Elder Research

Our CEO, Gerhard Pilcher, shares the 3d AI trend he’s seeing: the real-world impact of generative AI. While it’s a powerful tool for tasks like generating text or images, it’s important to understand its limitations. Generative AI can help generate ideas or content, but it’s not a replacement for precise, business-critical decisions. Human oversight is still necessary to guide these outputs toward practical application. 🧑💻 As businesses look to scale generative AI, they must understand the full creative process and the workflows required to deliver value. ☑️ Without the right guardrails, generative AI could create more work than it solves. 😶🌫️ Watch the video and share your thoughts in the comments! And stay tuned for the final 𝘈𝘯𝘢𝘭𝘺𝘵𝘪𝘤𝘴 𝘈𝘩𝘢 𝘔𝘰𝘮𝘦𝘯𝘵𝘴 episode next Tuesday. #AnalyticsAhaMoments #DataScience #DataAnalytics #GenerativeAI #ArtificialIntelligence

Are you at an organization with 𝗲𝗺𝗲𝗿𝗴𝗶𝗻𝗴 analytics or 𝗺𝗮𝘁𝘂𝗿𝗲 analytics? Data scientists Paige Spell and Shaylee Davis highlight the differences ⬇️ In their webinar, they share what emerging and mature analytics cultures look like and practical steps to drive analytics forward at any stage of analytical maturity: https://lnkd.in/ejUPkVzt 𝗧𝗵𝗲 𝗦𝘁𝗮𝗴𝗲𝘀 𝗼𝗳 𝗔𝗻𝗮𝗹𝘆𝘁𝗶𝗰𝗮𝗹 𝗠𝗮𝘁𝘂𝗿𝗶𝘁𝘆 🔍 𝗗𝗲𝘀𝗰𝗿𝗶𝗽𝘁𝗶𝘃𝗲 𝗔𝗻𝗮𝗹𝘆𝘁𝗶𝗰𝘀 – Shows what happened through reports and dashboards. Data is often separate across teams, with most reporting done manually. ☑️ 𝗗𝗶𝗮𝗴𝗻𝗼𝘀𝘁𝗶𝗰 𝗔𝗻𝗮𝗹𝘆𝘁𝗶𝗰𝘀 – Digs into why it happened, spotting patterns and root causes. Collaboration and data-sharing across teams start to grow. 📈 𝗣𝗿𝗲𝗱𝗶𝗰𝘁𝗶𝘃𝗲 𝗔𝗻𝗮𝗹𝘆𝘁𝗶𝗰𝘀 – Forecasts what’s likely to happen using data and machine learning. Decision-making becomes more proactive, with integrated, accessible data on cloud storage. 🛠️ 𝗣𝗿𝗲𝘀𝗰𝗿𝗶𝗽𝘁𝗶𝘃𝗲 𝗔𝗻𝗮𝗹𝘆𝘁𝗶𝗰𝘀 – Recommends what to do next for the best results. Advanced models and simulations make it easier to act on data in real-time, with workflows and teams working together seamlessly. 𝗗𝗼𝗻’𝘁 𝗳𝗲𝗲𝗹 𝗹𝗶𝗸𝗲 𝘆𝗼𝘂 𝗳𝗶𝘁 𝗶𝗻𝘁𝗼 𝗼𝗻𝗲 𝗼𝗳 𝘁𝗵𝗲𝘀𝗲 𝗯𝘂𝗰𝗸𝗲𝘁𝘀? Maybe your organization has an emerging analytics culture—somewhere between Diagnostic and Predictive or between Predictive and Prescriptive. Whether you’re just starting to incorporate analytics or working with a robust data infrastructure, there’s always room to grow. Interested in boosting your analytics culture? Catch the webinar for more insight. Thank you to The National Consortium for Data Science for hosting! #DataScience #DataAnalytics #DataMaturity #DataWebinar

