-stack startups discussed have created talent vortexes, and many alumni of these companies go on to build more technology-first companies in these categories. Full-stack startups often have the effect of developing talent that is familiar with the nuances and technicalities of their industry, but also well-versed in the Silicon Valley way of company-building and not necessarily stuck in the ways of the industry’s incumbents. Prominent examples include a number of energy and transportation startups founded by former Tesla employees; the startups in commercial space and adjacent verticals founded by former SpaceX employees; or the numerous companies founded by former employees of Palantir (which bears a number of similarities to the full-stack startups discussed here, though is also different in some ways).
There are a number of general patterns and attributes among full-stack American Dynamism startups that are worth discussing. They share these attributes even across industries because of the common qualities among these industries that make them particularly apt for the full stack approach. As characteristics like consolidation, barriers to entry, and fragmented value chains shape these markets and present an opportunity for full-stack startups, they also drive the following patterns to emerge across full-stack startups as they grow in these markets.
It’s common that processes and operations become just as important as product in these companies. Unlike with a vendor approach, full-stack startups often have to develop many more operational components of the business around their core technology advantage. Instead of selling some core product to incumbents, the full-stack startup effectively must build their own customer – the core technology looks more like an internal tool or platform that the business is built around. Consider, for example, a factory leveraging some level of automation for certain manufacturing processes. A vendor may sell or license their automation solution to other suitable factories, perhaps acting as a systems integrator as well; a full-stack startup will instead develop all the other components of the factory as well, effectively becoming their own customer and taking advantage of the technology leverage their solution provides.
Similarly, the process by which the product is built becomes a key differentiator against incumbents, as well. As full-stack startups typically end up building up the production or servicing capacity for their end product, magnified by some technology leverage, they also represent an opportunity to build new processes and operational advantages. In these cases, the startup’s production capacity and processes can be considered a product as well — in addition to the actual end product its customers buy.
These companies are often, though not always, more capital intensive. These greater capital needs typically result from a combination of needing to own more of the value chain and needing to navigate a more complex go-to-market motion.
Moreover, these factors also often mean the companies take longer to reach an inflection point in their growth. As such, a common challenge for full-stack startups is needing to raise more capital off less realized business traction than their peers. Here, a combination of historical capital efficiency, demonstrated product and technical progress, and validated demand can help the companies meet their capital needs.
As they scale, many of these companies become multi-product businesses. The transition makes sense as they have typically built both the relationships with customers and the systems and processes to expand into adjacent product areas. The multi-product model can also help justify the greater capital needs these businesses face.
The advantages of this full-stack approach are primarily around being able to compete on speed and flexibility, capturing more of the value the company creates, and building a number of moats at maturity. These companies are often extremely defensible, and we typically observe that they become n-of-1 companies. Moreover, as they enable more startups to be built in the vertical, it’s possible that growth also leads to more value accruing to the full-stack startup.
There are a number of markets and verticals that have some of the dynamics described earlier that lend themselves to full-stack startups, but have yet to see one reach massive scale. These verticals represent potential opportunities for full-stack startups to introduce dynamism to foundational and important industries.
Residential construction as an industry sees both high levels of fragmentation and barriers to entry, making it a compelling one for multiple types of full-stack startups to emerge. For example, Cover is a full-stack residential construction company that owns both the end customer relationship and production capacity, starting with accessory dwelling units (ADUs). There are also full-stack startups tackling specific markets and categories within the construction industry, such as Mosaic for general contracting, or companies like Agorus, Assembly OSM, and Diamond Age tackling different dimensions of automated residential construction (all of which go beyond being a pure technology vendor and operate in the business of homebuilding). Another slightly different example is Culdesac, which takes the full-stack approach not just to construction, but to urban development, by building neighborhoods from the ground up.
Perhaps the biggest challenge for the full-stack approach to residential construction is the web of state and local regulatory systems and processes that must be navigated in order to reach scale.
As discussed earlier, education has some of the market dynamics that make a full-stack startup approach suitable. In particular, higher education sees extremely high barriers to entry, and the most substantive work to create and deliver education of greater value is typically only doable by accredited, degree-granting institutions. To date, most education startups have focused on selling software, content, or services to schools (e.g. online program managers) or into supplementary consumer spending on education (e.g., most MOOCs). Full-stack startups in higher education often need to figure out ways to participate in this closed market — otherwise, they are confined to licensing curricula or selling software products to colleges.
Whereas one approach has been for a startup to partner with an existing college to deliver courses, Campus is taking the full-stack approach to two-year colleges by building a national online community college from the ground up. While the local nature of K-12 education makes a full-stack startup more challenging, it is also a market that represents an opportunity for new companies to introduce improvements to the status quo from the ground up.
