Ted Lorenzen

Today I want to ask Tableau about divergence between web edit and desktop. There are so many nice features in web edit I wish were in desktop as well. E.g. - Shape size percentage guage - slide shape size with precision. - Hyper links and spell check in text boxes. - Layout hierarchy for dashboards - move sheets up or down into the desired container with ease. Yet web edit doesn’t have the full formatting functionality that desktop does, to make the full switch. Switching between desktop and web is frustrating to get the full range of functionality. Tableau driven by agent force is a brand new UI coming next year. This is an opportunity to converge the product interfaces. For a better seamless workflow between desktop and Cloud. My question for Ryan Aytay would be is there plans to converge or continue to diverge web edit and desktop? #tableau #webedit #desktop

8

2 Comments

What's in a name? Falko Schulz creates a SAS Visual Analytics infographic on the popular names in the US collected by the Social Security Administration from 1920 to 2022. There's even a link to the interactive report if you want to delve into the data. #infographic #dataviz #SAS #SASviya #SASusers #SAStip #data https://lnkd.in/guhBVUPU

22

1 Comment

Tableau + map layers is spatial analytics at its best. Here are some quick insights about crimes in Chicago. Over two 1-hour classes in @Next-Level Tableau, we used data from the Chicago data portal about crimes in 2024. Overall, you can see a hot spot of crime on the east of Chicago. The highest crime area falls between a 1.25 mile buffer from two police stations. It might be a coincidence, but it's a signal that more analysis should be done. Despite there being more crime in the east, homicides trend towards the west and southeast. Almost all prostitution crimes happen in the SW corner of district 11, which is near one of the districts with more murders. This was created with 5 map layers: 1. Crime locations (density heat map) 2. Police station locations in each district (stars) 3. A buffer around each station colored by the number of crimes in that buffer. 4. The borders of each police district are colored by the number of crimes in that district 5. The streets across the city, grouped by police district. Because they were separate data sources, I used a parameter to filter across the data sources. Have you ever done that? It's amazing what you can do in a couple hours when you combine open data with Tableau. Check it out → https://lnkd.in/eqXtCVgF __________ Hi, I'm Andy Kriebel Click my name + follow + 🔔 I help people become Tableau experts 10x faster. Sound good? Book an appointment and let's see how I can help. Like this post? Like 👍 | Comment ✍ | Repost ♻️

98

20 Comments

🔍 Just read a fascinating paper on the future of data ecosystem business models in the emergent "Artificial Intelligence of Thinks" (AIoT) space! Very useful systematic research that yields some key insights on how companies are transforming data into value in this developing ecosystem. https://lnkd.in/ert4nU7X 🔄 FOUR DISTINCT BUSINESS MODEL ARCHETYPES IN THE AGE OF AI ARISING 🤝 Adaptive Partnership - High control + High customization - Focuses on developing close, collaborative relationships with customers - Offers deep insights and personalized solutions - Best suited for strategic, long-term business relationships 👥 User-Centric - Low control + High customization - Prioritizes customer experience and flexibility - Enables broad market reach and scalability - Emphasizes user autonomy in configuring solutions ⚡ Standardized Efficiency - High control + Low customization - Focus on operational excellence and cost reduction - Streamlined processes and standardized offerings - Ideal for high-volume, routine operations [3] 🔄 Flexible Transaction - Low control + Low customization - Transaction-focused with minimal ongoing engagement - Emphasizes automation and self-service - Suitable for commoditized offerings 💎 AIOT BUSINESS MODEL VALUE CAPTURE STRATEGIES: 📊 Responsive Monitoring: Goes beyond traditional monitoring by enabling instant actions based on real-time data, including automatic corrective functions and predictive maintenance to prevent failures 🤖 Object Self-Service: Drives objects toward independent functionality with less human intervention, incorporating self-adjustment, replenishment, and self-operation capabilities 💡 Insights as a Service: A data-driven value proposition that leverages AIoT to collect, analyze and deliver real-time and collective insights that give meaning to massive amounts of data, enabling customers to make informed decisions   ✨ Enhanced User Experience: Focuses on improving customer interaction through digital add-ons and customizable interfaces and content that adapt to individual users' preferences and behaviors over time 💡💡 The research shows successful AIoT implementations balance control and customization to create sustainable competitive advantages. 💡💡 🤔 Where do you see your organization fitting in this framework? #AIoT #DigitalTransformation #Innovation #BusinessStrategy

14

1 Comment

It can be useful to do a simple numerical analysis with qualitative data — particularly when you have several groups or a large number of IDIs. AI has been limited in its ability to create graphics from interview transcripts. Now with Claude 3.5 Sonnet we can create interactive graphs and export them for clients or colleagues. I’m going to be adding more videos to my new YouTube channel, so it would be great if you could subscribe. https://lnkd.in/eUeysha5

1

We often say that we slice data to the point of transparency, and so it is with our analysis of all non-Early Decision application outcomes. We link you to a viz on our Tableau Public page. Spoiler: the regular decision landscape is treacherous. https://ow.ly/pUgp50TsJK7

