Practical A/B testing frameworks, behavioral science, and conversion optimization — for growth leaders responsible for revenue.
38 practitioner-written articles covering everything the official Optimizely docs skip — Stats Engine mechanics, QA checklists, targeting gotchas, results interpretation, and program scaling. Built for experimentation leads who need answers, not tutorials.
In the AI-assisted building era, clean code isn't about readability — it's about context window management. Here's the hard constraint I run for every solo app: no file over 200 lines, ever. And here's exactly why that rule doubles my shipping velocity.
You don't need better ideas. You need a system that forces completion under constraint. Most non-execution is risk management in disguise — and the fix isn't more time or more polish. It's cutting scope until shipping becomes inevitable.
Most people pricing expert calls optimize the wrong variable. Rate is not the game — selection frequency is. Here's how the matching market actually works, and why the best consultants price lower than their ego wants them to.
Unicorns aren't created by talent or ideas. They're created by systems that allow long-term compounding under uncertainty. Here are the five constraints that quietly decide whether a country produces category leaders or just well-run local businesses.
Most recession forecasts don't fail because the data is wrong. They fail because analysts treat deteriorating indicators as signals instead of tracking whether the underlying system has actually broken. Here's the three-channel framework that works.
Most analytics dashboards break silently. When breakdown rows don't sum to your total users, you're not looking at a broken report — you're looking at overlapping populations pretending to be a funnel. Here's why, and how to fix it.
Smart people build productivity systems that optimize for storage instead of throughput — and their output drops. Here's why organization backfires, and how constraint-based execution fixes it.
Atticus Li reduced experimentation analysis time by 40% at NRG Energy by integrating AI tools including Claude, ChatGPT, and Optimizely AI into the testing workflow — with specific workflows, governance standards, and honest limitations.
Behavioral economics is a powerful tool for conversion optimization — but the field went through a replication crisis. Here's how to separate what actually works from what was built on shaky research, and how to test behavioral hypotheses in your own context.
Behavioral economics is powerful, but the field has had a reputation crisis. Here is how to use principles like loss aversion, anchoring, and social proof without getting burned by unreliable research.
The best products come from founders who use their own product every day. Here is why dogfooding is the real competitive moat — and why AI coding tools have changed the math for non-technical founders forever.
A data analyst's real job is not producing dashboards. It is helping stakeholders make better decisions with data. Here is how analysts should work with experimentation teams, translate data into stories, and earn a seat at the decision table.
Atticus Li shares data storytelling lessons from presenting experimentation results to C-suite executives at NRG Energy and Silicon Valley Bank — including the textbook analogy, the consultant mindset, and how financial storytelling grew the program.
Most experimentation advice assumes perfect statistical significance. Here is how to make the best decision when the data will never be complete — a pragmatic framework from 150+ enterprise experiments.
When something works, double down. But never become dependent on a single acquisition channel. Here is how to scale what is working while protecting against algorithm changes and platform risk.
Atticus Li designed NRG Energy's EBITDA impact estimation model that translates A/B test results into verified financial impact, turning experimentation from a cost center into a revenue driver generating $30M+ in projected value.
Atticus Li shares five real enrollment flow A/B tests from NRG Energy that collectively projected over $1M in annual revenue — with exact metrics, behavioral mechanisms, and the decision frameworks behind each test.
The experimentation team that treats itself as an internal consulting group outperforms the one that treats itself as a test execution shop. Here is how to reposition your team and why it matters for every stakeholder relationship.
Atticus Li shares lessons from leading marketing analytics at Silicon Valley Bank and NRG Energy — covering Google Analytics vs Adobe Analytics, data storytelling for executives, geo-incrementality testing, and what it means to be a consultant, not just an analyst.
Atticus Li designed a geo-incrementality experiment at Silicon Valley Bank to measure billboard and OOH advertising impact on digital demand, proving +97.8% web traffic lift in Miami and +94.4% in Austin versus the Seattle control market.
Atticus Li built Jobsolv from zero to 30,000+ users and $80K+ revenue without paid advertising. Here's the unglamorous playbook of product-led growth, channel diversification, and obsessive unit economics that actually worked.
Atticus Li built and managed a 27-member cross-functional team for Jobsolv while working full-time as experimentation lead at NRG Energy — coordinating a dev agency in Egypt, a UX/UI design agency, and freelancers across Upwork and Fiverr.
Atticus Li built the governance framework for running 100+ A/B tests per year across NRG Energy's five retail brands, serving 7M+ customers in 24 states.
Most experimentation teams write hypotheses that are really just disguised solutions. Here is why problem-first hypothesis design produces better tests — and how to stop solutionizing before you understand the leak.
Practical A/B testing frameworks, behavioral science, and CRO strategies for growth leaders responsible for revenue. Practical. Free. Weekly.
Free · No spam · Unsubscribe anytime