Optimizing Workflow Processes

How can businesses go beyond using AI for incremental efficiency gains to create transformative impact? I write from the World Economic Forum (WEF) in Davos, Switzerland, where I’ve been speaking with many CEOs about how to use AI for growth. A recurring theme is that running many experimental, bottom-up AI projects — letting a thousand flowers bloom — has failed to lead to significant payoffs. Instead, bigger gains require workflow redesign: taking a broader, perhaps top-down view of the multiple steps in a process and changing how they work together from end to end. Consider a bank issuing loans. The workflow consists of several discrete stages: Marketing -> Application -> Preliminary Approval -> Final Review -> Execution Suppose each step used to be manual. Preliminary Approval used to require an hour-long human review, but a new agentic system can do this automatically in 10 minutes. Swapping human review for AI review — but keeping everything else the same — gives a minor efficiency gain but isn’t transformative. Here’s what would be transformative: Instead of applicants waiting a week for a human to review their application, they can get a decision in 10 minutes. When that happens, the loan becomes a more compelling product, and that better customer experience allows lenders to attract more applications and ultimately issue more loans. However, making this change requires taking a broader business or product perspective, not just a technology perspective. Further, it changes the workflow of loan processing. Switching to offering a “10-minute loan” product would require changing how it is marketed. Applications would need to be digitized and routed more efficiently, and final review and execution would need to be redesigned to handle a larger volume. Even though AI is applied only to one step, Preliminary Approval, we end up implementing not just a point solution but a broader workflow redesign that transforms the product offering. At AI Aspire (an advisory firm I co-lead), here’s what we see: Bottom-up innovation matters because the people closest to problems often see solutions first. But scaling such ideas to create transformative impact often requires seeing how AI can transform entire workflows end to end, not just individual steps, and this is where top-down strategic direction and innovation can help. This year's WEF meeting, as in previous years, has been an energizing event. Among technologists, frequent topics of discussion include Agentic AI (when I coined this term, I was not expecting to see it plastered on billboards and buildings!), Sovereign AI (how nations can control their own access to AI), Talent (the challenging job market for recent graduates, and how to upskill nations), and data-center infrastructure (how to address bottlenecks in energy, talent, GPU chips, and memory). I will address some of these topics in future posts. [Original text: https://lnkd.in/gbiRs2mi ]

McKinsey & Company 𝗯𝗹𝘂𝗲𝗽𝗿𝗶𝗻𝘁 𝗳𝗼𝗿 𝗵𝗼𝘄 𝗯𝗮𝗻𝗸𝘀 𝗰𝗮𝗻 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝗲𝘅𝘁𝗿𝗮𝗰𝘁 𝗿𝗲𝗮𝗹 𝘃𝗮𝗹𝘂𝗲 𝗳𝗿𝗼𝗺 𝗔𝗜: ⬇️ This is a full-stack, enterprise-grade architecture — built on agents, orchestration, and rewired workflows. The AI bank stack consists out of 4 key layers: ⬇️ 𝟭. 𝗘𝗻𝗴𝗮𝗴𝗲𝗺𝗲𝗻𝘁 𝗟𝗮𝘆𝗲𝗿 This is the user layer — customers and employees. McKinsey calls for fully reimagined, intelligent, personalized experiences across all channels. → Multimodal chat (text, voice, image) → Omnichannel UX across mobile, contact center, branch → Digital twins for customer simulation and workforce training It’s all about a UI refresh and UX overhaul grounded in real AI. 𝟮. 