Improving Communication Through Data Analysis

New research from Tableau Research, led by Arjun Srinivasan and co-authored by Joanna Purich and Leilani Battle, based on an analysis of over 25,000 dashboards, offers crucial insights for every executive looking to maximize their data investment and move beyond the "#Dashboard Zoo": 🔶 Stop Chasing Complexity: Simple charts (bar, line) dominate for a reason. Clarity and familiarity drive adoption. Insist that your teams prioritize clean, accessible visualizations over bespoke or overly complex designs that confuse users and slow decision-making. 🔷 Elevate the Narrative: Text blocks are the second most common content element. Your data story is as critical as the data itself. Treat commentary, framing, and titles as first-class design elements to ensure strategic context is never missed. 🔶 Define the Archetype: Not all dashboards serve the same purpose. The research identified three main clusters: Analytic, Magazine, and Infographic. Ensure your teams align the dashboard's design archetype with its intended communication goal before development. A misalignment is a communication failure. The key takeaway for leadership: Scaling data impact requires intentional, user-centric design principles. Don't just measure the data—measure the quality of the communication. Read the full findings here: https://lnkd.in/ejN8gXA9

Data and Business teams operate in two different domains that speak different languages. For example, data analysts speak in terms of datasets, keys, joins, columns, filters, and aggregations, while data engineers speak in terms of pipelines, dependencies, and lineage. On the other hand, business domains use metrics, drivers, levers, segments, goals, tactics, strategy, initiatives, and features. Being data-driven or data-informed requires effectively translating between these two domains. This involves cross-pollinating data and business teams to align them with the same end goals, which is easier said than done. However, frameworks and tools can serve as powerful collaborative agents, much like engineering teams and how they interface with the business. One such framework that can help is metric trees, which maps the processes by which the business operates and serves its customers - including the inputs, outputs, relationships, entities involved, and their states. By building tools that help data and business teams collaboratively design and operate on metric trees, the translation friction can be significantly reduced. Or, even practically eliminated depending on the sophistication of the tooling. This incentivizes more people to participate leading to dramatic improvements in the quality of conversations and decision-making.

Ever watched eyes glaze over during your data presentation? All that beautiful math—missed the real question: "What should we do next?" I learned the hard way: translating analysis into action is everything. Curious how? Read on. I used to spend most of my meeting time explaining statistical methods, only to realize my audience just wanted a clear recommendation. Once, after 45 minutes on hierarchical models, a scientist asked, "So, should we move this compound forward or not?" I hadn't even calculated that probability. Lab scientists are experts in their own right—they need actionable insights, not a stats seminar. Now, I always start with the decision and probability, then offer details if asked. Trust and engagement have skyrocketed. Your communication budget is finite; spend it on what matters. Lead with the decision, not the methods. Use BLUF (Bottom-Line Up-Front): start with your recommendation and the probability behind it. Lab scientists operate in different modes: decision, learning, or validation. Tailor your approach to their needs—don't default to teaching when they're in decision mode. Translate your analysis through three layers: statistical reality (for you), scientific meaning (the bridge), and decision layer (for them). Only collapse to a single probability when it's time to make a decision. Build trust by being clear and actionable, not by over-explaining. Keep technical details in an appendix—share them only if asked. Anticipate the questions your collaborators always ask. Proactively address their concerns to build credibility and save time. If you've ever struggled to get your analysis heard, check out my latest post for practical frameworks and real-world examples. Would love your thoughts, likes, or shares! Full post here: https://lnkd.in/ex6-q4Hd What strategies have helped you bridge the gap between data and decision-making in your collaborations? #biotech #datascience #communication #leadership #decisionmaking

One of the first lessons I learned as a psychology researcher about humans is that we rarely think in terms of probabilities. Humans are inclined to think in patterns, stories, emotions, and lived experiences. So why do we communicate data as probabilities? In my years working as a data scientist, I didn’t see the best insights fail because the analysis was wrong; I saw them fail because the communication behind them didn’t match how people process information. So what does work? 👉 Translating probability into meaning 👉 Grounding insight in familiar patterns 👉 Pairing the data with the human 👉 Telling the story, the brain is built to hear When companies align with how people communicate and process information, even the most complex data becomes accessible, interesting, and actionable. How do you translate complex outputs into something people can grasp quickly? #DataCommunication #DataLiteracy #Analytics #Leadership

