I\u2019m a new data analyst trying to land my first full-time role, and I\u2019m building a portfolio and practicing for interviews as I apply. I\u2019ve done the usual polished datasets (Titanic\/clean Kaggle stuff), but I feel like they don\u2019t reflect the messy, business-question-driven work I\u2019d actually do on the job.<\/p>\n
I\u2019m looking for public datasets that let me tell an end-to-end story: define a question, model\/clean in SQL, analyze in Python, and finish with a dashboard. Ideally something with seasonality, joins across sources, and a clear decision or KPI impact.<\/p>\n
Datasets I\u2019m considering: – NYC TLC trips + NOAA weather to explain demand, tipping, or surge patterns – US DOT On-Time Performance (BTS) to analyze delay drivers and build a simple ETA model – City 311 requests to prioritize service backlogs and forecast hotspots – Yelp Open Dataset to tie reviews to price range\/location and detect \u201cmenu creep\u201d or churn risk – CMS Hospital Compare (or Medicare samples) to compare quality metrics vs readmission rates<\/p>\n
For presentation, is a repository containing a clear README (business question, data sources, and decisions), EDA\/modeling notebooks, a SQL folder for transformations, and a deployed Tableau\/Looker Studio link enough? Or do you prefer a short write-up per project with charts embedded and code linked at the end?<\/p>\n
On the interview side, I\u2019ve been rehearsing a crisp portfolio walkthrough with Beyz interview assistant, but I still need stronger datasets to build around. If you hire analysts, what makes you actually open a portfolio and keep reading?<\/p>\n
Last thing, are certificates like DataCamp\u2019s worth the time\/money for someone without a formal DS degree, or would you rather see 2\u20133 focused, shippable projects that answer a business question? Any dataset recommendations or examples would be hugely appreciated. <\/p>\n<\/div>\n
submitted by \/u\/Various_Candidate325 <\/a>
[link]<\/a><\/span> [comments]<\/a><\/span><\/p>
