Welcome to the Webinar Please join and post your questions in this chatter group https://sfdc.co/ea-ed-cert-fp Einstein Analytics & Discovery Consultant Certification Fast Path Ayhan Sahin Senior Manager , Partner Field Enablement Community Cloud & Einstein Divya Alavarthi Senior Manager, Partner Field Enablement Platform Vikas Roy Director, Partner Field Enablement Community Cloud & Einstein Agenda Webinar 3 ● Objective ● Certification breakdown ○ ○ Section 5 : Einstein Analytics - Dashboard Implementation Section 6 : Einstein Discovery - Story Design ● Q&A Objective Give you some guidance on where to find the content, the knowledge, share tip and tricks to start your journey with Einstein Analytics & Einstein Discovery and successfully pass the exam in a month. Certification Breakdown Certification Exam Sections Section Title Weighting % Nbr of Q 1 Data Layer 24% 15 2 Security 11% 7 3 Admin 9% 6 4 Dashboard Design 19% 11 5 Dashboard Implementation 18% 10 6 Discovery Story Design 19% 11 Section 5 - Analytics Dashboard Implementation Sections and Objectives Weighting % Section 5 Analytics Dashboard Implementation 18% 5.1 Given business requirements, define lens visualizations such as charts to use and dimensions and measures to display. 23% 5.2 Given customer business requirements, develop selection and results bindings with static steps. 22% 5.3 Given business expectations, create time series analysis. 10% 5.4 Given customer requirements, develop dynamic calculations using compare tables. 22% 5.5 Given business requirements that are beyond the standard UI, use SAQL to build lenses, configure joins, or connect data sources. 23% Bindings In a dashboard you usually have several charts (widgets). How do you link them to each other? You control the interactions by binding queries to each other. There are two types of interactions: selection binding and results binding. The selection or results of one query triggers updates in other queries in the dashboard. Use cases Bindings - Examples ● Binding by selection of a grouping ● Select Consumer and Fin Svcs Bindings - Examples ● Binding by selection of a grouping ● Select Consumer and Fin Svcs Bindings - Examples ● Binding by selection of a grouping ● Binding by selection of a measure Bindings - Examples ● Binding by multiple selection Bindings - Syntax "{{}}" "{{cell(Amount_3.result,\"avg_Amount\" "{{cell(Amount_3.result).asString}}" "{{cell(Amount_3.result)}}" "{{Amount_3.result}}" "{{.result}}" "{{cell(Amount_3.result, 0, \"avg_Amount\" ).asString}}" ).asString()}}" "{{column(Amount_3.result, ["avg_Amount\"] ).asObject()}}" "{{cell(Step_1.selection, 0, \"value\").asString()}}" or "{{column(Step_1.selection, [\"value\"]).asObject()}}" - A parameter that compiles/becomes a value at run time - This value can be a number (ex. 5000), a color (ex. #E6ECF2 ), a dimension name (ex. AccountId.Type), a measure (ex. Amount), etc… and is read/received from another query on the dashboard - It also has a specific format so that it “fits” correctly within the dashboard code (json) ex. “#E6ECF2” or [“AccountId.Type”] or [“sum”,”Amount”] All about bindings by Rikke Hovgaard Time Series ● Using existing data to predict the future. 1 result = timeseries resultSet generate (measure1 as fmeasure1 [, measure2 as fmeasure2 ...]) with (parameters); ○ Example : how many tourists will visit next year? Suppose that you run a chain of retail stores, and the number of tourists in your city affect your sales. Use timeseries to predict how many tourists will come to your city next year 11 q = timeseries q generate 'sum_NumTourist' as Tourists with (length=12, dateCols=('Visit_Year','Visit_Month', "Y-M")); All about Time Series by Rikke Hovgaard Compare Table ● View measures side by side, and perform math across the table’s columns and rows. Use string values to create labels, concatenate dimension values, provide simple buckets, or add image URLs. All about Compare Table by Rikke Hovgaard What is SAQL? What is SAQL? The Salesforce Analytics Query Language (SAQL) is a run-time query