𝗔𝗜-𝗣𝗼𝘄𝗲𝗿𝗲𝗱 𝗗𝗲𝗰𝗶𝘀𝗶𝗼𝗻 𝗠𝗮𝗸𝗶𝗻𝗴 This is the brain of the AI-first bank. And it’s not just predictive models anymore — it’s orchestrated agent ecosystems. → AI Orchestrators: Plan, reason, delegate across workflows → Domain Agents: Specialize in credit policy, fraud, risk, legal → Copilots: Embedded in workflows to guide users and automate decisions McKinsey reports 20–60% productivity gains in decision-making with this approach. 𝟯. 𝗖𝗼𝗿𝗲 𝗧𝗲𝗰𝗵 & 𝗗𝗮𝘁𝗮 The foundation layer most banks underestimate — until GenAI models stall in production. → Vector databases → LLM orchestration and FinOps → Search and retrieval engines → ML pipelines → Secure data architecture → API infrastructure The goal: make data accessible, tools reusable, and infra invisible to the business. Without this, nothing scales. 𝟰. 𝗢𝗽𝗲𝗿𝗮𝘁𝗶𝗻𝗴 𝗠𝗼𝗱𝗲𝗹 This is where the transformation wins or fails. Without rewiring the org, the tech doesn’t matter. → AI control towers to track value and set guardrails → Cross-functional teams across business, tech, and AI → Platform operating model for speed and alignment → Enterprise-wide reuse of AI capabilities If you're building isolated projects without shared assets or central coordination, you’re not transforming — you’re experimenting. 𝗪𝗵𝗮𝘁 𝘁𝗵𝗶𝘀 𝗮𝗹𝗹 𝗮𝗱𝗱𝘀 𝘂𝗽 𝘁𝗼? The banks that win won’t be the ones with the most pilots. They’ll be the ones that industrialize agents, orchestration, and rewired workflows, with full-stack coordination. Full McKinsey article: https://lnkd.in/dPaJzVK4 𝗜 𝗲𝘅𝗽𝗹𝗼𝗿𝗲 𝘁𝗵𝗲𝘀𝗲 𝗱𝗲𝘃𝗲𝗹𝗼𝗽𝗺𝗲𝗻𝘁𝘀 — 𝗮𝗻𝗱 𝘄𝗵𝗮𝘁 𝘁𝗵𝗲𝘆 𝗺𝗲𝗮𝗻 𝗳𝗼𝗿 𝗿𝗲𝗮𝗹-𝘄𝗼𝗿𝗹𝗱 𝘂𝘀𝗲 𝗰𝗮𝘀𝗲𝘀 — 𝗶𝗻 𝗺𝘆 𝘄𝗲𝗲𝗸𝗹𝘆 𝗻𝗲𝘄𝘀𝗹𝗲𝘁𝘁𝗲𝗿. 𝗬𝗼𝘂 𝗰𝗮𝗻 𝘀𝘂𝗯𝘀𝗰𝗿𝗶𝗯𝗲 𝗵𝗲𝗿𝗲 𝗳𝗼𝗿 𝗳𝗿𝗲𝗲: https://lnkd.in/dbf74Y9E

The Great Orange Processing Wars: ETL vs ELT vs EtLT Three factories. Three philosophies. One truth most data engineers miss. 𝗘𝗧𝗟 — The Juice Bar Squeeze before you store. Quality-first, schema-first, compliance-first. Works perfectly until your business logic changes — then you re-run everything. Best for: fraud detection, classical ML, regulated industries. Hidden cost: you can't un-squeeze juice. Raw fidelity is gone forever. These need strict compliance and clean, pre-processed data before loading. 𝗘𝗟𝗧 — The Warehouse Store everything raw. Transform strictly on demand. Pure flexibility. Works perfectly until nobody knows what's in Aisle 47 anymore. Best for: ad-hoc analytics, LLM training data, fast-moving startups. Hidden cost: "we'll clean it later" becomes never. Swamps, not lakes. Here, raw data is loaded first, then transformed as needed—great for massive, unstructured, fast-changing datasets.  𝗘𝘁𝗟𝗧 — The Gourmet Factory The lowercase t changes everything. E — Extract faithfully t — Fix only what must be fixed early (PII masking, dedup, type casting) L — Load into columnar storage T — Transform richly at query time, per use case The small t handles what belongs at ingestion: compliance, deduplication, partitioning. The big T handles what belongs at consumption: business logic, ML features, aggregations. Best for: RAG pipelines, LLM training, GDPR systems, real-time personalization. Used, when different data types need different strategies. The real choice isn’t just technical; it’s about timing. → ETL commits early. → ELT commits late. → EtLT commits minimally early — but maximally late. That tradeoff shows up everywhere in system design. Eager vs lazy. Compiled vs interpreted. Normalized vs denormalized 💡 The Modern Data Engineering Playbook Stop asking: "ETL or ELT?" Instead, start asking: "What fuels my AI use case?" 🍊 Traditional ML? → Serve up that perfectly squeezed juice (ETL) 🥤 Exploratory AI? → Keep the whole fruits on hand, blend them later (ELT) ⚡ AI at Scale? → Prepare smartly, create a variety (EtLT) 🎯 The Truth Bomb AI is like the promised land—bringing automation, intelligence, and scalability. But without the right pipeline? It’s just chaos. The best data engineers don’t play favorites. They choose the right tool for the right need. What’s your philosophy on data pipelines? 👇 Share it in the comments!

