In some ways, the dominant mode of legal service delivery resembles the old VHS or DVD model, whereby everyone rented movies by walking or driving to their local video store. For most legal services, a client identifies a need, contacts outside counsel, and receives a tailored product in response to a discrete request. The model reflects the nature of most legal work: fact-specific, judgment-intensive, and often difficult to standardize. But it also reflects the limits of the available delivery mechanisms. Legal expertise is mostly accessed matter-by-matter, lawyer-by-lawyer, and document-by-document, and therefore, like the old video rental store, it cannot scale exponentially.
That model remains indispensable for many legal services, especially those that are novel, sensitive, or strategically consequential. As we recently wrote, legal services are like waiters at restaurants in at least one respect: even when clients are fairly sure what they need, they often want an expert human to guide and validate their decisions. So traditional legal services will almost certainly survive the rise of AI.
But there is a category of legal work that is recurring, operational, and time-sensitive. For those matters, the friction of the traditional model can be limiting. The issue is not merely cost. It is latency, inconsistency, and the practical difficulty of making institutional legal knowledge available at the moment that it is most needed.
A useful analogy is the transition from video rentals to streaming. Instead of requiring a distinct trip for each transaction, streaming made content continuously available, searchable, and integrated into the user’s ordinary workflow. That shift did not eliminate the underlying product. It altered the way the product was delivered and experienced, creating the opportunity for significant scalability.
A similar transformative shift may be underway for some kinds of legal services that can be made available through a platform model that is more continuous, modular, and operationally useful than the traditional matter-based approach.
Debevoise’s STAAR, built with Legora’s Portal, does not displace legal advice with generic automation. Nor is it a marketing layer attached to conventional practice. Rather, it represents an effort to rethink how the same legal expertise that is needed by many clients is organized and delivered. The underlying premise is that valuable legal service does not necessarily require bespoke bilateral engagement, which has several implications.
First, it treats at least some legal know-how as capable of being structured into reusable assets rather than preserved only in lawyers’ memories, prior work product, or informal internal channels. Second, it assumes that clients benefit when legal knowledge is made more readily accessible. Third, it suggests that the relevant unit of value may sometimes be a system of access and triage, rather than a standalone memo, email, or call.
Building a platform like STAAR requires many decisions about which kinds of legal knowledge lend themselves to structuring and which do not. Legora has deployed Portal across a range of firms and practice areas, and the pattern is not binary. Some legal guidance is essentially rules-based (e.g., applicable regulations, jurisdictional triggers, disclosure thresholds, compliance checklists) and often can be structured. Other legal advice is judgment-based (e.g., novel or undecided areas of the law, or advice that is highly dependent on context, relationships, and strategic nuance) and therefore does not structure well. Our experience deploying Portal has shown that the line between those two categories is partly a design choice. Firms that draw it too narrowly—treating almost everything as judgment-based—leave potential efficiency gains unrealized. Those that draw it too broadly tend to automate in ways that erode quality and trust. Getting that boundary right, and revisiting it regularly as AI capabilities evolve, is one of the most consequential decisions a firm will make when building this kind of platform. Portal is designed for both sides of that boundary. It structures and delivers the work that can be systematized, and it serves as the platform where lawyers can conduct work on the complex, high-stakes matters that demand their full judgment—with the firm’s institutional knowledge organized and accessible beneath them rather than locked in someone’s inbox.
These platforms are more than just commoditized legal self-service, in the same way that Netflix provides users with personalized recommendations for movies based on their viewing history. For law, these platforms provide clients with access to high-quality legal content that can be easily customized to the individual client, with escalation options and access to firm lawyers for collaboration, clarification, and verification. So they sit between pure self-service and fully customized counseling.
That said, the streaming analogy has limits. Streaming succeeded in part because movies are standardized consumer goods, while legal advice is not. It is embedded in institutions, facts, incentives, regulatory environments, and relationships of trust. If Netflix had not been able to deliver the same high-quality movies that were in video stores, it probably would have failed, even though it has a superior delivery system. Success comes from improving delivery without sacrificing quality, which was easy to do for movies, but is very hard for legal services. And even when the content is the same, the experience matters. Netflix did not eliminate the movie theater. Sometimes the popcorn, the crowd, the big screen, and the previews are as important for an enjoyable evening as the movie itself.
For legal work, one additional benefit of platforms like STAAR is that they improve the allocation of legal expertise by allowing lawyers to focus more of their time on matters that are genuinely novel, contested, or strategic. Clients get direct access to a constantly improving and expanding database of trusted legal content, which works well with the cadence of modern business operations, and when they do talk to their lawyer, those calls are efficient and productive because they are informed by the platform content already accessed and reviewed.
There is also a broader institutional point. For decades, legal know-how has often been treated as something embodied primarily in individuals. That remains true in part, and it will remain true for the most complex forms of judgment. But platform-based delivery systems demonstrate that some significant portion of a law firm’s value lies in its ability to organize knowledge systematically, maintain it responsibly, and make it available in forms that are most usable to clients.
This is hard to do well. Netflix did not displace video stores because it had a good idea. It succeeded because it solved a series of extraordinarily difficult problems—content licensing, distribution logistics, recommendation architecture, quality consistency—over many years. Building a successful legal knowledge platform involves structuring content so that it is accurate, consistent, and usable. It also requires designing escalation pathways that lawyers and clients will actually trust, avoiding staleness as the law evolves, and building feedback loops that make the system better over time. Platforms that earn the confidence of sophisticated institutional clients—the kind that will put their name behind what they are accessing—are not built quickly or cheaply. That difficulty is not an argument against building them. It’s an argument for starting now.
The firms that succeed will be precise about where bespoke advice is necessary. They will take risks trying to build systems for the kinds of work that can be structured, recognizing that some of those efforts will be costly and will fail, but they will learn from those failures. Some firms may resist the switch to platform delivery because, within the billable-hour framework, there may be concerns about the loss in short-term profitability that comes with improved efficiencies from automation. And they may be right to wait, but at some point, they may risk the fate of the video stores.
On that reading, STAAR represents something beyond a single firm’s product innovation. It points to a model that we believe will help define the next generation of legal service delivery: a combination of structured, platform-accessible knowledge and highly skilled lawyers deployed where judgment is genuinely required. The firms that thrive will be the ones that use platforms like Portal to deliver the work that can be responsibly and effectively systematized, which will vary depending on the practice. But in doing so, the practice will earn the trust and attention of clients for the work that cannot be automated.
Whether particular legal platforms succeed will depend on the quality of the underlying legal content, the discipline of maintenance, the clarity of escalation pathways, and the platform’s day-to-day usefulness for clients. But on the whole, we believe that platforms like the STAAR Portal are not just reserved for the routine. They have the potential to provide an operating layer for a broad spectrum of legal work—delivering the recurring at scale and giving lawyers a better foundation for the complex, high-stakes matters that demand human expertise.
* * *
To subscribe to the Data Blog, please click here.
The cover art used in this blog post was generated by combination of ChatGPT Images 2.o and Nana Banana 2.
The Debevoise STAAR (Suite of Tools for Assessing AI Risk) is a monthly subscription service that provides Debevoise clients with an online suite of tools to help them responsibly fast-track their AI adoption. Please contact us at STAARinfo@debevoise.com for more information.