language that enables ad-hoc analysis of datasets. A SAQL query consists of a sequence of statements that are made up of keywords (such as filter, group, and order), identifiers, literals, or special characters. It’s JSON based and PIGQL familiar. A SAQL query loads an input dataset, operates on it, and outputs the results. When to use SAQL? Use SAQL when you need to do custom calculations across multiple datasets, advanced data manipulations/transformations on the fly, cogrouping (joining) data from different datasets, different aggregate levels on one chart, joining multiple streams of data, complex filters, and in conjunction with bindings for dynamic linking on charts and related, etc... Simple Use Case: a formula based on 2 metrics or more from different datasets on the SAME chart q = load ”My_Dataset"; q = group q by 'Division_Name'; q = foreach q generate 'Division_Name' as 'Division_Name', sum(’Amount’)/sum(’Quantity’) as 'sum_Amt/Qty'; q = order q by 'sum_Amt/Qty' desc; All about SAQL by Rikke Hovgaard Tips / FAQ ● If data are missing in a dataset, you can still use SAQL with the “fill” statement ○ ● https://developer.salesforce.com/docs/atlas.en-us.bi_dev_guide_saql.meta/bi_dev_guide_saql/bi_saql_statement_fill.htm Change null value with the coalesce()function ○ https://developer.salesforce.com/docs/atlas.en-us.bi_dev_guide_saql.meta/bi_dev_guide_saql/bi_saql_functions_coalesc e.htm#! Einstein Analytics/Discovery Consultant Path Section 5 – Dashboard Implementation – Helpful Links 5.1 – Visualizations https://help.salesforce.com/articleView?id=bi_visualize.htm&type=5 https://help.salesforce.com/articleView?id=bi_chart_reference_properties.htm&type=5 5.2 – Bindings https://developer.salesforce.com/docs/atlas.en-us.bi_dev_guide_bindings.meta/bi_dev_guide_bindings/bi_dbjson_bindings.htm https://help.salesforce.com/articleView?id=bi_dashboard_data_source_connections.htm&type=0 5.3 – Time Series https://developer.salesforce.com/docs/atlas.en-us.218.0.bi_dev_guide_saql.meta/bi_dev_guide_saql /bi_saql_statement_timeseries.htm https://help.salesforce.com/articleView?id=bi_limitations.htm&type=5 5.4 – Compare Tables https://help.salesforce.com/articleView?id=bi_explorer_compare_table_nav.htm&type=5 https://help.salesforce.com/articleView?id=bi_explorer_compare_table.htm&type=5 5.5 – SAQL https://developer.salesforce.com/docs/atlas.en-us.bi_dev_guide_saql.meta/bi_dev_guide_saql/bi_saql_functions_coalesce.htm?search_text =coal https://developer.salesforce.com/docs/atlas.en-us.bi_dev_guide_saql.meta/bi_dev_guide_saql/bi_saql_statement_group.htm# https://developer.salesforce.com/docs/atlas.en-us.bi_dev_guide_saql.meta/bi_dev_guide_saql/bi_saql_statement_fill.htm?search_text=fill Einstein Analytics/Discovery Consultant Path Section 5 – Dashboard Implementation – Links from the Exam guide ● ● ● ● ● ● ● ● ● ● ● Explore and Visualize Your Data in Einstein Analytics Build Einstein Analytics Dashboards Progressive Disclosure (Loading) Embed and Customize Einstein Analytics Analytics Bindings Developer Guide Analytics REST Query Resource Analytics SAQL Reference Wave Funnel Powered by Custom SAQL Timeseries SAQL Statement Analytics Extended Metadata (XMD) Reference Run Your Dashboards Faster with the Dashboard Inspector Section 6 - Einstein Discovery Story Design Sections and Objectives Section 6 Weighting % Einstein Discovery Story Design 19% 6.1 Given a dataset, use Einstein Discovery to prepare data for story output by assessing data and adjusting outputs. 35% 6.2 Given initial customer expectations, analyze the story results and determine suggested improvements that can be presented to the customer. 35% 6.3 Given derived results and insights, adjust data parameters, add/remove data, and rerun story as needed. 20% 6.4 Describe the process to perform writebacks to Salesforce objects. 