Selecting the right database is a crucial decision that impacts an application’s performance, scalability, and maintainability. With the growing complexity of data and diverse use cases, it is essential to understand the different types of databases and their strengths.  Relational Databases (SQL)   SQL databases follow a relational model, making them suitable for applications requiring structured data storage, consistency, and ACID compliance. These databases are widely used for financial systems, ERP, and CRM applications.   Examples: MySQL, Microsoft SQL Server  NoSQL Databases (Flexible and Scalable Storage)   NoSQL databases offer flexibility in data storage and retrieval, making them ideal for unstructured and semi-structured data. There are several types of NoSQL databases, each optimized for different workloads:  - Document Databases: Store data in a flexible, JSON-like structure, making them suitable for content management and applications with dynamic schemas.     Examples: MongoDB, Couchbase   - Key-Value Databases: Designed for high-speed lookups and caching, commonly used in session storage and real-time applications.     Examples: Redis, Amazon DynamoDB   - Columnar Databases: Optimized for analytical workloads and efficient query performance in data warehouses.     Examples: Amazon Redshift, Apache Cassandra   - Graph Databases: Best suited for applications requiring complex relationships and network analysis, such as social media platforms and recommendation engines.     Examples: Neo4j, Microsoft Azure Cosmos DB Specialized Databases for Specific Use Cases   - Time-Series Databases: Designed to store and analyze time-dependent data, such as monitoring and event tracking.     Examples: InfluxDB, TimescaleDB   - Vector Databases: Optimized for AI workloads, including similarity search and machine learning applications.     Examples: Milvus, Pinecone   - Spatial Databases: Enable geospatial data storage and querying for applications such as mapping and geographic information systems (GIS).     Examples: PostGIS, Oracle Spatial High-Performance and Emerging Databases   - In-Memory Databases: Store data in RAM to deliver ultra-fast performance, often used in real-time analytics and transaction processing.     Examples: SAP HANA, MemSQL   - NewSQL Databases: Combine the scalability of NoSQL with the reliability of SQL, making them suitable for distributed applications requiring strong consistency.     Examples: Google Spanner, CockroachDB Blockchain and Object-Oriented Databases   - Blockchain Databases: Ensure data integrity, immutability, and security, making them ideal for decentralized applications.     Examples: BigchainDB, Chainbase   - Object-Oriented Databases: Align with object-oriented programming paradigms and are suitable for applications handling complex data structures.     Examples: db4o, ObjectDB Database selection is not simply a choice between SQL and NoSQL but rather a strategic