1

OpenAI recently introduced two new models, OpenAI o1-preview and OpenAI o1-mini, enhancing ChatGPT with 'reasoning' abilities to improve its problem-solving capabilities. These models are part of the “Strawberry” series, aligning OpenAI's focus towards achieving #AGI. Some key points to note: 1️⃣ OpenAI claims that the new models excel at coding and assisting STEM professionals in problem-solving, but they are pricier and slower compared to GPT-4o. 2️⃣ While GPT-4o can handle various input types like text, images, and audio simultaneously, the new models are limited to text inputs only. 3️⃣ The o1-preview model represents a strategic shift for OpenAI, transitioning from a generic approach to tailored solutions for specific challenges. 📍For more details and insights on the capabilities of these new models, check out OpenAI's announcement: [Link to OpenAI o1-preview](https://lnkd.in/gNydZqNT) 👉🏼 Stay tuned for my upcoming post, where I will delve into the practical applications and use cases of these models, highlighting their ability to provide responses supported by critical reasoning. #openai #chatgpt #llm #o1

6

2 Comments

One of my favorite ML YouTubers is Joshua Starmer PhD. One thing he commented offhanded (or not,maybe he rehearsed that sentence 25 weeks, who knows), and I am totally paraphrasing here is that Stats is not just Data Science, and Data Science is not just Stats. Remember the formulas I was playing around with? The simple, updated formulas on Standard Deviation and Mean. Today, I put them to the test with simulated datasets mimicking real-world data additions. and the results, wow. This not only saves computing time, which adds up quick, but it also saves computing power. #DataScience #MachineLearning #DataAnalytics #IDoNotKnowWhatIAmDoing

14

What makes a tool great? There are a multitude of sites (and scores of books) on the subject, but if you want a quick and easy example, take a look at FRED, a 33-year-old data tool from the Federal Reserve of St. Louis. It’s popular across the political spectrum and the professional as well, beating out even expensive state-of-the-art alternatives. And while changes are coming in 2025, it’s still in heavy use both at home and abroad, with some 825,000 data series. Give this article 👇🏽 from Friday by Jeanna Smialek a read for more, and take a look at the FRED site itself. It’s clear, easy to read and use, and fast. It’s a reminder that great digital solutions can solve real problems, offer user-friendly interactions, facilitate seamless adoption, and enable exploration. Whether you’re migrating platforms or implementing AI automation, the best technology is not just powerful—it’s also highly usable. https://lnkd.in/gE_9tpdA #DataTools #DigitalInnovation #DataUsability #TechSimplicity #DigitalSolutions

7

2 Comments

When quantifying people data, things can get messy pretty quickly!! Zooming out is important. Data collected isn't as straightforward as just reading flat numbers. First up to bat in our consideration battery is *bias*. Arguably, no matter how carefully we measure or formulate research questions, there will always be unaccounted variables that can skew the data. These lurking biases often play a subtle but powerful role, influencing outcomes in ways we might not immediately recognize. For example, consider how the "Time is Money" study by Empower reported what individuals (by generation) claimed their time was worth in dollars. "How much money do you think an hour of your time is worth?" was the question—responses can be heavily influenced by several types of bias: 🧠 Anchoring bias might lead them to base their answer on recent financial discussions or known salaries. 🧠 Optimism bias shows up when people overestimate their worth based on aspirational earnings rather than current realities. 🧠 Moreover, self-serving bias can prompt individuals to inflate their answers to reflect better on their self-perceived value. Zooming out again, we see that data collected isn't as simple as just reading numbers; it’s about how people see themselves and their work in the context of broader economic and cultural landscapes. What percentage of each generation do you think is actually charging this much for their paid time? These biases remind us of the complexity involved in handling human-centered data. Before making what we love to call 'data-driven decisions' in organizations, keep in mind the biases that are crucial in shaping the context of the data. All data is insightful, but being cautious with lurking biases is key.

14

4 Comments

** Any Databricks based Data Engineering Guru's looking for a new opportunity? ** At Databricks Conference? :-) I am looking for a Data Engineer with deep experience (NOT KNOWLEDGE only) but real world practical experience. Can be based in LA/NYC (2-3 days per week) or possibly 100% remote. I'd love to hear (DM me) exactly, what, where, when and how you worked with Databricks. This is a good start... I am looking for people with practical experience and not people willing to learn. We need the person to hit the ground running. Therefore priorty will be given to suitable applicants. Feel free to tag anyone. *** PLEASE NOTE *** 1. Do not expect a reply to your message - why ? I anticipate there will be an abundance of messages/emails etc and, realistically, 90%+ will not be a fit (based on experience) however I will review EVERY messsage but only reply to those I can help at this moment. 2. Its not personal, however this is conditional on your databricks experience, this is nothing to do with you as a person. #databricks #aws #dataengineering #dataengineerjobs