10% Einstein Discovery Einstein Discovery is AI-Powered analytics that enables business users to automatically discover relevant patterns based on their data - without having to build sophisticated data models. Einstein Analytics is a visualisation tool to uncover details. Einstein Analytics allows you to explore all your data quickly and easily by providing AI-powered advanced analytics. Einstein Discovery Automated Analytics - Analyze millions of data combinations in minutes Unbiased Insights - Understand what happened, why it happened, what could happen, and what to do about it Narrative Explanations - Natural language-based insights and stories exported to Salesforce or Microsoft Office Recommended Actions - Take action, stay on top of changes, and iterate Einstein Discovery Automated Insights Narrative Explanations Data Prep Recommendat ions Einstein Discovery 1000s of combinations 1000s of charts 1000s of statistic runs Feedback Loop Operationalizi ng Results Einstein Discovery - Creating Stories ● Create Stories directly from the dataset ● Manage Stories ● Deploy to Salesforce Consider datasets loaded from native and/or external data sources, csv, enriched using dataflows and recipes. Einstein Discovery - Integration Examples Einstein Discovery - Integration Examples Einstein Discovery - Integration Examples Embedded Analytics Embedded Predictive, Explanatory and Prescriptive Write back 1 - Install the Managed Package Your Salesforce admin must install the Salesforce managed package to enable the Writeback feature. 2 - Create Custom Fields in Salesforce to Display Recommendations You create custom fields in Salesforce to display the outcome, explanation, and prescription information imported from Einstein Discovery. 3 - Add an Einstein Discovery Section to the Object’s Page Layout Add the custom fields to the object’s page layout. 4 - Enable Salesforce Writeback Before you can use the Writeback feature, you must enable it. 5 - Connect Einstein Discovery to Your Custom Fields To import the recommendations, you connect your custom fields with Einstein Discovery. 6 - Create an Apex Trigger The Apex trigger fetches the Einstein Discovery predictions when a record is inserted or updated. The Apex trigger is stored under the object with which it is associated. 7 - Troubleshoot the Integration To troubleshoot the integration, you can use the Developer Console. The Developer Console is an integrated development environment with a collection of tools for creating, debugging, and testing applications in your Salesforce org. https://appexchange.salesforce.com/appxListingDetail?listingId=a0N3A00000FOm9nUAD Write Back Write Back Tips / FAQ ● Two variables that behave the same can give similar outcome but they might be differences from the business perspective ● As a best practice, for more accurate model or outcome always remove data bias ● If you have 3 stories, segment your data first before deploying the appropriate model for each segment in Einstein Analytics Einstein Analytics/Discovery Consultant Path Section 6 – Discovery Story Design – Helpful Links 6.1 – Data Prep https://help.salesforce.com/articleView?id=bi_edd_prep.htm&type=0 https://help.salesforce.com/articleView?id=bi_edd_model_metrics.htm&type=5 https://help.salesforce.com/articleView?id=bi_edd_create.htm&type=0 https://help.salesforce.com/articleView?id=bi_edd_prep.htm&type=5 6.2 – Improvements https://help.salesforce.com/articleView?id=bi_edd_story_explore.htm&type=5 https://help.salesforce.com/articleView?id=bi_edd_story_graph_whatcould.htm&type=5 https://help.salesforce.com/articleView?id=bi_edd_story_interface.htm&type=5 6.3 – Adjust Parameters https://help.salesforce.com/articleView?id=bi_edd_prep.htm&type=5 https://releasenotes.docs.salesforce.com/en-us/winter19/release-notes/rn_bi_edd_model_metrics.htm 6.4 – Writeback https://help.salesforce.com/articleView?id=bi_edd_wb_native.htm&type=5 Einstein Analytics/Discovery Consultant Path Section 6 – Discovery Story Design – Links from the Exam guide ● ● ● ● ● ● Explore Stories View Model Metrics Einstein Discovery Limits Optimize Data for Predictive Analytics Display Einstein Discovery Predictions in a Salesforce Object Improve Your Einstein Discovery Models by Investigating Their Metrics and Performance Q&A GOOD LUCK!