Analytics teams spend weeks perfecting their reports and dashboards only to hear: “This is interesting, but what should we actually do?” Recently, a marketing professor DM’ed me about his students struggling with data storytelling. His marketing research class was comfortable with the reporting aspects. But when asked to offer a clear point of view or insight, they froze. Some worried it might come across as manipulating the data if they offered interpretations or recommendations. This hesitation isn’t limited to these students. Many data professionals feel uncomfortable pushing beyond the “what.” Here’s why: 👉 Fear of being wrong publicly, especially when data involves uncertainty 👉 Desire to appear objective and “let the numbers speak for themselves” 👉 Lack of business context or confidence in their domain knowledge 👉 Positioning as a support function rather than a strategic partner 👉 Not enough time to dig deeper 👉 Strong technical skills but underdeveloped communication skills As a result, analytics often stops before the diagnosis—just listing symptoms without explaining the cause, let alone the cure. We stop at reporting what happened: “Revenue dropped 18%.” 📉 And we hesitate to explain why it happened or what to do next. What we should say: “Revenue dropped 18% because our top customer segment shifted to a competitor with faster delivery options. We should pilot same-day shipping in three test markets.” Ironically, what stakeholders need most—interpretation and direction—is what analysts often avoid. And yet, we don't go to doctors just to confirm we're in pain. We go to understand the cause and find a cure. That’s where data storytelling comes in as it moves us from: ✅ 𝐖𝐡𝐚𝐭 = Symptoms (the metrics and trends) ✅ 𝐒𝐨 𝐖𝐡𝐚𝐭 = Diagnosis (why it’s happening) ✅ 𝐍𝐨𝐰 𝐖𝐡𝐚𝐭 = Treatment (what to do next) If you want your work to drive action, you can’t stop at symptoms. You need to offer meaning and a path forward. What’s one technique that’s helped your team move from reporting to storytelling and action? 🔽 🔽 🔽 🔽 🔽 Craving more of my data storytelling, analytics, AI, and data culture content? Sign up for my newsletter today: https://lnkd.in/gRNMYJQ7 Check out my brand-new data storytelling masterclass: https://lnkd.in/gy5Mr5ky Need a virtual or onsite data storytelling workshop? Let's talk. https://lnkd.in/gNpR9g_K

Over the last few years, I’ve been consulting with several clients on analytics and data-driven decision-making. What I’ve consistently observed is this: ✅ Teams put enormous effort into gathering, cleaning, and analyzing data. ✅ The findings are often powerful — they can unlock efficiency, reduce costs, improve customer experience. ❌ But when it comes to communicating those findings, the message often falls flat. That’s where data storytelling makes all the difference. In one recent engagement, the analysis showed clear gaps in payment collections. The raw numbers didn’t move anyone. But when we reframed it as a story — “For every 10 invoices raised, 3 are delayed, which means the equivalent of one full month of working capital is stuck” — suddenly the business heads sat up, understood the urgency, and acted on it. 👉 That’s the power of insight communication. It’s not just what the data says — it’s how you frame it so decision-makers can connect, remember, and act. My learning: Insights ≠ Impact. Impact happens when insights are translated into stories that resonate with business priorities.

Working smarter in senior living starts with data. A few years ago, one of our partner communities at Eldermark was seeing low satisfaction scores. At first glance, it looked like a staffing issue. But the data showed something else: -Care delivery was uneven across shifts -Staff were spending too much time on non-clinical tasks -Communication breakdowns were eroding trust Once we understood the “why” we made focused adjustments: -Clarified roles and reduced task overlap -Rebalanced schedules to match peak needs -Added simple touchpoints between caregivers and families The results? No overnight miracles. But over time, satisfaction rose, staff turnover dropped, and families reported more peace of mind. I’ve seen this play out over and over again. When you listen to your data, it starts to feel like you’re listening more closely to your people. How important is data in your decision-making process?

🎯 How to Approach Any Data Analyst Project & Maximize Learning Whether you're tackling your first data analyst project or your tenth, the real value isn’t just in getting the answer it’s in how you approach the problem and extract insights along the way. Here’s how I break down any project to maximize learning and deliver impact: 1️⃣ Understand the Problem Before Touching the Data Before diving into SQL, Python, or dashboards, take a step back: ✅ What’s the business question you’re trying to answer? ✅ Who will use this data, and what decisions depend on it? ✅ What are the key metrics or success criteria? 💡 Pro Tip: If you can't explain the problem in one sentence, you don’t understand it well enough. 2️⃣ Explore & Clean the Data (Don’t Skip This!) Most real-world data is messy. Spend time: ✔ Checking for missing, duplicate, or inconsistent values ✔ Understanding data types & distributions ✔ Identifying outliers that might skew results 📊 Learning Boost: Try different approaches (e.g., handling missing values via imputation vs. deletion) and compare how they impact the final analysis. 3️⃣ Analyze with a Hypothesis-Driven Approach Instead of randomly looking for trends, form hypotheses: ❓ Does A cause B, or are they just correlated? ❓ Which segments of users are behaving differently? ❓ What external factors could influence this trend? 🔍 Learning Boost: Every project should refine your ability to think critically and spot misleading conclusions. 4️⃣ Communicate Insights, Not Just Numbers Great analysts don’t just present numbers—they tell a story with data: 📌 Start with the key insight, not just the method 📌 Use visuals to simplify complex trends 📌 Tailor insights to your audience (executives, product teams, etc.) 🚀 Learning Boost: Challenge yourself to explain your findings in one sentence to a non-technical person. If you can’t, refine your messaging. 5️⃣ Reflect & Document Learnings Every project is an opportunity to improve: ✅ What assumptions did you make that turned out wrong? ✅ What techniques or tools would have made the process easier? ✅ What would you do differently next time? 📝 Learning Boost: Keep a project journal or start a blog sharing your key takeaways - it’ll reinforce your learning and build your personal brand. Final Thought Every data project is more than just a dataset it’s a chance to develop business acumen, problem-solving skills, and storytelling abilities. The best analysts aren’t those who know the most tools but those who think critically and communicate insights effectively. How do you approach your data projects? Would love to hear your strategies! 👇🔥 #DataAnalytics #SQL #Python # #CareerGrowth #DataScience #Jobs #PythonFunctions #DataAnalyst #CareerGrowth #InterviewTips #DataAnalysis #JobSearch #TechCareers #DataVisualization #projects