38

11 Comments

Does the rise of Generative AI research push data science into the background? 💭 Do we still need data science expertise to grasp the fundamentals of generative AI? 🤔 This article explores the intersection of data science and generative AI, offering insights into their overlapping areas and highlighting the crucial role data scientists play in the continued advancement of Generative AI 🤖 #GenerativeAI #DataScience #ArtificialIntelligence #medium

5

4 Comments

Data discrepancies can distort insights, but innovative tools like fuzzy matching in Tableau Prep turn chaos into clarity.   From phonetic groupings to spelling-based algorithms, fuzzy matching ensures your messy datasets deliver precise, actionable intelligence. In his latest blog Surendra Sharma breaks down the various methods Tableau Prep offers to clean and transform data for more accurate analysis.   Learn how to elevate your data prep game. Read Surendra’s blog here - https://lnkd.in/gpda-D5Y #DataPreparation #TableauPrep #FuzzyMatching #DataCleaning #Analytics #DataScience #TableauTips #DataQuality #BusinessIntelligence #DataManagement #BI #DataAnalytics #DataWrangling #InsightfulData #Blog #USEReady

14

Would you like to run MMM or run after MMMs? I think a lot of companies run AFTER MMMs rather than letting the MMMs run FOR them. Naavics helps companies in three ways. 1. Identify and fix the delays that creeps in your data 2. Work with your teams to set Data quality standards that apply for MMM 3. Making sure the results timeline aligns with your Media planning #MMM #attribution

2

An interesting study from Bill Schmarzo and Mark Stouse in Data Science Central "We have been vocal about my belief that organizations focusing on GenAI-driven productivity improvements rather than GenAI-driven innovation improvements are missing the bigger prize. GenAI tools can create value for an organization, but that increase in value is inconsistent. On one hand, GenAI tools can improve the productivity (efficiency) of below-average workers. On the other hand, they can significantly boost the innovation (effectiveness) of top-performing workers who better understand critical value-creation concepts such as data science, design thinking, and economics." #ai #genai #automation #productivity #innovation #teckedin https://lnkd.in/g7kgMnxf

8

1 Comment

Wake up! LLMs have jolted companies from their data slumber. But mountains of disjointed info, scattered silos, ignored guidelines, and skyrocketing costs are getting in the way. Sanjay Deshmukh tells us how Snowflake is stepping up as the new data sheriff and bring order to this new chaotic data frontier. #data #snowflake #LLM #generativeai #AI #datamanagement

14

There has been a lot of hype around a semantic layer or metrics layer as an independent and separate part of the analytics stack. We've been hearing about this for years, but I think nobody in the data ecosystem has really delivered on the promise of an independent semantic layer outside of our friends at Cube. I'm excited that we're launching our Hashboard partnership with Cube today. Now you can easily explore and find insights in the metrics you define in Cube. We'll be hosting a webinar tomorrow to show off how the products work together - link below Hashboard and Cube have a strong shared vision for governed and consistent metrics. While Hashboard has an integrated semantic layer (Hashquery), the purpose is just to drive consistent metrics and exploration inside of Hashboard and to make the most compelling reporting system we can. There is still a missing link for plugging metrics into other tools in the ecosystem. I see two basic needs for a *separate semantic layer: 1️⃣ SQL is not particularly composable, it's hard to share calculations (I've written a lot about this) 2️⃣ The analytics stack has a bunch of missing links: tools not being able to interoperate. Power BI, probably one of the most popular BI tools doesn't work particularly well with Snowflake for example. Hashboard is certainly not going to help you connect Power BI and Snowflake anytime soon. There are a few requirements for a semantic layer to be useful as a separate part of the data stack (these problems are *hard): 1️⃣ It's ultimately a network effects business: you need lots of integrations both upstream and downstream for all of this to work. 2️⃣ It's hard to get visualization tools to want to adopt you until you have lots of upstream integrations and customers. So you have to make integration really easy. I think Cube's secret sauce is its sql api: you can just write sql against Cube as a visualization tool which means that you don't have to do a highly customized integration. Their semantic layer speaks sql which is pretty familiar for all of the downstream tools in the ecosystem It has been delightful to get to know and work with the whole cube team Brian Bickell Artyom Keydunov and Tony Kau!

50

4 Comments

Have you tried Snowflake's Artic? Artic is the latest open sourced LLM launched by Snowflake. It's amazing to observe the innovation driving training and operating efficiency for LLMs. Some Specs about Artic: - Dense-MoE Hybrid transformer architecture - 480B parameters spread across 128 fine-grained experts - 17B Active Parameters - Apache 2.0 licensed The main use cases of Arctic are SQL, coding, and instruction following for enterprise applications. Generating SQL code presents an important challenge, demanding a deep understanding of both natural language and programming syntax. It involves crafting precise outputs that faithfully reflect the user's intentions. Artic is available on Hugging Face, Together AI, Microsoft Azure, NVIDIA API Catalog and Replicate. It's also coming soon on Amazon Web Services (AWS), Lamini and Perplexity.

26

6 Comments