Web Marketing Week1
- 1. Internet Marketing Practices Week 1: Introduction Alan Chen, 2011
- 2. Before Getting Started Thank you for joining
- 3. What will be covered
- 4. Time allocation My sharing
- 5. Your feedback Rewards In each section and the end of this topic
- 6. Introduction of Internet Marketing in the following weeks: Acquisition
- 7. Conversion
- 8. Retention
- 9. Why Important (to Reallusion)? The nature of Reallusion products
- 11. Fast to market
- 13. A Typical Flow and Challenges Users Will Face Before Becoming a Reallusion Customer?
- 14. Search term: 3d animation software Must be in first 2 pages to be seen
- 15. And good use experience! If they are interested enough...
- 16. Start from a need or event Find Reallusion Find the right product Convinced by the web page Find the page and start download Complete download and install Happy with the product! Find the purchase page Paid Every Single Point Can Go Wrong
- 17. Start from a need or event Find Reallusion Find the right product Convinced by the web page Find the page and start download Complete download and install Happy with the product! Find the purchase page Paid ? copy sold ? complete downloads ? register members ? purchase attempts ? download attempts ? product page visits 100,000 people want to find 3d animation software ? visitors visit Reallusion
- 18. Start from a need or event Find Reallusion Find the right product Convinced by the web page Find the page and start download Complete download and install Happy with the product! Find the purchase page Paid MD & ECD CPM & Art ECD RD & QA ECD & Sales
- 19. How to optimize the 3 steps? We will continue this topic in the following weeks http://www.matraxis.co.uk/
- 20. How Can We Know What Users Need?
- 21. Eye-tracking
- 22. Expert analysis
- 23. Lab usability testing The ultimate user experience is improved much more by three tests with 5 users than by a single test with 15 users - Jacob Nielsen Survey
- 24. If I had asked customers what they wanted, they would have told me they wanted a faster horse - Henry Ford
- 25. Web analytic tools Web logs
- 26. Web beacons
- 27. Java Script tags
- 28. Packet sniffing
- 29. Script based vs Log based analytics Script-based Analyze useful, easy-to-understand data
- 30. More meaningful data by using cookies (know what this user has done on your site)
- 31. Can filter traffic by sources (e.g. AP activation)
- 32. Can be used even you cant access the server log (e.g. a shared server) Log-based Setup once
- 33. No worries about page-level error
- 34. Always own the data
- 35. No Java-script required (2~6% users have it turned off)
- 36. No additional load to network or browser
- 37. Traditional marketing: I know 50% of my budget spent was in vain, but I don't know where
- 38. On the internet, you can know much more about users at lower cost. If you plan, and implement in a scientific way... Road banner: Est. exposure vs. Different URL
- 39. Commercial: Nielsen TV ratings vs. Search term
- 40. DM: Different coupon vs. coupon code plus A/B testing
- 41. You can know visitor navigation path
- 42. You can know abandon path
- 43. Why Google Analytics and What it can Provide?
- 44. A Javascript tag type web analytics system
- 45. Developed from Urchin Software Corporation, which was acquired by Google in April 2005
- 46. The USD 500/month service turned into free service, and keep involving
- 47. utmz: type of referral, session number, etc
- 49. utmc: visit session (close browser)
- 50. utmb: visit cookie (30 minutes)
- 51. utma: unique ID, time of visits, etc
- 52. Traffic overview by given time period
- 53. Compare different time period Effectiveness of events, promotions
- 55. Quick dashboard view and deeper info
- 56. Detail: Visitors Who are more interested in the site?
- 57. Detail: Visitors Who are more interested in the site?
- 58. Detail: Visitors How much time they spend on the site?
