- GA360 - enterprise version
- Google Tag Manager
- Requirements
- User must allow first party cookies
- Every page must have embedded JS tracking code
- Properties and views
- account > property > view
- should have at property for each site or platform
- views allow for different filters
- at a minimum should have these views:
- main reporting, testing, raw data
- "Insights" - tells you some interesting things about your data
- Reports
- Audience - who's visiting the site?
- Acquisition - how are users finding the site?
- Behavior - user engagement
- Conversion - if you've set goals
- Metrics and dimensions
- Metrics
- measures data
- e.g. page views, sessions
- Dimensions
- qualitative
- describes data in groups of that data
- e.g. browser, device type
- Scope
- Levels: hit, session, user, product
- metrics and dimensions should apply to same scope
- Graph data over time
- can overlay 2 different metrics for comparison
- can compare 2 time periods with the same metric
- can plot data for individual dimension (e.g. desktop and mobile)
- Using annotations in metric reports
- can be private (note for individual user) or shared
- e.g. when filters are changed, notes about anomalies
- Viewing data
- Acquisition > Channels
- can select two different metrics to compare
- can also view using pivot table
- there is a term cloud button
- Acquisition > Channels
- Metrics
-
View filters
- can be dangerous to use as it can filter out data from the views
-
basic/quick inline filters
- better to use the search bar to filter to get all instances
-
advanced inline filter
- can use reg expressions to search OR by using a pipe, "Matching RegExp"
- e.g. to find traffic only from facebook or youtube: "facebook|youtube"
- Acquisition > ALl traffic > source/medium
- Can filter by primary dimension or also by metric
- can use reg expressions to search OR by using a pipe, "Matching RegExp"
-
Segmentation intro
- intent
- landing page
- type of browser
- mobile vs desktop
- connection speed, screen resolution
- can segment and subsegment
- should think in terms of segments not all users as a whole
- GA has presegmented reports and also customizable options
- Can use secondary dimensions to combine multiple dimensions
- Under "Channels", you can "Add segments"
- Can then create your own segments
- e.g. US male, age 25-34, Windows
- can then compare segment against all users
- Can then create your own segments
- What segments do we want?
-
- active users
- lifetime
- cohort
- audiences
- can create audiences
- audience definitions
- similar to segments but across other properties
- user explorer
- look into individual user behavior
- Benchmark
- to compare to other industry benchmarks
- users flow
-
Audience reports
- overview
- use date range comparison
- not too much for analysis
- demographics and interests
- can break down by gender, age
- can sort by different dimensions (e.g. conversion rate)
- geographic location
- can drill down by location
- can break down by per user (e.g. per session value)
- to see user engagement (e.g. session duration)
- these metrics may not be reliable
- technology
- browser and OS
- can use to surface browser compatibility issues
- can use a "weighted sort" to filter out insignificant results
- use this to optimize site for user needs
- browser and OS
- mobile
- desktop, mobile, tablet
- devices report
- specific brands and devices
- e.g. Apple vs Android
- compare mobile vs desktop
- overview
-
Acquisition reports
- Overview
- channels
- All traffic
- Channels
- Q: Which channel/medium sends you the most traffic?
- Q: Which channel is the most bang for buck?
- organic search - i.e. Google search
- paid search - sponsored search results
- direct - user types in URL or lack of information
- referral - user clicking on a link
- social - social media sites
- affiliates - affiliate marketing efforts
- display - graphical ads (not search ads)
- Source/Medium
- identifies specific sources
- answers "who is the source?"
- Referrals
- to see sites that refer to your site
- add secondary dimension of landing page to see the page where users are coming to
- Campaign tracking
- to track by campaign name from same source and medium
- works by using UTM parameters that are added to the URL
- GA then uses this URL for tracking
- hint: tag your campaigns (e.g. tweets) to get more granular data by building campaign URLs
- Channels
- Behavior
- Q: How are your users engaging?
- Overview
- pageviews: user can view same page multiple times
- unique pageviews: per session
- bounce rate: when entire visit only had one page
- % exit: how many users are leaving the site from that page
- Behavior flow
- where are users dropping off?
- useful if you have content groups
- Site content
- All pages
- idea: click comparison button then select bounce rate to see problem areas by specific page
- can then click that page and see how users got to that page
- find pages with high bounce rates
- idea: click comparison button then select bounce rate to see problem areas by specific page
- Content drilldown
- good for sites with deep structure and hierarchy
- All pages
- Site speed
- measures load time and latency of pages
- Q: what pages load the slowest?
- we can then fix these first
- Click "Technical" stats to get more details
- Check load time of landing pages
- Change primary dimension to landing page
- Check load times by OS
- e.g. Chrome OS vs Android
- Check average load time over time range to look for anomalies
- Also check map overlay to see if there are any issues in specific locations
- speed suggestions
- gives suggestions for how to improve speed
- Events
- any user behavior that you want to track
- to add new events see Advanced course
- need to add JS
- use Google Tag Manager
- Dev guide: https://developers.google.com/analytics/devguides/collection/analyticsjs/events
- categorize events by category, action, label
- event reports
- Conversion reports
- Overview
- can configure 20 goals per view
- need to configure goals to use this
- can assign monetary value to goal
- can also use goal conversion rate, not monetary/revenue
- to find what segments/users are outperforming or underperforming
- Goals
- Goal URLs
- to see the exact location where goal was completed
- Reverse goal path
- use this to find the path user took to achieve goal
- Funnel visualization
- funnel report
- allows you to see where users abandoned the site or the funnel (set milestones)
- funnel report
- Goal flow report
- can drill down to specific parameters
- see more details such as funnel exits
- IDEAS for goals
- anytime someone sees a product page
- time on site
- number of pages viewed
- TODO: Must have goals configured and set up to make use of these reports
- Overview
- Real-time reports
- see instantaneous results
- could be important for sites with content that changes multiple times daily
- can be useful to track interaction during promotions or campaigns
- advanced tip: can use real-time API to publish this data
- Basic configuration
- 100 accounts, 50 properties, 25 views
- In properties, can create audience definitions
- In view section
- TIP: set up goals here
- can do user management at account/property/view level
- IMPORTANT: Set up goals
- View > Goal
- can use templates for ideas
- e.g. successfully complete contact us page, i.e. form completion
- e.g. duration of 5 min or longer
- e.g. pages per session
- e.g. goals based on event interactions
- if goal has clear steps, set up a funnel
- if goal has a $ value, add value
- View > Goal
- Overview