How is AI reshaping data governance? 🔄 Our CEO, Gerhard Pilcher, shares how AI is changing the way we manage and govern data: “What I’m seeing as a trend is that there are a lot of companies beginning to address—and use AI actually to do it—to make the process of data governance more automated,” says Gerhard. 𝗔 𝗳𝗲𝘄 𝘁𝗮𝗸𝗲𝗮𝘄𝗮𝘆𝘀: ⚬  New guidelines and regulations mean formal AI and data governance are quickly becoming must-haves. ⚬ AI is taking over tagging, cataloging, and tracking data, making it simpler and less manual as data scales up. ⚬ With more data sources, AI can track how data is used and keep companies in compliance with agreements and regulations. Gerhard notes that governance can start strong but quickly fade if it’s not kept current. With AI, organizations can keep governance up-to-date and ready for evolving regulations. Watch the video for more of Gerhard’s thoughts, and share your thoughts in the comments! 💭 Missed any videos in the series? Catch up now, and stay tuned for another episode next Tuesday! #AnalyticsAhaMoments #DataGovernance #DataScience #DataAnalytics #MachineLearning #ArtificialIntelligence

17 reasons data teams should play pickleball … Actually, we really only need 1 😉: It’s a great way to spend time with each other beyond the day-to-day projects. Last Friday our Charlottesville office had an amazing time on the pickleball courts. 🏓 There were some epic matches AND some epic dives along with them (check the photos). 👟 Taking time to get to know each other beyond the work we do matters a lot to our team. So moments like this are worth the investment. What’s one way your team builds community? Happy Friday! #Pickleball #TeamBuilding

LLMs might be the AI heavyweights, but SLMs pack a punch too. 🥊 For government agencies exploring AI, small language models (SLMs) provide an efficient, lower-cost way to dive in without overcommitting resources. In our latest blog, data scientists Harrison Blondeau and John Moroney show how SLMs help government agencies tap into AI—all while staying in control of costs, data, and security.   𝗛𝗲𝗿𝗲 𝗮𝗿𝗲 𝗮 𝗳𝗲𝘄 𝗼𝗳 𝘁𝗵𝗲 𝗯𝗲𝗻𝗲𝗳𝗶𝘁𝘀: ⚙️ 𝗗𝗲𝘀𝗶𝗴𝗻𝗲𝗱 𝗳𝗼𝗿 𝘀𝗽𝗲𝗰𝗶𝗳𝗶𝗰 𝘁𝗮𝘀𝗸𝘀 - SLMs focus on solving specific problems, making them the perfect fit for government projects where precision matters. ⚙️ 𝗠𝗼𝗿𝗲 𝗰𝗼𝗻𝘁𝗿𝗼𝗹 𝗼𝘃𝗲𝗿 𝗱𝗮𝘁𝗮 - SLMs can be deployed on-premises, giving agencies the security they need while advancing their AI initiatives. ⚙️ 𝗖𝗼𝘀𝘁-𝗲𝗳𝗳𝗲𝗰𝘁𝗶𝘃𝗲 - With lower resource requirements, SLMs provide a way to explore AI without straining budgets.   Want to learn more? Check out our blog: https://lnkd.in/gA9w7jqP 𝗣.𝗦. If your team has had experience with SLMs, we’d love to hear about it in the comments! #SmallLanguageModel #SLM #AI #ArtificialIntelligence