While there have been full-stack companies built in both passenger and freight transportation, there are a number of verticals within transportation that have the market dynamics suitable for a full-stack approach. The full-stack approach here might involve building entire vehicles instead of being a supplier to OEMs. For example, startups like Intramotev and Parallel Systems are introducing autonomous, electric railcars for the rail industry, and Fleetzero is electrifying freighters, with a long term plan to build their own ships. These companies represent a starting point for full-stack startups to develop in the freight industries.
There are few industries as highly consolidated, or with as high of technical and economic barriers to entry, as semiconductor fabrication. Given the complexity and capital intensity involved in building foundries, this is a market typically left to incumbent integrated device manufacturers like Intel (itself an example of the full-stack approach) or third party foundries like TSMC. However, some early-stage companies, like Atomic Semi, are pursuing a full-stack approach of building foundries.
As a market, utilities have often been considered natural monopolies, and are highly regulated by utilities commissions, making them an particularly extreme form of a market where new technology might be introduced via a full-stack startup. Moreover, the modern history of public utilities markets in the United States has seen the introduction of more competition since the de-monopolization of electrical utilities in the late 1990s. It remains to be seen how a technology-first startup might be built in these categories, but it’s possible it may look very different to historical incumbents given the challenges in these markets.
None of this is to say that all startups in these categories need to take the full-stack approach — far from it. In fact, the vast majority of successful businesses built in these markets probably won’t take this approach. Full-stack startups are interesting precisely because they are anomalies, and often enable a number of other technology companies that don’t have to be full-stack. There are a few reasons we believe that to be the case.
First, they are simply hard businesses to build. They often require founders and teams to develop and exercise a variety of different competencies, and perform them all well — there cannot be only one core competency. By extension, there are simply more pieces of the business that need to be built before reaching product-market fit. Teams that can successfully execute on these visions are rare.
Second, as discussed earlier, the capital needs for many of these full-stack startups are comparatively greater than peer companies. Moreover, these companies often take a longer time to reach an inflection point as a business — after which they would likely have more business traction from which they can aggregate the capital needed. Prior to inflection, however, these companies can represent contrarian visions; founders that can successfully raise the capital needed for these visions are also rare.
Finally, many people will root for the entrepreneur building a full-stack startup to fail, reacting with skepticism and rooting against these companies. Many will look at a serious attempt to introduce ambitious technologies to important industries and dismiss it as “Silicon Valley hubris” — especially if these attempts are led by entrepreneurs who are industry outsiders. They’ll ask, cynically, whether these founders seriously think they can change an industry by building a venture-backed technology startup that differs from incumbents.
We certainly believe they can. If you’re building a full stack startup, we’d love to hear from you.
* * *
The views expressed here are those of the individual AH Capital Management, L.L.C. (“a16z”) personnel quoted and are not the views of a16z or its affiliates. Certain information contained in here has been obtained from third-party sources, including from portfolio companies of funds managed by a16z. While taken from sources believed to be reliable, a16z has not independently verified such information and makes no representations about the enduring accuracy of the information or its appropriateness for a given situation. In addition, this content may include third-party advertisements; a16z has not reviewed such advertisements and does not endorse any advertising content contained therein.
This content is provided for informational purposes only, and should not be relied upon as legal, business, investment, or tax advice. You should consult your own advisers as to those matters. References to any securities or digital assets are for illustrative purposes only, and do not constitute an investment recommendation or offer to provide investment advisory services. Furthermore, this content is not directed at nor intended for use by any investors or prospective investors, and may not under any circumstances be relied upon when making a decision to invest in any fund managed by a16z. (An offering to invest in an a16z fund will be made only by the private placement memorandum, subscription agreement, and other relevant documentation of any such fund and should be read in their entirety.) Any investments or portfolio companies mentioned, referred to, or described are not representative of all investments in vehicles managed by a16z, and there can be no assurance that the investments will be profitable or that other investments made in the future will have similar characteristics or results. A list of investments made by funds managed by Andreessen Horowitz (excluding investments for which the issuer has not provided permission for a16z to disclose publicly as well as unannounced investments in publicly traded digital assets) is available at https://a16z.com/investments/.
Charts and graphs provided within are for informational purposes solely and should not be relied upon when making any investment decision. Past performance is not indicative of future results. The content speaks only as of the date indicated. Any projections, estimates, forecasts, targets, prospects, and/or opinions expressed in these materials are subject to change without notice and may differ or be contrary to opinions expressed by others. Please see https://a16z.com/disclosures for additional important information.