Ever merged code at 3 PM on a Friday and immediately regretted your life choices? 🥲 Yeah, that's what happens when your team has no real Git branching strategy. Here's the Good, Bad, and Ugly about the 3 main approaches teams use: 𝟭) 𝗚𝗶𝘁𝗙𝗹𝗼𝘄 → 𝗧𝗵𝗲 𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲𝗱 𝗼𝗻𝗲 ↳ Perfect when you need to support multiple versions at once (think mobile apps where users have v2.3, v2.4, v3.0 all running) ↳ You create separate branches for features, releases, and hotfixes. ✅ Good: Multiple versions? No problem. ❌ Bad: Slows you down if you need speed. 𝟮) 𝗚𝗶𝘁𝗛𝘂𝗯 𝗙𝗹𝗼𝘄 → 𝗧𝗵𝗲 𝘀𝗶𝗺𝗽𝗹𝗲 𝗼𝗻𝗲 ↳ One rule: main is always deployable. ↳ Create branch → Make changes → Review → Merge → Deploy. ✅ Good: Fast, clean, easy to understand. ❌ Bad: Needs discipline. One bad merge breaks everything. 𝟯) 𝗧𝗿𝘂𝗻𝗸-𝗕𝗮𝘀𝗲𝗱 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗺𝗲𝗻𝘁 → 𝗧𝗵𝗲 𝗯𝗼𝗹𝗱 𝗼𝗻𝗲 ↳ Everyone commits small changes to main, multiple times a day. ↳ Feature flags hide incomplete work. ✅ Good: Super fast integration, forces you to build solid tests. ❌ Bad: Junior devs can accidentally blow things up. My advice ↳ Pick what fits YOUR team, not what's trendy. ↳ Start simple. You can add complexity later (but you'll never successfully remove it). 💬 Which strategy does your team use? And honestly... is it working for you? — 🔗 And if you missed this video, it shows you how your Git strategy actually powers your entire CI/CD pipeline. Check it out if you want to see the bigger picture: https://lnkd.in/ddsjPsUa

Building Data Pipelines has levels to it: - level 0 Understand the basic flow: Extract → Transform → Load (ETL) or ELT This is the foundation. - Extract: Pull data from sources (APIs, DBs, files) - Transform: Clean, filter, join, or enrich the data - Load: Store into a warehouse or lake for analysis You’re not a data engineer until you’ve scheduled a job to pull CSVs off an SFTP server at 3AM! level 1 Master the tools: - Airflow for orchestration - dbt for transformations - Spark or PySpark for big data - Snowflake, BigQuery, Redshift for warehouses - Kafka or Kinesis for streaming Understand when to batch vs stream. Most companies think they need real-time data. They usually don’t. level 2 Handle complexity with modular design: - DAGs should be atomic, idempotent, and parameterized - Use task dependencies and sensors wisely - Break transformations into layers (staging → clean → marts) - Design for failure recovery. If a step fails, how do you re-run it? From scratch or just that part? Learn how to backfill without breaking the world. level 3 Data quality and observability: - Add tests for nulls, duplicates, and business logic - Use tools like Great Expectations, Monte Carlo, or built-in dbt tests - Track lineage so you know what downstream will break if upstream changes Know the difference between: - a late-arriving dimension - a broken SCD2 - and a pipeline silently dropping rows At this level, you understand that reliability > cleverness. level 4 Build for scale and maintainability: - Version control your pipeline configs - Use feature flags to toggle behavior in prod - Push vs pull architecture - Decouple compute and storage (e.g. Iceberg and Delta Lake) - Data mesh, data contracts, streaming joins, and CDC are words you throw around because you know how and when to use them. What else belongs in the journey to mastering data pipelines?