Data/MLOps Engineer Joseph Okeno-Storms shares 3 words of wisdom he’s learned in his career: 𝟭. 𝗣𝗲𝗿𝗳𝗲𝗰𝘁𝗶𝗼𝗻 𝗶𝘀 𝘁𝗵𝗲 𝗲𝗻𝗲𝗺𝘆 𝗼𝗳 𝗽𝗿𝗼𝗴𝗿𝗲𝘀𝘀. “It’s okay if your first iterations of a project or task aren’t perfect,” says Joseph. “Progress comes from taking steps forward, learning, and improving along the way. Perfectionism, however, can trap you in the early stages, causing you to stall or even abandon ideas that could lead to success.” 𝟮. 𝗬𝗼𝘂 𝗰𝗮𝗻 𝗮𝗹𝘄𝗮𝘆𝘀 𝗰𝗼𝗻𝘁𝗿𝗶𝗯𝘂𝘁𝗲 𝘁𝗼 𝘁𝗵𝗲 𝘀𝗼𝗹𝘂𝘁𝗶𝗼𝗻 𝘄𝗵𝗲𝗻 𝗳𝗮𝗰𝗲𝗱 𝘄𝗶𝘁𝗵 𝗮 𝗽𝗿𝗼𝗯𝗹𝗲𝗺. “In this role, you’ll often face errors, blockers, and failures. When discussing these with clients, colleagues, or supervisors, always propose a solution or begin that conversation. This shows initiative and commitment,” notes Joseph. “For example, ‘The pipeline failed in production because of this bug, but here’s what we’ve done or can do to fix it.’ Or ‘I anticipate this challenge in the proposal, but here’s how we can address it.’ Even if you’re unsure of the solution, it’s important to share potential paths you’re exploring.” 𝟯. 𝗙𝗼𝗰𝘂𝘀 𝗼𝗻 𝘄𝗵𝗼 𝘆𝗼𝘂 𝘄𝗮𝗻𝘁 𝘁𝗼 𝗯𝗲 𝗮𝘀 𝗮𝗻 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿, 𝗻𝗼𝘁 𝗷𝘂𝘀𝘁 𝘄𝗵𝗮𝘁 𝘆𝗼𝘂 𝘄𝗮𝗻𝘁 𝘁𝗼 𝗱𝗼. “Inspired by Simon Sinek’s Golden Circle model, think about the values that guide your actions as an engineer,” says Joseph. “The ‘what’ you do should be the outermost layer, driven by the ‘who’ you choose to be. When you encounter challenges or unexpected roadblocks, focusing on your commitments and values helps maintain direction and congruence in your role. This mindset keeps you grounded, even when the ‘what’ changes or faces difficulties.” What thoughts do you have on Joseph’s words of wisdom? Let us know in the comments! #DataEngineering #MLOpsEngineering #DataScience #DataAnalytics

Ever deployed a system thinking it’s the easy button only to find it didn’t change a thing? Gerhard Pilcher shares why that happens: Scaling AI isn’t just about plugging in tech—it’s about business transformation. 🔄 This week in 𝘈𝘯𝘢𝘭𝘺𝘵𝘪𝘤𝘴 𝘈𝘩𝘢 𝘔𝘰𝘮𝘦𝘯𝘵𝘴, Gerhard dives into the first of four AI trends he’s seeing: scaling AI to actually deliver business value. He points out that many organizations fail to realize the full potential of AI because they don’t invest enough time in examining their business processes. 🔬 Adopting AI effectively requires a detailed look at your processes to understand where AI can help—and where human expertise is still essential. Without this, you risk AI being just another tool that doesn’t move the needle. But by aligning AI with the right processes, businesses can unlock significant value. 📊 Watch the full video to hear Gerhard explain more, and stay tuned for next Tuesday’s video. #AnalyticsAhaMoments #DataScience #DataAnalytics #MachineLearning #ArtificialIntelligence #AI

