50 CRM Analytics and Einstein Discovery Consultant Practice Questions: Question Bank 2025

A consultant needs to create a dataset that combines Opportunity data from Salesforce with a CSV file containing quarterly target quotas by Region. The combined data will be used in dashboards. Which approach is recommended?

A dashboard consumer reports that a chart is showing totals that are much higher than expected after a consultant joined Accounts to Opportunities in a recipe. What is the most likely cause?

A consultant is designing an executive dashboard to monitor pipeline health. Executives want to quickly identify outliers and drill into details without leaving the dashboard. Which design approach best meets this requirement?

A consultant needs a dashboard to show performance trends over time and compare current period to the previous period for the same metric. Which visualization is typically the best fit?

A manager wants a dashboard to show Top 10 sales reps by Closed Won amount, but the ranking must update when users apply filters (e.g., Region, Product Family). How should the consultant configure this?

A consultant is deploying an Einstein Discovery model to explain and improve Case Resolution Time. Business stakeholders want prescriptive guidance on what changes would most improve outcomes for specific cases. Which Einstein Discovery feature addresses this need?

A consultant is troubleshooting an Einstein Discovery model that performs well during training but poorly on new data. Which issue is the most likely root cause?

A company needs to restrict CRM Analytics dashboard data so that sales reps only see records for their assigned territory, while managers see all territories in their region. What is the best approach?

A consultant wants to use Einstein Discovery in a Salesforce record page to score opportunities in real time and write the prediction and top drivers back to fields for downstream automation. Which architecture best supports this requirement?

A large dataset is refreshed nightly. After adding multiple calculated fields and a complex multi-join transformation, refresh time exceeds the available window. Which optimization is most appropriate?

A business user wants to combine two dashboard steps so they can be filtered together and used as a single input for a third step. What should the consultant recommend?

An administrator needs to prevent users from downloading dashboard data as CSV while still allowing them to view and interact with the dashboard. Where should this be controlled?

A consultant is asked to create a dashboard that lets users quickly compare performance against a goal and immediately spot underperforming regions. Which visualization is the best fit?

A dataset in CRM Analytics contains a field named "Close Date" with spaces. A user wants to reference it in a formula in a dashboard step. What is the recommended approach?

A CRM Analytics app uses a single dataset shared across multiple departments. Each department should only see its own records based on a "Department" field. The security rule must update automatically as users move departments. What should be implemented?

An analytics team wants a dashboard to show "Period-over-Period" growth (e.g., this month vs. last month) while allowing users to change the date range using a global date filter. What design pattern best supports this?

A consultant builds an Einstein Discovery model to predict Opportunity win probability. During evaluation, the model shows strong overall accuracy but significantly worse performance for smaller regions with fewer records. What is the best next step?

A dataflow fails intermittently with a query timeout when extracting a very large object from Salesforce. Which change is most likely to improve reliability?

A consultant needs to deploy a CRM Analytics app from a sandbox to production. The app includes dashboards, datasets, and bindings that reference dataset IDs. After deployment, some widgets show errors because IDs differ. What is the best practice to avoid environment-specific references?

An organization wants to write Einstein Discovery predictions back to Salesforce fields on Opportunity nightly. Security requires that the writeback respects field-level security and only updates records the integration user can edit. What architecture should be used?

A CRM Analytics dashboard includes a binding that filters a step by Account Industry. Users report the filter sometimes returns blank results even though matching records exist. The data model contains both an Industry field on Account and an Industry field on a related custom object. What is the most likely cause?

An analyst needs to combine daily website traffic (CSV) with Salesforce Opportunity data in CRM Analytics. The website file has a Date column and a Region column, but no Salesforce IDs. The requirement is to trend traffic alongside pipeline by region and day. Which approach is recommended?

A consultant is configuring row-level security for a shared "Global Sales" app. The requirement is: users can only see Opportunities for their Region, but executives can see all Regions. What is the best way to implement this in CRM Analytics?

A dashboard contains a compare table and a chart that must always show the same top 10 products by Revenue for the selected date range. Users complain that the table and chart show different top 10 lists. What should the consultant do?