🔮 How To Prioritize UX Work (Framework) (https://lnkd.in/eGQrPm2N), a very practical guide on how to choose and estimate the right level of research and UX work needed for a successful outcome of a project — along with the process to follow and UX estimates to set. Kindly shared by Jeremy Bird. 🤔 Planning is typically done for the delivery phase only. 🤔 Design, research, discovery, ideation are not planned. 🤔 Effort, estimates, roadmaps, capacity are rare for UX work. 🚫 Not every project needs the same level of research/design. ✅ Goal: set realistic expectations for UX work in a timeframe. Jeremy suggests to estimate research and design efforts separately, and across different dimensions: we assess research by mapping Risks and Problem Clarity. And we estimate design effort needed by mapping Risk and Level of Complexity: 🔮 Clarity: Low ↔ High New, unknown problems usually come with a lot of assumptions and very low clarity. Well-known problems with shared understanding in the team and some extensive research have higher degree of clarity. 🔥 Risk: Low ↔ High Some projects are relatively easy to roll back and they don't really affect business-critical workflows (low risk). Others are much more difficult to reverse and operate within users' key journeys (high risk). 🚀 Complexity: Low ↔ High Self-contained projects in well-understood workflows are typically straightforward (low complexity). Some projects that involve many systems, external dependencies, stakeholders scattered across teams with little existing knowledge (high complexity). ✅ We start by defining a problem to solve + business impact. ✅ Then, we shape desired user outcome and success criteria. ✅ Next, we assess design effort and research effort levels. ✅ Run a kickoff meeting to prioritize and decide the scope. ✅ Designers break down UX work, estimate it, add to Jira. Personally, I always find it remarkably difficult to estimate the effort for research or design work. Even after so many years, with 20–30% buffer, I’m often underestimating the little nuances, blockers, constraints and bottlenecks hidden away somewhere between complex dependencies and external stakeholders. One thing is certain though: considering risk early is a very, very effective way to guide UX work in the right direction. High risk always requires some level of research and discovery. And early prioritization helps UX teams focus their effort where they add most value — saving time on resources for projects that deliver value to users and businesses. Finally: I can highly recommend to consider John Cutler's Effort vs. Value curves (https://lnkd.in/evrKJUEy) for prioritization work as well. Much of the work isn’t completed once it's delivered. More often than not, it will significantly add to maintenance costs over time. We better account for it early. #ux #design

2026 planning starts now. If I was the CRO of a $50M business looking to grow 30% next year (i.e. add $15M of net new ARR to end the year at $65M) here’s exactly what I’d do: ASSUMPTIONS: - Selling into SMB and Mid-Market but with a small Enterprise sales effort. - 82% Gross Revenue Retention and 95% Net Revenue Retention (NRR). - A CS team that has renewal targets but expansion is handled by the AEs. - New business team is hitting quota in total but unevenly distributed. 1. Stress Test the Targets and the Revenue Model Look at 2025 growth and compare to total investments in sales and marketing focused on new business growth to understand CAC to ARR growth. Confirm the ratios map to the budget — e.g. you’re not being asked $15M in growth on the *same* CAC investment.    Assume CAC will degrade by 10% and ensure your fully weighted S&M investment is Pro-Rata + 10% to the growth. We’re looking for rough confirmation we’re not being asked to perform miracles. 2. Stress Test Pipeline Coverage and Marketing Performance Ensure we understand Lead to Closed Won Cycle and we have coverage. If we have a 3 month sales cycle and it’s mid-September, we’re on track. But if we wait much longer we’ll be drifting into Q1 and will immediately be behind. As usual, we’re looking for 3-5x pipeline coverage. 3. Understand Demand Generation Channels Word of mouth is not (really) a channel. It’s a “channel” if you can put $ behind it and the more you spend the more you get. If we have our basic framework in place, it’s time to get out the precision tools, modeling CAC, retention, and LTV by *investable channel*. At higher ACVs, we put muscle behind in-person travel, ABM, and targeted field marketing. At lower ACVs, we need investments in data and enrichment to enable effective paid acquisition and AI-enabled inside reps. 4. Review Gross and Net Revenue Retention targets If nothing changes with NRR, we are looking at $47.5M end of year run-rate. Let’s figure out if we can push NRR up to 105%, lowering the burden on new business. How? - Segment accounts by Red, Yellow, Green - Assign commercial support to CS to expand Green through more seats, new products, or deeper usage. Take one high performing AE and turn them into an Account Management expansion focused hunter whose sole job is converting upsells. 5. Drive Our AEs with Great Variable Comp and Route Our Best Leads to Our Best People Top sellers are 5-7x more productive than average sellers. And sellers with unlimited upside and generous accelerators, do better. I'd design our comp plans to pay for over-performance and route our leads to our best people. Target 80%+ quota attainment and be willing to part with the bottom 20%.  Get confidence every lead we send to the sales team closes at a higher rate with a higher deal value. The last step? Pop bottles because we hit our number 🍾 P.S. Want to learn how to do this as a scaleup CRO? Pavilion's CRO School starts 10/2. DM me to join.

