KnowledgeShop

Learn & Share

Web Analytics

Concepts

  • What is a Web Tag?
    • A web tag is a 1x1 pixel-sized image beacon or JavaScript snippet that is embedded into the code of a web page to capture and report data. A web tag is triggered whenever a predetermined action occurs on a specified platform in the web. It captures the data and sends the information back to your preferred data collection system. A web tag is usually implemented to: track online user behavior for marketing analytics, instruct web browsers and implemented technologies to collect data, to set cookies and to integrate third-party content.

Web Analytics 1.0

  • Web analytics evolution
    • Web server logs parsing -> WebTrends
    • Javascript Tags
    • Site overlay or Click Density
      • display where visitors clicked
      • segments based on where users came from
  • Traditional Analytics KPI Metrics
    • Page views
      • frustrating navigation means more page view, but no buy
      • better naviation means faster decision, and yet no buy
    • Hits
      • Pages with images & media has more hits to the server
      • With CDNs, servers receive lesser hits
      • more hits != more visitors
    • Top Exit Pages
      • Top Exit Pages does not mean pages are suboptimal or product is bad
      • Exit rates doesn’t tell content is good or bad
    • Website Engagement
      • Engagement = # of sessions / # of visitors
      • Repeated visitors means they can’t find what they want or the site is good?
    • Visitor screen resolution
      • Seldom this data changes. Instead of wasting time and money capturing this, use Forrestor Research/Gartner data

Traditional KPI can’t accommodate for strategic business goals. We now have KIA (Key Insight Analysis)

  • KIA (Key Insight Analysis)
    • Click Density Analysis
      • using the site overlay feature, it helps you walk in the shoes of your customer
      • segment your traffic and personalize content differently for visitors from Google vs. Facebook, increasing the customer engagement
    • Visitor Primary Purpose
      • ask customers why they are visiting your site
      • run a survey, do phone interviews, etc.
    • Task Completion Rates
      • Page views != users found what they were looking for
      • Do survey, lab usability, create tests, …
    • Segmented Visitor Trends
      • old world embedded attributes in JS tags
      • new tools = Click Tracks, Visual Sciences
      • Reports on “Average time on site”, “Top search keywords”, “popular content”- based on real segmentation of data.
    • Multichannel Impact Analysis
      • Measuring the impact of non-web channels (like TV, newspaper) on your website
      • e.g. how many people use your site but buy your product via retail or phone channel?

Web Analytics 2.0

Web Analytics 2.0 Multiplicity
  • Quantitative, Qualitative, Competetive analytics
  • Paradox of Data
    • Lack of data -> cannot make complete decisions
    • Lots of data -> less insights

Multiplicity

  • one tool from each layer/pillar for better insights
  1. Clickstream data (WHAT)
    • click-level data
    • visits, visitors, source, time on site, page views, etc.
    • great at answering “WHAT”, but not “WHY”
    • e.g, tools like WebTrends, Google Analytics
  2. Multiple Outcome Analysis (HOW MUCH)
    • measuring outcome means connecting customer behavior to the bottom line of the company
    • outcome types of a webiste
      • increased revenue
      • reduced cost
      • improve customer satisfaction/loyalty
  3. Experimentation and Testing (WHY)
    • Launch fast, fail/succeed fast Experiment or Die
    • A/B and multivariate testing
    • Tools: Google Website Optimizer, Adobe Analytics Target
  4. Voice of Customer (WHY)
    • What customers wanted, but did not find
    • e.g., surveys, lab usability testing, remote usability testing, card sorts, etc.
  5. Competitive Intelligence (WHAT ELSE)
    • compare your performance with your competitors
    • gether info on your direct and indirect competitors

Knowing how you are performing is good; Knowing how your competitors are performing is priceless.

  • Maxamine/ObservePoint
    • missing JavaScript tags, duplicative content, security & privacy compliance, black holes not crawled by search engine, broken links/forms.
  • Coradiant
    • Why conversion rates are low? Pages are slow or missing?

References

  • Books
    • Web Analytics - An hour a day - Avinash Kaushik
    • Web Analytics 2.0 - Avinash Kaushik