![Hypothesis Formula
By changing [problem] into [solution] we will increase [metric ]
All hypotheses should come as a result of conversion research: KNOW, don’t GUESS](https://image.slidesharecdn.com/meetup16dec-datadrivenprocesstooptimization-160127083543/85/Meetup-16-dec-data-driven-process-to-optimization-14-320.jpg)
Meetup 16 dec data driven process to optimization
- 1. The data driven process to identify, prioritize, launch and monitor your CRO experiments
- 2. What is conversion optimization? Conversions – Actions you want your visitors to take on your site ◦ Signs up to your newsletter ◦ Buys a product ◦ Downloads a file ◦ Etc. Conversion optimization – understand your customers, provide them with a better experience on your site and simplify everything so they can convert more.
- 3. The Big Picture Conversion optimization is about doing better marketing Assuming Knowing Data
- 4. The Big Picture “Data are just summaries of thousands of stories – tell a few of those stories to help make the data meaningful.” – Chip & Dan Heath Data Information Insight
- 5. Don’t just assume It makes an ass out of u and me
- 6. Mindset of the Optimizer Rule #1: There is no one size fits all
- 7. Mindset of the Optimizer There is no “this always works” Ego and Bias – not your friends when it comes to optimization We need to know, not assume
- 8. Typical CRO Process If you can’t describe what you are doing as a process, you don’t know what you’re doing. – W. Edwards Deming
- 9. Determine Business Objective – KPIs, Metrics to increase Setup Data Gathering Insight Phase – Sources of traffic, patterns, personas Define Hypotheses DesignTechnical IntegrationTesting Learning and Improving Customer Theory Back to Step 2 Prioritize Hypotheses – Testing Plan
- 10. Conversion Research 101 Experience based assessment ◦ Relevance ◦ Clarity ◦ Value ◦ Friction ◦ Distraction
- 11. Conversion Research 101 -Online surveys -Web traffic surveys -Live chat transcripts -Customer support insights -User testing Qualitative Research
- 12. -Web analytics research -Mouse tracking analytics Quantitative Research
- 13. Developing a hypothesis With each hypothesis we want to understand: ◦Which problem we’re solving ◦What’s the solution ◦Which metric we are trying to improve with this
- 14. Hypothesis Formula By changing [problem] into [solution] we will increase [metric ] All hypotheses should come as a result of conversion research: KNOW, don’t GUESS
- 15. Prioritizing Experiments Optimizing the optimization process is often just as important as the tests themselves WiderFunnel uses this PIE Framework: PIE Framework: Potential Importance Ease
- 16. T.I.R. Method ◦ Time 1 – longest time 5 – shortest time ◦ Impact 1 – lowest impact 5 – highest impact ◦ Resources 1 – most resources 5 – least resources Multiply these scores together and start working on the projects that have the highest scores as those will have the most lift with the least amount of time and the least amount of resources.
- 17. Test It Right Wait until statistical significance Make sure your sample size is big enough Sample size calculator: http://www.evanmiller.org/ab-testing/sample-size.html Don’t call tests too early
- 18. Validity Threats History effect Instrumentation effect Selection effect Compatibility issues – broken code