Modernize 2018: Advanced Experimentation Workshop - Web Marketing ROI & Optimizely
0 likes•82 views
The document details a workshop led by experts on optimizing data and campaign development using Optimizely, focusing on common data issues, measures of analytics, and improving efficiency through extensions and custom implementations. It highlights significant misconceptions surrounding Google Analytics data, emphasizing the importance of accurate tracking and integration with Optimizely to ensure correct analysis. Practical advice for troubleshooting, campaign development, and enhancing performance with advanced techniques is also provided.
![1. Translation
[:en]
Hello World
[:zh]
(Traditional Chinese)
2. Markdown
3. Extending extensions
Cool Use Cases
experience](https://image.slidesharecdn.com/webroi-modernizeadvancedexperimentationworkshop-newfinal-180527162756-180825105443/85/Modernize-2018-Advanced-Experimentation-Workshop-Web-Marketing-ROI-Optimizely-34-320.jpg)
![jQuery – what you need to know about
Optimizely X
Var $ = window.optimizely.get(‘jQuery’) [Optimizely jQuery] when
instrumenting jQuery in your Optimizely campaigns goes wrong versus
var $ = window.jQuery;
var $ = window.jQuery || window.optimizely.get(‘jQuery’);
experience](https://image.slidesharecdn.com/webroi-modernizeadvancedexperimentationworkshop-newfinal-180527162756-180825105443/85/Modernize-2018-Advanced-Experimentation-Workshop-Web-Marketing-ROI-Optimizely-38-320.jpg)
Modernize 2018: Advanced Experimentation Workshop - Web Marketing ROI & Optimizely
- 1. experience Advanced Experimentation Workshop James Spittal, CEO, Web Marketing ROI Holly Joshi, Sr. Customer Success Manager, Optimizely Mansoor Omar, Technical Support Engineer, Optimizely
- 2. 1. Tackling data problems • Misconceptions/theories • Data between GA and Optimizely • How to ensure what you’re seeing is accurate 2. Overcoming common challenges faced operationally • Writing regular expressions • Cross-domain Optimizely 3. Making your campaign development faster and more efficient • Optimizely Extensions • Reducing flickering Agenda experience
- 3. CRO/MarTech Leadership = GSD experience
- 4. experience
- 6. experience All About The Data What Optimizely measures and integrating with your data systems
- 7. 1. Resolving the misconceptions 2. Implementation is the solution 3. Get the most common/most critical right Optimizely Data experience
- 8. 73% … the percentage of Google Analytics implementations that are wrong (leading to incorrect traffic numbers, incorrect revenue, etc). Source: https://econsultancy.com/blog/10925-80-of-online-retailers-are-using-google- analytics-incorrectly
- 11. If you’re not comparing the same thing – you’re not going to see the same numbers. Apples & Oranges experience
- 12. Google Analytics Session Time-Outs (and Optimizely End User Id) By default, a Google Analytics session expires after 30 minutes of user inactivity. This is entirely different to how Optimizely tracks a “unique visitor” (tied to Optimizely End User Id cookie, 10 year expiry). experience
- 13. Impressions (new Optimizely measurement model) – How it actually works Optimizely logs an event whenever an experiment activates (this can be seen as a ‘decision’ under the hood as an XMLHTTPRequest) The decision carries a few pieces of information, including: Visitor ID The ID of the experiment / campaign which is activating The ID of the variation the visitor is allocated to UUID: A unique decision ID (used for deduplication) Again, decision events are generated whenever an experiment activates, even if the visitor has been targeted by the same experiment previously. So in a non-technical sense, you can imagine that a decision is generated each time an experience is delivered to a visitor.
- 14. Impressions (new Optimizely measurement model) select count(distinct(UUID)) from decisions; This deduplication on UUID is important because Optimizely clients sometimes send multiple decisions for the same experiment activation (this is due to the batching behavior of the web snippet). Typically, impressions are allocated on an annual basis. So we just count up the number of unique decision events, subtract them from the annual quota, and voila, you've got your "impressions remaining" count for the year. experience
- 15. Optimizely Results Screen Example
- 17. Audience Conditions – Custom JavaScript experience
- 18. QA everything (DIY) 1. Make a spreadsheet of all of your events, pages and audiences 2. Create a QA campaign for each, to verify it is triggering when intended and not triggering when it shouldn’t. experience
- 20. Optimizely Accelerate Total cost around $15k AUD, includes full events/pages/URLs/audiences QA and campaign build, delivered by approved Optimizely Solutions Partners. experience
- 22. 1. Regular expressions (versus Simple Match, Exact Match and Substring Match) 2. Multi-page tests 3. Making Optimizely work cross-domain (and/or cross-project) Pages & URLs experience
- 23. 1. Matching too freely 2. ‘example.com’ Matches (Substring) for: yourdomain.com.au/example.com Common ‘Pages’ mistakes experience
- 24. Regular Expressions Cheat Sheet experience
- 26. 1. This makes Optimizely cross-domain work 2. In short - Injects an iFrame that allows local storage/cookie to traverse 3. Put waitForOriginSync into your Project JS WaitForOriginSync experience
- 28. experience Campaign Development Extensions, Efficiencies and Performance
- 29. 1. Extensions = more campaigns, faster • Extensions 101 • Why extensions can be tricky • Extension of extensions 2. Experiment flickering minimisation Campaign Development experience
- 30. Extensions
- 32. 1. Templates, modularity, performance, strategy, operational savings… 2. Object- orientated 101 experience
- 33. 1. How many input fields should we have? 2. Why is my extension not showing up/not accessible? 3. Where did my extension go? 4. Data drops out of an extension Challenges experience
- 34. 1. Translation [:en] Hello World [:zh] (Traditional Chinese) 2. Markdown 3. Extending extensions Cool Use Cases experience
- 37. 1. WaitForElement & ES6 2. WaitUntil 3. See the Optimzely Developer Docs 4. Async vs. Sync Document object model challenges experience
- 38. jQuery – what you need to know about Optimizely X Var $ = window.optimizely.get(‘jQuery’) [Optimizely jQuery] when instrumenting jQuery in your Optimizely campaigns goes wrong versus var $ = window.jQuery; var $ = window.jQuery || window.optimizely.get(‘jQuery’); experience
- 39. 1. Use ‘utils’ (WaitForElement, etc) 2. Use ES6/Native JavaScript 3. Asynchronous versus synchronous Most important things to know experience
- 40. JavaScript Hell
- 42. Don’t always believe what Google PageSpeed Insights tells you experience
- 45. 1. If you need help with data analysis problems or auditing your GA data, speak to an Optimizely Solutions Partner or GACP 2. Manny and Holly are available for on- demand Optimizely support for Optimizely customers Next steps experience