𝟴 𝗼𝗳 𝘁𝗵𝗲 𝘁𝗼𝗽 𝗮𝗻𝗮𝗹𝘆𝘁𝗶𝗰𝘀 𝗽𝗿𝗼𝗷𝗲𝗰𝘁 𝗱𝗲𝗿𝗮𝗶𝗹𝗲𝗿𝘀 𝗔𝗡𝗗 𝘄𝗮𝘆𝘀 𝘁𝗼 𝗮𝘃𝗼𝗶𝗱 𝘁𝗵𝗲𝗺: ❌ 𝗗𝗲𝗿𝗮𝗶𝗹𝗲𝗿 𝟭: 𝗟𝗮𝗰𝗸 𝗼𝗳 𝗖𝗹𝗲𝗮𝗿 𝗦𝘁𝗿𝗮𝘁𝗲𝗴𝘆 & 𝗢𝗯𝗷𝗲𝗰𝘁𝗶𝘃𝗲𝘀 ✅ 𝗦𝗼𝗹𝘂𝘁𝗶𝗼𝗻: Ensure that every analytics project is driven by well-defined business objectives and a clear strategy aligned with organizational goals. ❌ 𝗗𝗲𝗿𝗮𝗶𝗹𝗲𝗿 𝟮: 𝗢𝘃𝗲𝗿𝗲𝗺𝗽𝗵𝗮𝘀𝗶𝘀 𝗼𝗻 𝗧𝗼𝗼𝗹𝘀 ✅ 𝗦𝗼𝗹𝘂𝘁𝗶𝗼𝗻: Choose tools that are best suited to solve specific business problems rather than selecting a tool and then searching for a problem to apply it to. ❌ 𝗗𝗲𝗿𝗮𝗶𝗹𝗲𝗿 𝟯: 𝗙𝗮𝗶𝗹𝘂𝗿𝗲 𝘁𝗼 𝗜𝗻𝘁𝗲𝗴𝗿𝗮𝘁𝗲 𝗔𝗻𝗮𝗹𝘆𝘁𝗶𝗰𝘀 𝘄𝗶𝘁𝗵 𝗕𝘂𝘀𝗶𝗻𝗲𝘀𝘀 ✅ 𝗦𝗼𝗹𝘂𝘁𝗶𝗼𝗻: Foster collaboration between analytics teams and business units to ensure that models and insights are aligned with real business needs and can be effectively implemented. ❌ 𝗗𝗲𝗿𝗮𝗶𝗹𝗲𝗿 𝟰: 𝗜𝗻𝗮𝗱𝗲𝗾𝘂𝗮𝘁𝗲 𝗢𝗿𝗴𝗮𝗻𝗶𝘇𝗮𝘁𝗶𝗼𝗻𝗮𝗹 𝗖𝗵𝗮𝗻𝗴𝗲 𝗠𝗮𝗻𝗮𝗴𝗲𝗺𝗲𝗻𝘁 ✅ 𝗦𝗼𝗹𝘂𝘁𝗶𝗼𝗻: Communicate the value of analytics initiatives across the organization and providing the necessary training to ease the transition. ❌ 𝗗𝗲𝗿𝗮𝗶𝗹𝗲𝗿 𝟱: 𝗣𝗼𝗼𝗿 𝗗𝗮𝘁𝗮 𝗤𝘂𝗮𝗹𝗶𝘁𝘆 & 𝗠𝗮𝗻𝗮𝗴𝗲𝗺𝗲𝗻𝘁 ✅ 𝗦𝗼𝗹𝘂𝘁𝗶𝗼𝗻: Put data governance practices in place to ensure data is accurate, relevant, and ready for analysis. (But don’t focus on the elusive “perfect data.” Data doesn’t have to be perfect to deliver insights.) ❌ 𝗗𝗲𝗿𝗮𝗶𝗹𝗲𝗿 𝟲: 𝗟𝗮𝗰𝗸 𝗼𝗳 𝗕𝗿𝗼𝗮𝗱 𝗔𝗻𝗮𝗹𝘆𝘁𝗶𝗰𝘀 𝗘𝗱𝘂𝗰𝗮𝘁𝗶𝗼𝗻 ✅ 𝗦𝗼𝗹𝘂𝘁𝗶𝗼𝗻: Provide ongoing training to build a culture where everyone understands the value of data, trusts the process, and can contribute to analytics initiatives. ❌ 𝗗𝗲𝗿𝗮𝗶𝗹𝗲𝗿 𝟳: 𝗧𝗲𝗰𝗵𝗻𝗶𝗰𝗮𝗹 𝗣𝗶𝘁𝗳𝗮𝗹𝗹𝘀 ✅ 𝗦𝗼𝗹𝘂𝘁𝗶𝗼𝗻: Involve domain experts throughout the model development process to ensure that models are built on sound assumptions and avoid common technical errors. ❌ 𝗗𝗲𝗿𝗮𝗶𝗹𝗲𝗿 𝟴: 𝗜𝗻𝘀𝘂𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝗥𝗶𝘀𝗸 𝗠𝗮𝗻𝗮𝗴𝗲𝗺𝗲𝗻𝘁 ✅ 𝗦𝗼𝗹𝘂𝘁𝗶𝗼𝗻: Approach projects with flexibility, buffer deadlines, and keep stakeholders informed to manage their expectations and maintain support. Remember, the key to a successful analytics project isn’t just in the data or the tools—it's in how you align your strategy, team, and organization to truly unlock the value. Ready to get your next project off the ground? Our team would love to chat about ways you can get the most value. #DataAnalytics #DataScience #DataProjects

