Floria Hau

tell me you're a data professional without telling me you're a data professional... when I build brio train tacks, I build them only using track pieces. Once it's up and running then I will swap out pieces of track for the fancy bits i.e a windmill. Foundations over fancy features for release 1.0. #dataleaders #dataprofessionals #dataleadership #cdo #dataanalytics

54

25 Comments

NEW: Questions in Dataviz: How do we bring joy to data visualisations?  https://lnkd.in/efGpzyC5 Alternatively titled "What is visual prosody?" - a review of Joyful Infographics by Nigel Holmes

14

3 Comments

This months instalment of my Causal AI series is up on Towards Data Science, this time exploring optimising non-linear treatment effects in Pricing & Promotions (or Marketing): https://lnkd.in/etYraNxt

101

2 Comments

Snowflake Cortex is now available to Snowflake customers, and you may have thought - yawn - another big software company has launched Generative AI capabilities. But this is actually a game-changer for marketers. With Snowflake's Cortex now in place, brands can use Google AI, Mistral AI, Meta AI, Snowflake's own Artic, and other leading generative AI products in a secure, efficient, and user-friendly way... #CDP #customerdataplatform #snowflake

17

1 Comment

Anyone building AI agents? Would love to share notes on stack and data flow. Explored a few all-in-one platforms but need more flexibility and honestly don’t live the saas model. Currently leaning into n8n as the hub. Using pinecone for the vector db but will prob need to shift to Snowflake’s vector format to fit into our enterprise stack. Still early in my research. Would also appreciate any links to companies/folks on the edge of this space. #aiagent #n8n #automation

15

19 Comments

Self Serve Analytics... We've been having the same conversation on this for probably over 15 years so Andy Crossley and Nick Latocha reopened the doors of The Data Pubcast and we had another conversation on it. As you can see from the screenie below, it was a hoot. Go and have a listen on your lunch break 🎧 #podcast #dataanalytics #dataanalysis #dataleaders #dataleadership #businessintelligence

27

3 Comments

🚀 Hiring the Right Data Scientist 🚀 I had a call with a new client this week who has hired a data scientist without fully understanding their needs. They now need to revisit their search, and we have a plan ensure sure we find the right person. Here are the key takeaways: Clearly define what you need the data scientist to achieve - Specify the tasks, goals, and outcomes expected from the role. Identify the skills required - Data scientists have different strengths. Determine if you need expertise in building forecasting models, analyzing customer data, optimizing processes, or applying NLP techniques. Once you understand your specific requirements, focus on finding candidates with those exact skills. It may take more time and thorough qualification, but the right fit is worth the effort. #DataScience #Recruitment #AI #Analytics #NLP #Forecasting

46

2 Comments

Data Vizzers - Can you give me 20 mins of your time? I am looking for volunteers to spend 20 mins to build some charts. Basically I have a little excel worksheet with 4 viz exercises in and I want some people to test it out. Comment or Message me if you fancy having a play with some totally made up data. TIA. #dataanalysts #datavisualisation #training #dataviz #datastorytelling

19

17 Comments

Since joining DT Consulting I've been analysing lots of research data so thought I'd go back in the Workout Wednesday backlog to find a survey data challenge (2023 - W4) for my first viz of the year. Really enjoyed this one, displaying NPS using a floating stacked bar chart with the likert (pronounced Lick - ert rather than Like - ert, thanks Steve Wexler) scale. Tableau public link: https://lnkd.in/ePA3j-m3 Need to get back into the swing of these community challenges again! #datafam #tableau #dataviz

53

How do you obtain buy-in and sponsorship for data management? And, does buy-in and sponsorship automatically mean you'll have success? In Episode 35, of Season 4, of Driven By Data: The Podcast I was joined by Rupal S., Head of Data Management at ASOS.com Where we discuss how to obtain buy-in and sponsorship for data management, which includes; 🎤Ending up in Data Governance by accident 🎤Rebranding from Data Governance to Data Management 🎤The recognition that Data Management can support business goals 🎤Why repetition, persistence and resilience are the key to driving buy-in and sponsorship 🎤Using different content and analogies to communicate with different teams 🎤The importance of getting into as many meetings as possible 🎤Using a variety of different mediums to communicate with your business teams 🎤Why data management success requires a fundamental mindset shift for the entire organisation  🎤Communication differences between business teams and executives 🎤Why the perception of legacy organisations versus digital natives isn’t always accurate 🎤Prioritising which data management use cases to tackle first  🎤Why willing partners are critical and how you identify them 🎤Why people skills are so critical when working in data management 🎤How personality can take you a long way 🎤Why the perception of job titles can make a difference 🎤Balancing quick wins with long-term strategic initiatives 🎤The role of technology and tooling in data governance 🎤Why data discoverability is a critical component to success 🎤The importance of retaining and sharing company knowledge 🎤The differences between data governance and AI governance 🎤How GenAI will impact data governance  Full episode ⬇️ #datamanagement #datagovernance #sponsorship

