certified Developer Certification Tutorial # 2: Analytics as Service

This blog post is part of the certified Developer Certification tutorial series and this is the second tutorial titled ‘Analytics as Service’. The following is the list of tutorial series.


II. Analytics as a Service:

1. Introduction to Reports

  • Reports can include both standard & custom objects
  • two users with different access levels see different data with the same report
  • objects & fields are available immediately after they get created
  • reports run in real-time data
  • reports are saved in folder, report security is determined by folder where report is stored.
  • System administrator and users with ‘Manage public reports’ permissions can manage the folders
  • Custom Report
    • 3 panes – fields, fitlers & preview
    • 4 types of reports:
      • tabular – simple listing of data
      • summary – sorting and sub totalling
      • matrix – summarizes data in grid (to compare related totals)
      • joined – can contain data from multiple standard or custom report types
    • can customize by adding more columns, but cannot deselect action & name columns
    • the way the type of report to be created can be identified by:
      • Select Data
      • Select Report Type
      • Select Summaries
      • Select Groupings
      • Select Filters
    • scheduled reports run at user’s timezone who scheduled it
    • to run a report, the report must be in a public folder and the user has to have access to it
    • Hide Details – will show only summary rows, Show Details will show all the data
    • report shows only upto 2000 reports, beyond that the user can export to excel
    • Printable view – exports data to excel with the report formatting
    • Export Details – exports raw data to excel (no report format will be retained)

2. Introduction to Dashboards

  • Dashboard – contains components based on report or chart, can use VF pages to present data, shows data as of the last time it refreshed, can be refreshed manually
  • dashboard can have upto 20 components
  • can choose either 2 or 3 columns
  • dashboard components
    • chart – good for comparisons, is based on report data and the field which you summarize the report is the field that you see on the dashboard
    • table – shows the top/bottom n records, can be used to display totals also, can include upto 4 columns
    • guage – shows progress towards goals, display percentage or total
    • metric – shows single number (grand total)
    • visual force page – pulls data from another source
  • can follow individual components through dashboard alerts and snapshots
  • can post component’s snapshot to dashboard feed to chatter or dashboard feed
  • chart types
    • vertical column – summary report with single grouping. variations: vertical col: grouped, stacked, stacked 200%, good for showing dates
    • horizontal bar – good for summary with single grouping. horizontal grouped, stacked, stacked 200%
    • line – specially used to show data over time
    • donut chart – good for multiple groupings with total amount
    • funnel – good for datasets for multiple groupings in ordered sets and want to show the proportions among them
    • pie – multiple groupings and want to show proportion of single value for each groupings against total
    • scatter – illustrate degree of correlation between two axis
    • combination charts – that includes different sets of data
  • dashboard security
    • dashboard folder – controls who sees the folder
    • running user – controls what data is displayed on the dashboard
  • clicking a dashboard component takes users to underlying source report, filtered source report or another url that the dashboard creator specifies
  • can drill down to single record, but user will see that the security model will allow to see
  • report access is determined by folders
  • it is possible to see a dashboard component, but not the underlying report
  • Dashboard filters
    • let users chose which data to chose
    • each filter based on single field, can specify upto 10 filter options
    • enables users to view different subsets of data on the same dashboard
    • each dashboard can have upto 3 filters
    • filters can be created on date, datetime, currency, pickup, lookup and text fields
    • filters cannot be added with visual force components, scontrol conponents
    • cannot use filters for bucket fields, filter components cannot be followed on chatter
    • scheduling or emailing a filtered dashboard will return unfiltered data
  • Dashboard display info as of date
  • can be refreshed daily, weekly, monthly

3. Custom Report Types

  • Standard Report Types
    • created, when an object is created, allows to report relationships between objects are created (similar to inner joins)
    • the report types cannot be modified
  • Custom Report Types (CRT)
  • unique templates for creating reports, created by administrators or users with ‘Manage Custom Report Types’
  • can include standard or custom objects, allows users to select the objects and fields that should be related for reporting purposes
  • Enable the creation of “with or without” reports
  • Three steps to create a report based on CRT:
    • select the primary object
    • select related records from other objects (optional)
    • can chose upto 4 objects, chose “with or without” for each related object (similar to outer join)
    • Add fields related via lookup (optional)
  • traverse multiple objects
  • use lookups to join other objects
  • can go upto 4 levels deep, supports many to many relationships
  • primary object cannot be changed after the CRT is created
  • there is a limit on # of CRTs  (depends on SF edition)
  • counting rows
    • standard reports show all rows from both the objects (positions & job applications)
    • what if you want to see only unique rows so that within a report of both job applications & positions, you could see how many job apps you have or how many positions you have.
    • in this case, need to create a report that uses custom field to  count the # of records
    • two step process
      • create a new formula field
      • run a report that utilizes this new field
  • to create custom summary formula: 3 steps
    • create formula fields to count the number of records for each object
    • create custom summary formula to calculate the ratio between the number of records for each object
    • optional: create a dashboard component that will show the ratio between the two numbers
  • Bucketing – helps qualifying the data (for e.g. Not qualified, Qualified, Highly Qualified – for candidates)
    • create a new formula field to group the information
    • create or update a report that utilizes the new field
    • create or update a dashboard component using the new or updated report

4. Analytic Snapshots

  • provides trending, can be scheduled
  • schedule the report and capture the data in custom object and report on the data in the custom object to view historical data and analyze trends
  • can be created in 3 steps:
    • create source report (tab/summary)
    • create the target object (custom). Fields in target object should have same type as in the source object that included in the source report
    • setup the analytic snapshot
    • select the source report, select the target object, map the fields on the report to the fields on the custom object and schedule the frequency for taking the snapshot

5.Going Beyond Salesforce Reports

  • exception reports are available in a few places, but not everywhere
  • no reporting on multiple related lists (using crt, you can include custom objects that are hierarchically related
  • there is only one layout or UI for the reports
  • Salesforce reports don’t provide trending capabilities beyond analytic snapshots
  • Salesforce reports provide limited analysis of what changed bet two dates
  • can’t include data in sf report from another source
  • RaaS -that uses Salesforce as a source – then reports on the data by running multiple queries
  • BIaaS – copies all data to local repository, then runs queries off of the copy of the data  (data repository is provided as a service)
  • Data Warehousing – similary to BIaaS, but you own the servers

Tagged: , , , ,

3 thoughts on “ certified Developer Certification Tutorial # 2: Analytics as Service

  1. Jeff Rogers March 8, 2013 at 12:29 am Reply

    Hi Hari,

    Great blog series you have here! I noticed that the link to the Data Management blog is broken. If you could fix the link it would be much appreciated! Thanks!


  2. Liva SEO May 13, 2014 at 6:35 am Reply

    HI Hari,
    It’s nice content, really i appreciate. Thanks for you great information. Salesforce Services in India

  3. Sreekanth G September 11, 2014 at 11:24 am Reply

    Hi Hari,

    Thanks a lot for your efforts in sharing such valuable information. Much appreciated..!!!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: