Building a Web Analytics Framework that Works

Building a Web Analytics Framework

that works!

Pradeep  Chopra  CEO,  Digital  Vidya  

Aloke  Bajpai  CEO,  iXiGO  

3 Fundamental Questions

• Why? (Objectives)

• What? (Metrics)

• How? (Tools, Techniques, Dashboards…)

Lets Learn from Real Examples!

1. E-commerce Tracking

Source   Visits   Trials   Transac9ons  

SEO   33413   481   79  

Direct   7041   193   82  

Google  Adwords   1664   204   8  

Referral  (SiteA)   732   28   6  

Referral  (SiteB)   384   11   4  

2. UTM Tracking Source   Visits   Sign-­‐Ups  (%)   Bounce  Rate  

Source  A   150   7   65  

Source  B   35   9   45  

Source  C   46   12   48  

Source  D   95   8   56  

Source  E   140   5   67  

digitalvidya.com?utm_source=tb;utm_medium=email&utm_campaign=jun-­‐12    

3. Revenue Modeling Monthly   India   Asia   Total  

Revenue  (Rs)   800,000   400,000   12,00,000  

No  of  ParUcipants   100   50   150  

No  of  Leads   2000   1000   3000  

No  of  Visitors   20000   10000   30000  

Cost  of  Adv  (Rs)   160,000   80,000   240,000  

Daily  Calls   150   75   225  

No  of  Callers   2.5   1.25   3.75  

The  odds  of  contac9ng  a  lead  if  called  in  5  minutes  versus  30  minutes  drop  100  9mes.      The  odds  of  qualifying  a  lead  if  called  in  5  minutes  versus  30  minutes  drop  21  9mes.  

4. Top 10 Metrics for CEOs 1.  Daily Visits 2.  Daily Unique Visitors 3.  Unique Pageviews for important pages 4.  Multi-Channel Conversion Funnel 5.  Bounce Rate 6.  Direct Visits (including brand keyword visits) 7.  SEO Visits + Keywords Base 8.  Referring Sites (Deep-Dive) 9.  No. of server errors (404s / 500s) 10.  Qualitative Feedback/Sentiment Report

5. Using Advanced Segments & Custom Reports

•  Advanced Segments –  Segment a section of visitors for deep-dives (e.g. only visitors

who viewed 3+ pages, only visitors who visited a particular page, only visitors who landed on a particular page

–  Useful for understanding behaviour for various source/campaign segments

•  Custom Reporting –  Very useful for deep analysis of paid search, SEO, Browsers,

Device type data –  Choose metrics (columns) –  Choose dimensions (drilldowns) –  Choose filters (for any metric)

•  Most of the time you want to use both of above •  Please mind the sampling gap !

Examples

6. Qualitative Analytics •  Measuring Happiness •  How do my users behave ? What do they spend

their time looking at ? Why do they do what they do ?

•  User Interaction / Behaviour (Clicktale/UserFly) •  Net Promoter Score •  Surveys (on-site / off-site) •  Social actions on the website (Likes/Shares) •  Social Media Mentions / Sentiments

NPS Example

NPS  =  46  

Qualitative Analytics

7. The HEART Framework •  Happiness

–  User attitude, qualitative metrics, perceptions •  Engagement

–  Behavioral signals, Depth of Interaction, Frequency, Clicks •  Adoption

–  % Users who adopt new product / feature, “Get” the product •  Retention

–  Returning users / Churn •  Task-Action

–  Efficiency, Error Rates, Time Taken, % who complete a specific goal –  Specific metrics / signals critical for your product’s success

HEART Metrics in Action

Thank you!

[email protected] | [email protected]