46

6 Comments

In my latest Towards Data Science article, we shift our focus to understanding how multi-collinearity can damage the causal inferences you make, particularly in marketing mix modelling. Should be a good read for those of you interested in MMM: https://lnkd.in/eVErcCit

99

4 Comments

How will LLMs change the future of the Market Research industry? 🤖 A comprehensive article here from Kantar which covers these key points: 🔍 Summarization & Reporting: Quickly condense survey responses and organize data. 🔍 Theme Identification & Prediction: Extract insights and predict outcomes from large datasets. 🔍 Text Cleaning & Creative Writing: Improve data quality and assist in drafting materials. 🔍 Conversational AI: Enhance intelligent interviewing and natural language queries. My key takeaway from this is that the potential is immense, but we must navigate risks like data quality, biases, and ethical concerns. The future is bright, but careful implementation is key. Let's embrace the AI revolution in market research! 🌟 https://lnkd.in/eMMQHKN3 #AI #MarketResearch #FutureOfWork

9

2 Comments

Really enjoyed chatting with Hana M. K. - it was a pleasure to meet and talk all things dataviz. That’s me below with my thoughts on testing yourself in dataviz competitions, but check out our full conversation at https://lnkd.in/divRZQDP

14

1 Comment

Really chuffed to have been accepted again thanks DSF. Looking forward to this one. A new topic to talk about with a familiar twist on data visualisation. Best get building those slides. Hopefully see some familiar faces on the day, and if you're interested in coming along, check out the free ballot link below. Enter the ticket ballot below 🎟️ https://ow.ly/FvB550QF1Ot #DSFmayday #datasciencefest #datasciencefestival

28

3 Comments

Interview: The power of data storytelling in modern business Most people will know I love data visualisation and storytelling and most of you will be familiar to my disliking of the old pie chart. After finding myself in the Raconteur | B Corp™ Data Leaders 25 list, the lovely people over there and I had a nice chinwag on data viz, storytelling, leadership and data literacy. Article is in the comments... #DataLeaders #DataVisualisation #DataStorytelling

69

6 Comments

I've just started to learn dbt and it really looks great. One of my favourites so far is modularity that I've struggled a lot so far using only SQL statements in Fabric warehouse. There I should have chosen to use references which made is essential to run them sequentially in a pipeline or to duplicate the code for multiple view, that requires parallel maintenance of the codes. With modularity in dbt it seems pretty easy to handle. Looking forward for the next lessons.

5

Requirements are key. I’ve been building data things for over 15 years and as we all know, getting clear and concise requirements can be difficult but it is key. At the weekend I was the one giving requirements… Question : How would you like to feel after your massage? My response: like I’ve been tortured Clear, concise and to the point and guess what… I got what I asked for. There’s a message in there somewhere. #dataleaders #dataanalytics #dataproducts #dashboard #datascience

15

Today's Challenge: Penultimate Order - Interview Question from an E-commerce Data Analyst Role 🚀 Today is Day 7 of the 21-day SQL Challenge! 🎯 Problem Statement: You are a data analyst working for an e-commerce company, responsible for analysing customer orders to gain insights into their purchasing behaviour. Your task is to write a SQL query to retrieve the details of the penultimate order for each customer. However, if a customer has placed only one order, you need to retrieve the details of that order instead, display the output in ascending order of customer name. Table Structure: The orders table contains: order_id: Unique identifier for each order order_date: The date of the order customer_name: Name of the customer product_name: Product purchased sales: The revenue from the order 🧠 My Thought Process: 1️. Rank the Orders For each customer, rank their orders by order_date in descending order (most recent first). I used ROW_NUMBER() to assign a rank to each order (rank = 1 for the latest order, rank = 2 for the penultimate). 2️. Filter for Penultimate Orders Retrieve the order with rank = 2 if a customer has more than one order. If the customer has only one order, retrieve the one with rank = 1. 3️. Handle Edge Cases Ensure the query handles customers with only one order gracefully. 4️. Sort the Results Order the output by customer_name in ascending order for readability. Key Steps: Created a CTE (common table expression) to rank the orders and calculate the total number of orders per customer. Filtered for either the penultimate order (rank = 2) or the only order (rank = 1 where total_orders = 1). Ensured the output is clean and sorted by customer name. 💡 Closing Thoughts: This challenge was an excellent opportunity to leverage the power of window functions like ROW_NUMBER() and COUNT(). These tools allow us to rank and group data efficiently, uncovering insights into customer behavior with ease. What do you think of my approach? Would you solve it differently? Let’s exchange ideas below! 💬 Thanks Sylvia Wutche for this great oppoturnity #21daysSQLchallenge #SQL #DataAnalytics #LearningSQL #DataAnalyst

23

2 Comments