# 33sticks.com > AI-optimized mirror of 33sticks.com containing 25 pages totalling 19,839 words of clean markdown content, structured data, and semantic HTML. Original source: https://33sticks.com/. Last updated: 2026-06-15T23:54:36.054Z. Each page is available as HTML (with JSON-LD structured data) and Markdown (text-only, ideal for LLMs and RAG). ## Homepage - [**Your Analytics Partners, Not Just Another Agency**](/content/site-root.html): 33 Sticks is a data intelligence company that fundamentally changes how people feel about analytics and their futures., 33 Sticks is a data intelligence company that fundamentally changes how people feel about analytics and their futures. (268 words) ## Articles & Blog Posts - [articles/how-to-spot-when-someone-cherry-picks-data-to-tell-you-what-they-want-you-to-believe.html](/content/articles/how-to-spot-when-someone-cherry-picks-data-to-tell-you-what-they-want-you-to-believe.html) (1 words) - [articles/the-map-that-never-dies-why-this-misleading-electoral-visualization-still-works-in-2025.html](/content/articles/the-map-that-never-dies-why-this-misleading-electoral-visualization-still-works-in-2025.html) (1 words) - [articles/how-to-spot-bad-data-a-case-study-in-corporate-research-theater.html](/content/articles/how-to-spot-bad-data-a-case-study-in-corporate-research-theater.html) (1 words) - [articles/the-new-compliance-challenge-hiding-in-your-optimization-stack.html](/content/articles/the-new-compliance-challenge-hiding-in-your-optimization-stack.html) (1 words) - [articles/we-took-googles-new-ai-analyst-for-a-test-drive/index.html](/content/articles/we-took-googles-new-ai-analyst-for-a-test-drive/index.html) (1 words) - [articles/we-tore-apart-two-hotel-booking-experiences-heres-what-we-actually-look-at.html](/content/articles/we-tore-apart-two-hotel-booking-experiences-heres-what-we-actually-look-at.html) (1 words) - [[We Took Google's New AI Analyst for a Test Drive](/content/articles/we-took-googles-new-ai-analyst-for-a-test-drive/index.html)](/content/articles/index.html) (1,556 words) - [We Asked 3 LLMs to Build a Data Viz. Here's What Happened.](/content/articles/we-asked-3-llms-to-build-a-data-viz-heres-what-happened.html): If you've been on LinkedIn lately, you've probably seen your fair share of posts about how "AI will replace your analysts!" "ChatGPT just made my whole data team obsolete!" We got curious. Not about whether AI can theoretically do analysis, we know it can be a very useful tool,, If you've been on LinkedIn lately, you've probably seen your fair share of posts about how "AI will replace your analysts!" "ChatGPT just made my whole data team obsolete!" We got curious. Not about whether AI can theoretically do analysis, we know it can be a very useful tool, but whether it can deliver an actual work product. The kind of thing a marketing director might hand to their exec team. So we ran a simple test. (1,329 words) - [What If Your Optimization Team Could Answer Their Own Data Questions?](/content/articles/what-if-your-optimization-team-could-answer-their-own-data-questions.html): The organizations that will lead over the next few years aren't the ones with the best point solutions. They're the ones that connect their tools into a system where data flows to wherever decisions are being made — without requiring a human intermediary for every question., The organizations that will lead over the next few years aren't the ones with the best point solutions. They're the ones that connect their tools into a system where data flows to wherever decisions are being made — without requiring a human intermediary for every question. (1,581 words) - [More A/B Tests Won't Save You](/content/articles/more-ab-tests-wont-save-you/index.html): The best testing programs we've seen aren't the ones that run the most tests. They're the ones where every test has a reason, every result gets analyzed with intellectual honesty, and the people doing the work are trusted to do it well. Volume is a byproduct of maturity, not a substitute for it., The best testing programs we've seen aren't the ones that run the most tests. They're the ones where every test has a reason, every result gets analyzed with intellectual honesty, and the people doing the work are trusted to do it well. Volume is a byproduct of maturity, not a substitute for it. (1,824 words) - [[Market Basket Analysis With Adobe Analytics