- 59. Detail: Visitors Browser, OS, Flash version, network, etc
- 60. Detail: Visitors Browser, OS, Flash version, network, etc Different countries, analytics profile have different value. E.g. If we count top 5 countries only, 1024x768 may drop
- 61. Detail: Traffic source Direct traffic, referral, or search
- 62. Detail: Traffic source Organic keywords (SEO) vs Paid (AdWords)
- 63. Detail: Content Top Content
- 64. Detail: Content Top Content
- 65. And More Customizable Tracking
- 66. Detail: Visitors Member loyalty
- 67. Detail: Campaign Knowing event performance by time
- 69. Medium
- 70. Source
- 71. Keyword (optional) http://www.reallusion.com/event/09_war_peace/?utm_ source =facebook&utm_ medium =cpc&utm_ campaign =war_peace&utm_ term =girl
- 72. Detail: Traffic source 3 compulsory and 2 optional variables Campaign Source Medium Keyword AD content war_peace youtube web email ppc (adwords) banner textad buy now facebook download home landing newsletter rc_0901 email buy now download ic_landing (mail subject) ccd_0901 edu_0901 icfu_0901 partnership digitaljuice email Source & Medium MUST be exactly match (all lower-capital) E5, element5, Element5, are all different
- 73. 3-dimension variables Source Medium Campaign Campign= war_peace Banner vs. email Or Facebook vs. Digitaljuice Source=facebook Banner vs. textAD War_peace vs. film_the_world Medium=email RL vs. Digitaljuice Realitycheck vs. CCD NL
- 74. Detail: Traffic source Medium & Source Comparing source= google with different medium Comparing campaign= newsletter with different source
- 75. 2,000 visits 1,000 visits 500 visits 3 pages/visit 4 pages/visit 5 pages/visit USD 3/visit USD 8/visit USD 5/visit Event A Event B Event C Changing focus from Attractive ADs to Effectiveness Messages to the ultimate result, Purchase Detail: Campaign A new method to measure traffic effectiveness vs. linkcount
- 76. Detail: Goals Can be used to track: Form submission
- 78. Member registration (activation page)
- 79. Clicks (e.g. file download attempts)
- 80. and so on
- 81. Detail: Goals Goal conversion: BackStage purchase flow 16.98%
- 82. Detail: Goals Goal conversion: EX member registration 55.8%
- 83. Detail: Goals As a warning
- 84. Use virtual page to track page view as a goal Use java script to create a virtual page view
- 85. Same technique can be used to track Flash events
- 86. Track file downloads (Virtual page view): http://www.google.com/support/analytics/bin/answer.py?answer=55529 Event tracking: http://code.google.com/apis/analytics/docs/gaJS/gaJSApiEventTracking.html Library for tracking flash: http://code.google.com/apis/analytics/docs/tracking/flashTrackingIntro.html Trade-offs: Over-estimate the page views (same ID)
- 87. Create broken links in GA (different ID)
- 88. Can't be view as goals (event tracking)
- 89. Detail: E-commerce BackStage overview
- 90. Detail: E-commerce Popular products
- 91. Detail: E-commerce Popular categories
- 93. Detail: E-commerce Visits/days to purchase
- 94. Cross-referencing other Data Monthly or weekly change in: Traffic vs. download
- 95. Download attempts vs. complete download
- 96. Complete download vs. new member registration
- 97. Some Do's and Don'ts with Site Tracking
- 98. Source: cshome;
- 99. Medium: banner;
- 100. Campaign: topseller Any thing unusual in the following source/medium? If you are not sure, check reference file or ECD Dos: Use Consistent Parameters
- 101. How come a English page PV increased 30 times without update or promotion? A DE page copied the page without change GA code ID before newsletter send-out ID URI Path
- 104. Keep consistent internal page links /crazytalk and /crazytalk.aspx looks no difference in creating an internal link
- 105. But they are different in GA /crazytalk 1,000
- 106. /crazytalk/default.aspx 2,000 (another issue occurred when we were upgrading .asp pages to .aspx ones) Dos: Use Consistent Links
- 107. Don'ts: Forget Putting GA Code All data captured are replying on the GA code
- 108. Wrong data can lead to wrong decision Interpret user behavior incorrectly
- 109. Under/Over-estimate marketing results In most case, GA code are embedded in the top navigation, unless: Event pages that do not have a navigation bar
