Anonymity of Clickstream data

1
Anonymity of Clickstream data
2

• Traffic from different countries
• People with different intent
• Buyers
• Browsers
• Competitors
• Visitors with no apparent reason
• Students
• Information seekers
• Potential employees
• (employees, crawlers, machine traffic)
• One of the reasons why senior management dont
  trust the data
  

3
Solution?

• SEGMENTATION!

4
Benefits

• Data that pertains to your target audience
• Take action
• Different strategies for different segments
• Triggers for successful behavior
• Market intelligence

5
Segmentation

• Segment based on what?
• Geography, Time, Referrer, Campaign, Content,
  Behavior, Success Visitors, Paid traffic, Natural
  SEO traffic.
  
• What metrics to look at?
• What possible actions to take?
• How this is done?

6
1. Geography
7
Metrics

• Visits / Visitors
• Top content
• Top Products
• Key phrases
• Internal search phrase
• Orders

8
Scenario

• IT company focused on certain countries
• Eliminate data from other regions
• All metrics would be more accurate with narrower
  data
  
• Less skew (example career traffic)

9
Scenario

• Services company looking at new regions
• Search key phrase know what people are looking
  at
  
• Internal search phrase know what people are
  looking specifically from the company
  
• Top Downloads / Pages Which service is
  relevant
  
• Top supporting pages What supporting
  information are they looking for?
  

All this can go as market intelligence to the
sales/marketing team

10
Scenario

• Product company delivering in multiple countries
• Traffic source pie decide where the marketing
  dollars should be spent
  
• Top Viewed services inputs to sales
• Search terms customer intent
• Top viewed supporting pages inputs to sales /
  marketing
  
• Visits to non-English sites decide to start /
  sustain / shut-down.
  
• Download data decision on which collaterals to
  print
  

11
2. Time / Day / Date
12
Metrics

• Visits / Visitors
• Top products viewed
• Top products ordered
• Orders for previous year / month / day
• Email Open ratio / CTR / Conversions
• Conversion ratio for messages (internal ads)

13
Scenario

• E-commerce site
• Message conversion ratio
• What offers and where
• Top products viewed / ordered
• Timing for running PPC ads
• Product display on key landing pages
• Stock
• Email metrics When to send and what to talk
  about
  

14
3. Referrer
15
Metrics

• Visits / visitors (share, trend)
• Conversion (orders / subscriptions / leads)
• Top landing pages
• Top content consumed
• Geography

16
Scenario

• News website
• Engagement Understand which referrer is better
• Content consumed Useful to know what kind of
  traffic is of interest
  
• Geography data Relevant referrers for specific
  regions
  
• Visit trend Predict when the traffic is going to
  happen (cricket)

17
Scenario

• B2B Website / Content Website
• Region data Useful to evaluate PR effectiveness
• Engagement metrics Identify high-value referrers
  and pursue relationship with them
  
• Content consumed links from the landing pages

18
4. Success Visitors
19

• Success Visitors
• Ordered
• Subscribed
• Inquired
• Looked at x number of pages
• Looked at key pages
• Used a particular key phrase to reach
• Very powerful segment to help initiate action

20
Scenario

• E-commerce Website
• Product ordered-Key phrase matrix Useful for
  choosing products / key phrases for PPC
  
• Geography product ordered Choosing regions for
  PPC
  
• Supporting pages viewed Push pages
• Campaign Looking at where to spend money
• Paid key phrases Which terms to bid
• Products ordered Figuring out related products
  to push (through email, landing pages, etc)
  
• Offers Know what to push and where

21
Scenario

• Lead generation website
• Content consumed knowing which pages influenced
• Trigger page For cross-linking
• Search engine key phrases action point for SEO
• Geography Knowing where the customer is
• Search Engine date Choosing key phrase engine
  combination
  

22
Tools
23
Tools
24
To Conclude

• Segmentation options for your tool
• Start looking / reporting segmented data
• Segmenting based on geography / referrer /
  success visitors easy to start
  
• Consider Clicktracks (for analysis and not
  reporting)
  
• Explore multiple profiles and custom variables
  for GA

25
Questions

• For more on this topic, please visit
  www.nabler.com/blog
  
• If you visit, please leave a comment so that I
  know you had come
  
• If you have questions, please post a comment and
  I will try to answer them.
  
• Seby Kallarakkal