What does “maturity” really mean when it comes to data analytics? And how can it support or hinder an organization’s efforts? 🤔 Tomorrow, October 17, at 12 pm (ET), data scientists Paige Spell and Shaylee Davis are digging into that very topic. 𝗧𝗵𝗲 𝘄𝗲𝗯𝗶𝗻𝗮𝗿 𝗲𝘅𝗽𝗹𝗼𝗿𝗲𝘀: 💡 How analytics shift across organizations with different levels of maturity 💡Real-world challenges (and opportunities!) when working with data at these varying stages 💡Practical tips for making a bigger impact—no matter where your organization is on its data journey The DataBytes webinar series is all about sharing fresh ideas in data science—whether you’re sharpening your skills or just looking for new perspectives. Big thanks to The National Consortium for Data Science for hosting. We hope to see you there! 𝗥𝗲𝗴𝗶𝘀𝘁𝗲𝗿 𝗵𝗲𝗿𝗲: https://lnkd.in/gxsHmeyN #DataScience #DataAnalytics #DataMaturity #DataWebinar

Is your AI strategy keeping up with the rapid changes in the industry? 💡 In the latest episode of 𝘈𝘯𝘢𝘭𝘺𝘵𝘪𝘤𝘴 𝘈𝘩𝘢 𝘔𝘰𝘮𝘦𝘯𝘵𝘴, Gerhard Pilcher dives into the three foundational types of AI—giving us a better framework to understand the fast-paced developments happening today. 𝗚𝗲𝗿𝗵𝗮𝗿𝗱 𝗯𝗿𝗲𝗮𝗸𝘀 𝗔𝗜 𝗶𝗻𝘁𝗼 𝘁𝗵𝗿𝗲𝗲 𝗸𝗲𝘆 𝗰𝗮𝘁𝗲𝗴𝗼𝗿𝗶𝗲𝘀: 1️⃣ 𝗔𝘂𝘁𝗼𝗺𝗮𝘁𝗶𝗼𝗻: AI has been automating business processes for decades. Think back to the Apollo space missions. Without the help of computers automating complex calculations, astronauts couldn’t have achieved their mission. 2️⃣ 𝗣𝗿𝗲𝗱𝗶𝗰𝘁𝗶𝘃𝗲: This is about forecasting the future. It helps us predict what will happen next, whether it’s classifying data, making forecasts, or predicting consumer behavior. 3️⃣ 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝘃𝗲: Using AI to generate new content, like text and images, is the hottest topic right now. But Gerhard reminds us that AI is more than the latest trend. It’s important to look at the big picture, not just focus on what’s making headlines. In the coming videos, Gerhard will dive into four trends he’s seeing in the AI space. These trends are shaping how organizations use AI to innovate, automate, and stay competitive. 𝗣.𝗦. If you missed earlier episodes, check out the hashtag #AnalyticsAhaMoments to catch up. #DataScience #DataAnalytics #MachineLearning #ArtificialIntelligence #AI