𝗕𝗿𝗶𝗱𝗴𝗶𝗻𝗴 𝗕𝘂𝘀𝗶𝗻𝗲𝘀𝘀 & 𝗗𝗮𝘁𝗮: 𝗖𝗼𝗺𝗺𝘂𝗻𝗶𝗰𝗮𝘁𝗶𝗻𝗴 𝗜𝗻𝘀𝗶𝗴𝗵𝘁𝘀 𝘁𝗼 𝗡𝗼𝗻-𝗧𝗲𝗰𝗵𝗻𝗶𝗰𝗮𝗹 𝗦𝘁𝗮𝗸𝗲𝗵𝗼𝗹𝗱𝗲𝗿𝘀 Data analysts often face a big challenge not just analyzing data, but explaining it in a way that makes sense to business team. A great analysis is useless if decision-makers don’t understand it! Here are some ways analysts can communicate better with non-technical stakeholders: ↳ 𝗧𝗲𝗹𝗹 𝗮 𝗦𝘁𝗼𝗿𝘆, 𝗡𝗼𝘁 𝗝𝘂𝘀𝘁 𝗡𝘂𝗺𝗯𝗲𝗿𝘀:– Instead of sharing raw data, focus on the key takeaway. What does the data mean for the business? ↳ 𝗔𝘃𝗼𝗶𝗱 𝗝𝗮𝗿𝗴𝗼𝗻:– Terms like "p-value," "ETL," or "normalization" might not be familiar to everyone. Use simple language that connects with your audience. ↳ 𝗨𝘀𝗲 𝗖𝗹𝗲𝗮𝗿 𝗩𝗶𝘀𝘂𝗮𝗹𝘀:– A well-designed chart is more powerful than a table full of numbers. Choose the right visual to highlight the key insight. ↳ 𝗨𝗻𝗱𝗲𝗿𝘀𝘁𝗮𝗻𝗱 𝗧𝗵𝗲𝗶𝗿 𝗡𝗲𝗲𝗱𝘀:– Before presenting data, ask stakeholders what decisions they need to make. This helps you focus on relevant insights. ↳ 𝗘𝗻𝗰𝗼𝘂𝗿𝗮𝗴𝗲 𝗤𝘂𝗲𝘀𝘁𝗶𝗼𝗻𝘀:– A two-way conversation ensures stakeholders fully understand the data and feel confident using it. Great analysts don’t just crunch numbers, they bridge the gap between data and decision-making. What strategies have helped you communicate better with non-technical teams? #dataanalytics

There's this one underrated skill I figured every data professional should have. Stakeholder Engagement. The clients or business partners reach out to you with their concerns, and you give them data-backed solutions. Great! But do they actually use all of it? Maybe because they didn't fully understand your solution.   Maybe 2 out of 20 graphs would suffice for their requirements.   Or maybe you gave an orange when they asked for an apple. Anyhow, an unhappy user is equivalent to poor value and grading of your work. Here’s how we can do better: 1. Keep them in the loop from day 0 - even while understanding the requirements. Ask a lot of questions and make them feel heard. Trust starts with you stepping over to their side of the boat. 2. Explain the data layer - they are the business experts, and you are the data expert. Explaining what each field is and how it's retrieved helps users draft better and more realistic requirements. 3. Educate - explaining how you built that KPI really boosts clarity. Explain the logic, show them the process, and ask for feedback on how we could make this better together. 4. Connect beyond meetings - recurring weekly updates might feel enough, but constant communication - be it a call, quick text, or an ad-hoc in-person conversation - results in better alignment. This ensures that the final solution you deliver is not a surprise handover; instead, they'll feel it's their own project - co-built. Happy insights, y'all! #dataanalytics #datascience #stakeholders