And RSiteCatalyst](/content/articles-restore/data-analysis/market-basket-analysis-using-r-sitecatalyst/index.html)](/content/articles-restore/index.html): Explore strategic analytics and data analysis resources to elevate your insight-driven decisions. Discover actionable insights today., Explore strategic analytics and data analysis resources to elevate your insight-driven decisions. Discover actionable insights today. (1,217 words) - [Market Basket Analysis With Adobe Analytics And RSiteCatalyst](/content/articles-restore/data-analysis/market-basket-analysis-using-r-sitecatalyst/index.html): Any merchandiser can tell you their top-selling product. If you ask what the top product combination is sold on their website, they might actually know (there’s a lot of sharp people out there!), but you’re also likely hear a guess like “multiple <variations> of <top-selling pro, Any merchandiser can tell you their top-selling product. If you ask what the top product combination is sold on their website, they might actually know (there’s a lot of sharp people out there!), but you’re also likely hear a guess like “multiple <variations> of <top-selling product>.” Obvious, but not very insightful. The better question to ask is “What are some interesting product combinations that are being purchased, and should we be offering those to our customers?”. To answer this question, we can use Market Basket Analysis/Association Rules*, which will be the technical focus of this blog post using the arules package in R, along with RSiteCatalyst. (1,188 words) - [[Market Basket Analysis With Adobe Analytics And RSiteCatalyst](/content/articles-restore/data-analysis/market-basket-analysis-using-r-sitecatalyst/index.html)](/content/articles-restore/category/data-analysis/index.html) (1,217 words) - [Data Theater: What McKinsey's AI Report Actually Says (And What LinkedIn Gurus Won't Tell You)](/content/articles/when-accurate-data-becomes-a-weapon/index.html): McKinsey's AI report says 57% of work could be automated—but viral LinkedIn posts aren't telling you the whole story. Here's what the research actually says., McKinsey's AI report says 57% of work could be automated—but viral LinkedIn posts aren't telling you the whole story. Here's what the research actually says. (1,420 words) - [The Power of Advanced Calculated Metrics in Adobe Analytics](/content/articles-restore/data-analysis/adobe-analytics-advanced-calculated-metrics/index.html): 33 Sticks Co-Founder and Principal Analyst, Hila Dahan, presented an extremely popular session at Adobe Summit 2019 in Las Vegas on leveraging Adobe Analytics Advanced Calculated Metrics. Hila's session focused on using advanced Calculated Metrics to drive deeper insight into consumer behavior. F, 33 Sticks Co-Founder and Principal Analyst, Hila Dahan, presented an extremely popular session at Adobe Summit 2019 in Las Vegas on leveraging Adobe Analytics Advanced Calculated Metrics. Hila's session focused on using advanced Calculated Metrics to drive deeper insight into consumer behavior. Forecasting - Introducing plan data into Analysis Workspace and creating pacing visualizations. Statistical Significance - Calculating test results in Analysis Workspace and introducing discipline into a testing program. Churn - Visualizing retention and churn in Analysis Workspace and understanding loyalty patterns. (368 words) - [Under the Hood at 33 Sticks: What Four Months Taught Me About Data, Strategy, and What Really Matters](/content/articles/under-the-hood-at-33-sticks-what-four-months-taught-me-about-data-strategy-and-what-really-matters.html): I was hired without a CV. I learned that 2-hour meetings are a scam. And I discovered that being called 'difficult' often just means refusing to shrink into systems that weren't built for real innovation. A reflection on four months that permanently raised my standards for how I want to work., I was hired without a CV. I learned that 2-hour meetings are a scam. And I discovered that being called 'difficult' often just means refusing to shrink into systems that weren't built for real innovation. A reflection on four months that permanently raised my standards for how I want to work. (1,096 words) - [Your Data Team Structure is Probably Wrong](/content/articles/your-data-team-structure-is-probably-wrong/index.html): There are two dominant philosophies for structuring data teams: centralized