Olga Berezovsky

We're facing more and more questions about the use of synthetic data in market research. I'd recommend this blog post, which really resonated with me. So much of the conversation is "how closely can this replicate human data?" But there are so many other considerations researchers should be thinking about. https://lnkd.in/eSQfSTPj

31

3 Comments

I Made It Up new feature drop: Analysis. This is a new tool that processes synthetic interview transcripts for quantitive (numbers) and qualitative (stories) analysis: • Quantitative - means + variance + statistical comparisons across demographic groups + outlier detection • Qualitative - theme extraction + clustering into higher-level patterns + supporting quotes Insights before you're back with coffee. If you don't know: I Made It Up is a synthetic research toolkit that plugs straight into AI chat like ChatGPT and Claude. (If that's jargon, 'synthetic research' is methodology that uses AI to simulate digital personas as avatars to explore how people might think. Made up? Yes. Useful? We think so.) Big words. What's it like in practice? Say you interview 36 personas about a product idea. Your AI then passes the transcripts to the Analysis tool for processing. You + AI then get results back that enable further analysis + thinking. Quantitative tells you: "Gen Z rated this 2.1 out of 5, while Millennials gave it 3.8 - that's statistically significant with a large effect size". Qualitative tells you: "Three themes emerged - price sensitivity, trust concerns, and feature confusion - here are the quotes that back each one up". Super early stage. Interviews are limited to six questions for six personas. This goes up soon and when it does, the Analysis tool becomes indispensable. Want to know more? If you're already using it, just ask ChatGPT or Claude what you can do with I Made It Up's Analysis and Retrieve tools. If you're not yet, check Getting Started on website linked in comments. Available on Pro accounts and above. Lots under-the-hood to unpack. Blog soon. Questions? Hit me up. Would love to talk and if you're local, can drop in to your office for a chat. 

7

2 Comments

You should be rewriting your prompts We talk about overfitting models, but never overfitting prompts to models. Models have specific formatting preferences - OpenAI favors markdown while Anthropic uses XML - and might weight different parts of the prompt differently. Newer models won't necessarily perform better out-of-the-box until you do this testing and optimization. https://lnkd.in/eNq6DqXJ

2

One more for today - just because it’s too insightful and valuable to miss for anyone wanting to step up their agentic analytics game. Thanks to @Pankaj this repository teaches you everything from A-Z on ontologies, Enterprise AI Architectures, approaches and concepts - from high level to deep down the rabbit hole. https://lnkd.in/d3uxNfcR

4

Orthogonal ETL pipelines triangulate to truth. Most data quality frameworks are outdated. Coverage. Timeliness. Accuracy. Schema compliance. These metrics are artifacts of the 2010s data stack. In the age of agentic ETL, they are increasingly cheap. Our agents can scrape more sources, normalize schemas, and expand coverage at scale. But none of that guarantees truth. The most reliable datasets emerge from orthogonal pipelines reconstructing the same dataset independently. This principle already exists in modern statistics and AI: ensembles, mixture-of-experts, multimodal models, and orthogonal estimators. The insight is the same. Single pipelines produce data. Orthogonal systems converge on truth.

5

1 Comment

Are Your Data Leaders Thinking… or Just Connecting? Here’s an uncomfortable truth. Many data teams spend most of their time not analyzing… not predicting… not innovating… But connecting systems. Fixing flows. Maintaining pipelines. That’s not a talent problem. That’s an architecture problem. When ingestion, governance, and analytics live in different tools, Your smartest engineers end up stitching instead of strategizing. That ratio is broken. Modern platforms like Microsoft Fabric shift the balance: • Built-in ingestion • Built-in governance • Built-in analytics Less system stitching. More business thinking. Because scaling insight shouldn’t require scaling headcount. 📩 datafabric@quadranttechnologies.com Write to us for a quick walkthrough or POC discussion. And if you’re heading to FabCon Atlanta (March 18–20), let’s connect in person; see you there. Microsoft | Quadrant Technologies | Ram Paluri, MBA | Vamshi Reddy | Bhaskar Gangipamula | James Kass | Prakash Nagarajan | Lavina DSilva | Mithun P N | Dr. Madhavi Gundavajyala | Siva Kanuru | Sivani Pamidi | Varsha Panguluri | Ajay kumar Erukulla | #Microsoft #DataStrategy #MicrosoftFabric #CIO #CDO #ModernData #FabCon

91

2 Comments

🔍 The power of Causal Inference is the real game-changer. We're drowning in data. We have dashboards, KPIs, and reports coming out of our ears. We can tell you what happened—sales dropped, user engagement spiked, churn increased. But can we truly answer why? This is where the statistical superpower of Causal Inference is taking over the conversation. It's the difference between seeing a correlation and proving a cause. Let me break it down without the complex math: ➡️ Correlation vs. Causation (The Classic): We all know the meme: "Ice cream sales correlate with drowning deaths." They happen at the same time, but one doesn't cause the other. The hidden cause? Summer heat. ➡️ Causal Inference in Action: This is the framework for finding that "summer heat" in your business. It moves beyond "what" to answer critical questions like: "Did our new marketing campaign cause the increase in sales, or was it just the holiday season?" "Did that new feature update cause a reduction in customer churn?" "Will offering this training cause an improvement in employee performance?" Why is this trending NOW? Because in a world of A/B testing and complex customer journeys, guessing is a luxury we can't afford. Businesses that understand the why behind their data make smarter, more profitable decisions and build products people actually want. The Tool You'll Hear About: Propensity Score Matching (Don't be scared by the name!). Imagine you want to test a new drug but can't do a random trial. You could find people who didn't take the drug but are statistically identical in every other way (age, health, lifestyle) to those who did. That's PSM in a nutshell—it helps simulate a randomized experiment from observational data. Your Takeaway: The next time you're looking at a data insight, push yourself and your team. Don't just accept the "what." Ask: "How can we test if this is truly the cause?" Embracing causal thinking is the next frontier in data literacy. It’s what separates data-informed companies from data-driven ones. What's your experience with moving from correlation to causation? Let's discuss in the comments! 👇 #Statistics #DataScience #CausalInference #DataAnalytics #BusinessIntelligence #AI #MachineLearning #DataDriven #KPIs

1