--- title: "Industry Perspective: Why is Context so Important? | EP 02 Recording | Atlan" url: "https://atlan.com/wtf-context-layer/why-is-context-important-recording/" description: "Watch the full EP 02 recording. Every vendor, analyst, and framework has landed on \"context\" at the same time. Sanjeev Mohan gives his unfiltered take on what's real, what's hype, and what the industry is getting wrong." keywords: "context layer, AI governance, enterprise AI, data context, atlan events, context layer series, sanjeev mohan" --- > Atlan is hosting Context Conference, bringing together the leaders and builders at the frontier of giving AI the context it needs to understand their business. It runs online on October 28, 2026, from 11:00 AM to 2:00 PM ET. Atlan co-founder Prukalpa Sankar opens and closes the day. Leaders from AstraZeneca, BNY and Verizon share why they invest in context and what they get from it. Registrants get early access to The AI Context Gap, a new study from MIT Technology Review Insights. Register: https://atlan.com/context-conference/ Recording and write-up of Episode 2 of Atlan's WTF is the Context Layer? series, "Industry Perspective: Why is Context so Important?". Every vendor, analyst, and framework has landed on "context" at the same time. Independent analyst Sanjeev Mohan joins host Austin Kronz with his take on what's real, what's hype, and what the industry is getting wrong. The page carries the full video, key takeaways, and a blog article of the main Q&As. **Recording:** [EP 02 on Vimeo](https://player.vimeo.com/video/1195644090?h=9de30751c9) (the video is embedded on the page, with chapters). Episode page: [Industry Perspective: Why is context so important?](https://atlan.com/wtf-context-layer/why-is-context-important/) ## Key takeaways - Semantics tells you what; context tells you why - Zero context means zero agent performance - Agents don't understand disparate signals like humans do - ECL: extract context, not data - Regulated industries have the best AI foundation - The best strategy is execution - Build vs buy framing is the first adoption blocker - Context will be federated, not centralized ## Blog article: Industry Perspective: Why is Context So Important? In March, Gartner declared that [2026 is the year of context](https://www.linkedin.com/pulse/gartner-da-2026-where-context-layer-became-budget-line-prukalpa--n20tf/), but there's still confusion across the industry about what that means. Sanjeev Mohan, principal at SanjMo and former Gartner Research VP, is still fielding questions from enterprise buyers and vendors who conflate context with semantics; he even [published an FAQs blog](https://sanjmo.medium.com/faq-on-metadata-semantics-taxonomy-ontology-knowledge-graphs-and-context-c4a53bfda395) on it. With Austin Kronz he untangled six terms the industry treats as synonyms, pushed back on overcomplicated context engineering approaches, and gave an honest view of where enterprise teams stall when taking AI into production. ### Q: What's the difference between semantics and context? Semantics defines things: MRR means monthly recurring revenue; a customer is someone with an active subscription. Sanjeev described semantics as built for the BI world, where business users needed consistent definitions they could trust without understanding or writing SQL: rules, formulas, shared meanings. Context is everything semantics leaves out: the "why" behind a definition or event. His example: his recent purchase of a MacBook Air M5. Semantics can tell you the transaction happened, but not why he chose the M5 over the Pro, which sites he visited to research it, or what influenced the final decision. "All the 'why' questions that help complete a transaction have got nothing to do with semantics. That's context." (Sanjeev Mohan) The [semantics versus context distinction](https://atlan.com/know/context-layer-vs-semantic-layer/) matters more now because software is shifting from being built for humans to being built for agents. A human support rep can patch together disparate signals on the fly; an AI agent can't. Agents need scattered context assembled and delivered explicitly, which is why context engineering has emerged as a real job title. ### Q: What is ECL and why does it matter? [ECL stands for Extract, Context, Link](https://www.linkedin.com/posts/sanjmo_it-depends-105-from-etl-to-ecl-extract-share-7435284052963880961-FhO9/), Sanjeev's riff on ETL (Extract, Transform, Load). Extracting and moving data in ETL introduces latency, compliance overhead, and cost, and it was never meant for unstructured data. The context enterprises need lives mostly in unstructured sources: PDFs, parts manuals, support tickets, internal wikis. ECL extracts the context instead of moving the data, so agents can traverse it without the underlying data moving. "Instead of extracting and moving data, let's extract context. If it's a parts manual in a PDF, I extract the entities, I create vector embeddings, I do similarity search. I now know what's related to what. Then I link it so an agent can use it." Converging infrastructure shifts make this feasible: cloud separation of compute and storage, the Apache Iceberg standard, cheaper compute and object storage, faster networks, and lower egress costs between major cloud providers. ### Q: How much context do organizations need to start building agents? Sanjeev proposed a six-layer context architecture stack: metadata, semantics, taxonomy, ontology, knowledge graph, and context. Taxonomy, ontology, and knowledge graphs should not block you from starting. For most organizations, three layers are sufficient for a proof of concept: metadata as the foundation, semantics for consistent definitions, and context for situational awareness. Pick a single use case, connect those three layers, and start building. "Pick a business use case. Collect your metadata, create the semantic layer, and then do context engineering on top. Ontology, knowledge graph, taxonomy - those are important, but if they slow you down, get started without them." Organizations with large, frequently changing datasets and user populations may eventually need the full stack, but the "all six or nothing" framing causes hesitation. A working proof of concept built on three layers beats a stalled architecture debate. The more complex the framework, the higher the chance enterprises decide they lack the budget or expertise, and it never happens. ### Q: Where do teams actually hit the first blocker? Messaging: context is framed as something you must build, not buy. Technically accurate, practically damaging. "Build" means a new project, team, budget, and timeline, and most teams stop before they start. Sanjeev's counter: context is an extension of work teams are already doing. "You're not embarking on a different project. You're upskilling. You've done the metadata, you've got the glossary, the lineage. Now you're asking: how do I expand my semantics to bring in context from other sources? When you do that, you get the context." Adoption sticks when teams have an on-ramp that starts where they are and follows a smooth curve from metadata to semantics to taxonomy to context. Generative AI has removed the excuse that starting is too hard. The hard part now is finishing: taking a working agent to an enterprise-scale context layer. The way through is to start building and run headfirst into the real problems. The article closes by pointing to Episode 3, where David Mariani, co-founder and CTO of AtScale, examines how semantic layers and context layers relate: [watch the EP 03 recording](https://atlan.com/wtf-context-layer/is-semantic-layer-context-layer-recording/). ## Conversation leaders - **Austin Kronz** (host), Director of Data Strategy, Atlan. Former Gartner Director who spent years advising Fortune 500 data leaders on analytics strategy. He's the person large enterprises call when they can't figure out why their AI keeps getting context wrong. - **Sanjeev Mohan** (guest), Independent Industry Analyst. Left Gartner to call data trends honestly. Spent thirty years watching the industry name new categories, and watching half of them quietly disappear. ## Next session (as listed on the page) **Can a Context Graph Alone Make AI Reliable?** Thu, Sep 17, 12:30 PM ET. Late last year, Jaya Gupta called the context graph AI's next trillion-dollar opportunity: the reasoning behind a company's decisions that systems of record never stored. Jaya joins Austin to work through what a context graph must hold and where it stops. The registration form is on [this page](https://atlan.com/wtf-context-layer/why-is-context-important-recording/).