Sometimes, the simplest methods teach the most powerful lessons. Take the Japanese multiplication method: instead of numbers alone, it uses lines and intersections. You visually break down problems, count connections, and combine results to get the answer. Do you use this method? Why does this matter for tech leaders today: ✅ Logic & Decomposition – Breaking complex problems into manageable pieces mirrors how we design algorithms, optimize systems, and lead projects. ✅ Memory & Pattern Recognition – Visualizing relationships strengthens cognitive agility, crucial for data-driven decisions. ✅ Error Detection & Attention to Detail – Seeing patterns makes mistakes obvious—just like in code reviews, system architecture, or financial modeling. Leadership in tech isn’t just about tools; it’s about training your mind to see structure in complexity. Centuries-old logic can still sharpen the skills that drive modern innovation. 💡 Next time you face a complex challenge, think like a line multiplier: visualize, decompose, connect, and solve. #TechLeadership #Logic #ProblemSolving #Innovation #CognitiveSkills #BusinessStrategy

2025 was the year of the enterprise agent. 2026 will be the year of the workflow. Businesses are excited about AI agents, and rightfully so. But the most successful deployments share one common trait: They all started with rock-solid workflows. Workflows are simply the series of steps in a process to complete a goal or a task. Some can be agentic; many aren’t. But all are important. Imagine an Olympic relay race. You can have the world’s fastest runners, but if they fumble the baton handoff, you lose. The same is true for AI agents in your business. An agent might be powerful, but if it's operating within broken processes, it's going to drop the baton. At Docusign, we've seen this firsthand with our Intelligent Agreement Management (IAM) platform. The biggest customer beneficiaries have built structured, end-to-end AI workflows where agreements are generated, routed for approval, and integrated with downstream systems without manual intervention. The payoff isn't just internal. By pulling their data into workflows from start to finish, their clients or new hires get faster onboarding and can engage sooner. Perceptyx, an employee experience platform company, saw this when it modernized its manual client-facing workflows using Docusign IAM. It spent 99% less time with customers generating documents and processing NDAs. Why Agreement Workflows Break Agreement workflows are messier than most people realize. Legal needs to approve, finance has to review, compliance might weigh in, and the process can repeat multiple times. An RFP process might start in CRM, move to email and Slack, trigger Zoom calls and loop in multiple departments. They break because the systems usually don’t connect and lack context. Bad agreement management practices cause a nearly $2 trillion drag on the global economy. The Docusign Advantage We power the entire agreement workflow, from contract generation to signature, obligation management, renewal and beyond. Our newly launched Agreement Desk serves as a hub where AI contract agents handle routine tasks like intake, review, and routing. Instead of agents navigating chaos, they operate on a platform that knows how to route decisions and trigger actions. Agreements flowing through our platform give us deep insight into their structure, context, and history. Our AI engine Iris powers Navigator, which has ingested approximately 150 million customer agreements and extracts key information like obligations, renewal management, and key deadlines. And we power the signature, the most critical moment in any agreement workflow. If you're already using Docusign for signing, adopting our workflow capabilities is a natural extension. The Bottom Line Workflows will become the next major focus in enterprise AI. The question isn't whether your organization is ready for AI. It's whether you have the tools to make your AI deployment extraordinary. Because it doesn't matter how fast your runners are if they can't pass the baton.