Determine if the timeseries required parameter, prediction interval, should be 80 or 95.

Investigate which prediction model is best for the data available.

Validate that there is enough data in the dataset to make a good prediction.

Include the fill SAQL statement to fill any missing dates.

Aggregate by a measure

Count rows by measure you want to visualize.

Filter to narrow down your results.

Grouping by a dimesnsion

Display the results graphically

Add the system permission "Manage Analytics Custom Maps" to the permission set used.

Enable "Custom maps with geoJSON" in the analytics settings.

Upload the geoJSON via the API because it is not a function in the UI.

Contact Salesforce support and request to get custom maps and geoJSON enabled for the org.

Use the Show Summary option in the compare table.

Add a formula field to calculate the summary.

Use a SAQL query since the summary is not available in standard interface.

Create a computeExpression field in the Dataflow to calculate the summary.

5,000

1,000, but more can be purchased

75, but more can be purchased

500, but more can be purchased

flatten

compute Relative

compute Expression

append

q = timeseries qgenerate 'sum_Sales' as Forecasted_Sales' with (dateCols=('Date_Year, Date_Quarter', "Y-Q"), partition=Region', ignoreLast=true);

q = timeseries q generate 'sum_Sales' as Forecasted_Sales' with (Iength = 12, dateCols=('Date_Year', 'Date_Month". "Y-M"),partition='Region');

q = timeseries q generate 'sum_Sales' as Torecasted_Sales' with (dateCols=('Date_Year', Date_Quarter', "Y-Q"), partition='Region', seasonality=4);

q = timeseries q generate 'sum_Sales' as Torecasted_Sales' with (dateCols=('Date_Year', Date_Quarter', "Y-Q"), partition='Region');

Create a trigger on Opportunity and use the Salesforce External Connector to get predictions from Einstein Discovery.

Create an Apex batch on Opportunity and use the REST API to get predictions from Einstein Discovery.

Create a trigger on Opportunity and install the Einstein Discovery Writeback managed package from the AppExchange.

Create a trigger on Opportunity and use the REST API to get predictions from Einstein Discovery.

Create separate datasets for attainment and commission and apply security predicates and/or sharing inheritance.

Apply a security predicate on the existing single dataset.

Use sharing inheritance.

Add the sales organization to the attainment dataset access list.

It's a read-onlyfile with all the user documentation for a template.

It's the boss of the template with all the information needed to create the app.

It makes sure that the app includes all the right Salesforce data.

It's automatically populated every time you create an app from the template.

No. Einstein Discovery is impervious to collinearity, so the story and subsequent model will be fine.

No. Although it is ideal to eliminate collinearity as soon as possible, Einstein will give a warning post-build and the ridge regression will prevent collinearity from over-fitting.

yes. If all collinear variables are not excluded, the model will over-fit and not make any sense.

Yes. If the collinear variables are not removed, the Einstein Discovery model build will fail.

Ant Migration Tool

Analytics External Data API

Analytics REST API

Change sets

Analytics dashboard connector

Modify dataflows to exclude unused fields and include a filter to extract relevant data.

Scale down the number of datasets to help improve data sync (replication) performance.

Start overagain with new dataflows that exclude unused fields with a filter to extract relevant data.

Update the existing data sync (replication) process by removing unused fields with a filter to extract relevant data.

Use the Salesforce SOAP interface.

Use the Analytics External Data API.

Use Oracle database connector.

Load the data as a CSV file.

Customize tooltip

Chart Markers

XMD

Compare Table

Upload External Data to Analytics

Manage Analytics Templated Apps

EditAnalytics Dataflows

Edit Dataset Recipes

Use a dataflow to rename the stages with a preceding number.

Use a dashboard JSON to rename stages.

Use a dataflow to sort the stages by their internal name.

Use a dataset XMD to rename stages.

Create and assign Salesforce permission sets according to the three types of roles defined.

Assign the app permissions"viewer/' "editor," and "manager" to the three types of roles defined.

Create and assign three new Salesforce profiles according to the three types of roles defined.

Assign "Einstein Analytics Explorer" licenses to users that should only view thedashboard, and assign "Einstein Analytics Developer" licenses to users that should be able to create datasets and dashboards.

The model isvery strong. A GINI coefficient of .535 shows that this model is very effective.

The model is pretty good; the accuracy rating of .7597 means we can predict both wins and losses at a fairly high rate.

The strength of the model cannot be determined with the metrics shown.

The model is very weak and doesn't provide useful predictions due to the low threshold.

Bar chart

Flat gauge

Pivot table

Pie chart

10,000

50,000

1,000

20,000

Use the Analytics External Data API and leverage the "query" resource.

Export the dataset as an XLS file and use the Einstein Analytics Connector for Excel.

Use the Analytics REST API and leverage the -query" resource.

Export the dataset as a CSV file and load it to an external database.

Third-order analysis

Object-oriented analysis

Second-order analysis

First-order analysis

Spectrum analysis

Create one application for each department and use roles to share applications.

Create one application for each department, put common datasets in the shared app, and use profiles to share applications.

Create one application for each department, put common datasets in the shared app, and use roles to share applications.

Create one application for each department, put common datasets in the shared app, and use permissionsets to share applications.

Predicting outcomes that you don't actually have the right data for

Exploring your existing data to see what already happened

Drilling down into the underlying reasons behind a prediction

Choosing between all possible outcomes for asingle variable

An account of imaginary or real people and events told for entertainment

A collection of graphs, statistical analysis, and text that helps you understand the relationships between multiple outcome variables and the factors that influence those variables

A collection of graphs, statistical analysis, andtext that proves exactly what caused an outcome

A collection of graphs, statistical analysis, and text that helps you understand the relationships between a single outcome variable and the factors that influence that variable