- 110. Pop-up pages that do not have a navigation bar
- 111. Pages in a frame that share the same URL
- 112. .. And more
- 113. Don'ts: Use Linkcount Typically: From www.siteA.com From search engine Newsletter From site A, PV: 4 Keyword A, PV: 3 rc_1103, PV: 3
- 114. Don'ts: Use Linkcount We can be misled: From www.siteA.com From search engine Newsletter From site A, PV: 2 Direct, PV: 2 Keyword A, PV: 3 rc_1103, PV: 1; bounce Direct, PV: 2
- 115. Unless in AP. Linkcount creates the dilemma: One click generates two page views, or
- 116. Broken link in tracking user's navigation Don'ts: Use Linkcount
- 117. Bounce Rate: visitors hit a page and leave with out a click It's currently not accurate to our site because they might clicked but not tracked: Linkcount, redirecting pages
- 118. Download links in event pages
- 119. External link (e.g. add to cart, YouTube) ( Solution )
- 120. Use Campaign tracking only when it is necessary: Campaign cookie overwrites each other by default, you then can't track the real contribution of each campaign
- 121. The default life span of a campaign cookie is 6-month
- 122. ( Cookie types ) Don'ts: Over-use Campaign Tracking We will start talking Acquisition in 2 weeks
- 123. That's all for today, any question?
- 124. More material if you are interested in Analytics... Google Analytics online training: http://www.google.com/support/conversionuniversity/ Web Analytics: An Hour a Day (book) http://www.amazon.com/Web-Analytics-Hour-Avinash-Kaushik/dp/0470130652 Useful blogs: http://cutroni.com/blog/
- 126. What will we do next week? Week 2: Acquire visitors
- 127. Exercise: come out one thing you want to know about Reallusion site, may be but not limited: An existing figure has been monitored
- 128. A useful goal that should be added
- 129. An KPI we should regularly checked
Editor's Notes
- #4: Acquisition: get quality traffic Conversion: clear guide, convincing message/graphic Retention: encourage repeat purchase
- #5: Some key factors does not apply to Reallusion, e.g. location, gender, delivery, etc Online sales plays an important role to the company Reallusion site is our most profitable channel You can't control your partners or channels, but you can 100% control your web site (e.g. member DB, behavior, interaction)
- #7: Convert concept to the keyword / search term. e.g. 3d animation
- #9: Not enough users want to find 3d animation tool We can't list in the top 2 pages, server down, etc The flow isn't clear, broken link, too slow to load The web page isn't attractive, not enough information, they don't think the product meets their needs They can't find download link or download link don't work Too slow to download, can't download, can't install They don't want to register, they are not happy with the product They can't find the purchase link, price isn't sounding Can't paid, complicated pay flow
- #14: Web logs (almost) no effort to implment Not optimized to analyze business info Web beacons (e.g. 1*1 image) Easy to implement Cross-site issue and less info than Javascript Javascripit tags Browser-side script (e.g. turn off javascript) Not as accurate as web log if not implemented well Packet sniffing Fast and less implemntation for page design Additional H/Sware required
- #16: Web logs (almost) no effort to implment Not optimized to analyze business info Web beacons (e.g. 1*1 image) Easy to implement Cross-site issue and less info than Javascript Javascripit tags Browser-side script (e.g. turn off javascript) Not as accurate as web log if not implemented well Packet sniffing Fast and less implemntation for page design Additional H/Sware required
- #55: Top_1009 is a banner in content store Isn't a standard/understandable source Can't compare result with others
- #57: I recommand skip the default page name.
- #58: One of the disadvantages of using scrip-based tracking is that you get nothing if you did not put the code correctly
- #63: - e.g. if a newletter's cookie been over-written by content store home, then you under-estimate the result of newsletter - 6 month is good for tracking external partner effectiveness if not overwritten