and distributed. Both are wrong. Centralized teams have the expertise but lack business context. Distributed teams have proximity but lack leadership, governance, and consistent data. The answer is a hybrid, centralized expe, There are two dominant philosophies for structuring data teams: centralized and distributed. Both are wrong. Centralized teams have the expertise but lack business context. Distributed teams have proximity but lack leadership, governance, and consistent data. The answer is a hybrid, centralized expertise with embedded liaisons in the business units you serve. Here's why, and how to rank the three models. (1,338 words) - [Your A/B Test Won. But Did It Actually Win?](/content/articles/your-ab-test-won-but-did-it-actually-win/index.html): A mobile checkout test drove $6.2M in incremental bookings. It also killed 41,000 loyalty enrollments a year. Most optimization teams celebrate the first number and never measure the second., A mobile checkout test drove $6.2M in incremental bookings. It also killed 41,000 loyalty enrollments a year. Most optimization teams celebrate the first number and never measure the second. (1,539 words) - [Your A/B Test Is Counting People Who Never Saw It](/content/articles/your-ab-test-is-counting-people-who-never-saw-it/index.html): A/B testing platforms bucket every visitor the moment a page loads, whether they interact with your test or not. If you're testing a mini-cart experience that only 15% of visitors ever open, the other 85% are noise in your results. We've spent years writing custom JavaScript to work around this. Opt, A/B testing platforms bucket every visitor the moment a page loads, whether they interact with your test or not. If you're testing a mini-cart experience that only 15% of visitors ever open, the other 85% are noise in your results. We've spent years writing custom JavaScript to work around this. Optimizely's Conditional Activation replaces all of that with a dropdown. (1,228 words) - [Your Customers Aren't Abandoning Their Cart — They're Thinking](/content/articles/your-customers-arent-abandoning-their-cart-theyre-thinking.html): Cart abandonment rate is hiding deliberate buyers who spend 33% more per order. Here's how to build the segments that separate them — and why your retargeting strategy needs to change., Cart abandonment rate is hiding deliberate buyers who spend 33% more per order. Here's how to build the segments that separate them — and why your retargeting strategy needs to change. (1,098 words) - [How to Prioritize A/B Tests: A Simple Framework for Choosing What to Run First](/content/articles/how-to-prioritize-ab-tests-a-simple-framework-for-choosing-what-to-run-first.html): Learn how to prioritize A/B tests using the ICE framework. Score tests on impact, confidence, and ease to build a strategic testing roadmap that drives results., Learn how to prioritize A/B tests using the ICE framework. Score tests on impact, confidence, and ease to build a strategic testing roadmap that drives results. (772 words) - [What’s the Minimum Sample Size You Need Before Trusting Your Test Results?](/content/articles/whats-the-minimum-sample-size-you-need-before-trusting-your-test-results.html): Learn how to calculate the right sample size to make A/B test results trustworthy without wasting time or traffic., Learn how to calculate the right sample size to make A/B test results trustworthy without wasting time or traffic. (583 words) - [Optimizely Opal x Adobe Analytics: Ask Your Data Anything with Natural Language](/content/articles/optimizely-opal-x-adobe-analytics-ask-your-data-anything-with-natural-language.html): A demo of the Optimizely Opal connector for Adobe Analytics. Query any dimension, metric, or segment using plain English. No report builder, no drag-and-drop, no workspace setup. Just ask., A demo of the Optimizely Opal connector for Adobe Analytics. Query any dimension, metric, or segment using plain English. No report builder, no drag-and-drop, no workspace setup. Just ask. (210 words) - [sitemap-xml.html](/content/sitemap-xml.html) (1 words) ## Resources - [Full Page Index](/index.html): Browse all cached pages with rich metadata - [About This Cache](/content/about.html): Methodology, technical details, and usage guidelines - [XML Sitemap](/sitemap.xml): Machine-readable sitemap for crawler discovery - [Robots.txt](/robots.txt): Crawler directives