An Einstein Discovery model was created to predict Opportunity win probability. After deployment, sales users say the predictions are not available on some Opportunity records. Which configuration is most likely missing?

A recipe joins a large Orders dataset to a smaller Products dataset. After an update, row counts in the output dataset dramatically increase beyond the number of Orders. What is the most likely cause?

A consultant needs to ensure a KPI card displays a month-to-date value that always respects the viewer’s locale and fiscal month settings. Which approach is best?

An Einstein Discovery model shows strong accuracy metrics, but business stakeholders notice it recommends actions that are not feasible (for example, increasing discount beyond policy). What should the consultant do to make recommendations actionable?

A company uses multiple connected apps with different audiences. They must ensure that a sensitive dataset can be used by a specific dashboard but cannot be discovered or reused in other apps by the same users. Which design best meets this requirement?

A dataflow loads an Opportunities dataset nightly. Business users need near-real-time updates for a subset of fields (Stage, Amount) during business hours, but the rest of the dataset can remain nightly. What architecture is most appropriate?

A CRM Analytics admin notices a dataset is producing inflated totals after a recipe step that brings in Opportunity Line Items. The business expects Opportunity-level totals, not Line Item-level totals. What is the best way to prevent double counting in downstream dashboards?

A consultant is designing a CRM Analytics dashboard for executives who primarily view results on mobile devices. Which design choice is a recommended best practice?

An Einstein Discovery story is built to predict Case Escalation. A user reports that when they click "Why" (explanations), the top driver is a field that should not be used because it is only populated after escalation occurs. What should the consultant do?

A user can access CRM Analytics but receives an error when opening a specific dashboard: they can see the app but not the dashboard contents. Other dashboards work. What is the most likely cause?

A dashboard includes a table showing Top 20 Accounts by Revenue. The business wants the table to automatically re-rank when the viewer applies page filters (Region, Industry). What configuration supports this behavior?

A dataflow fails intermittently with errors related to too many input rows during a join. The source objects are large, and only the last 90 days are needed for analytics. What is the best mitigation?

A consultant needs to build an Einstein Discovery model for renewal likelihood. The dataset contains multiple rows per Account (one per product subscription). The business wants predictions at the Account level. What is the most appropriate approach?

A team wants to ensure that a dashboard filter selection (e.g., Fiscal Quarter) applies consistently across all widgets, including those built from different datasets. What is the recommended configuration?

A CRM Analytics implementation requires row-level security so that users only see Opportunities they own or that are on their team. The organization already maintains sharing in Salesforce. What is the best approach to align CRM Analytics security with Salesforce sharing at scale?

A recipe outputs a dataset where a date field is being treated as text, causing time-series charts to sort incorrectly. What is the simplest fix?

A CRM Analytics consultant needs to create a dataset that appends a constant column called Region with value "NA" to every row coming from a CSV upload. What is the simplest way to do this in a dataflow recipe?

A dashboard designer wants users to click a bar in a chart and navigate to a record page in Salesforce for the selected Account. Which dashboard feature best supports this requirement?

An admin needs to ensure that only members of the "Sales Ops" team can edit a specific CRM Analytics dataset, while other users can still view dashboards built on it. What is the recommended approach?

A consultant is building a dataflow that unions two sources: one provides Amount as a number, the other provides Amount as text (e.g., "1000"). The union step fails due to a schema mismatch. What should the consultant do?

A dashboard includes multiple pages with related but different analyses. Users complain that when they change a global filter on page 1 and then go to page 2, the filter does not apply there. What design change best addresses this?

A dataset contains multiple date fields (CreatedDate, CloseDate, LastActivityDate). The business wants a single dashboard date filter that can switch which date field is used for filtering without duplicating the dashboard. What is the best approach?

A consultant creates an Einstein Discovery story to predict Opportunity win probability. After deployment, users report that some predictions are missing for Opportunities. The dataset contains null values in several key fields. What is the most likely reason predictions are missing?

A business wants to take action when Einstein Discovery predicts a high churn risk for an Account. The requirement is to automatically create a Case for the customer success team only when the predicted churn risk exceeds a threshold and the model is considered reliable. What is the best implementation approach?

A security review finds that users can view aggregated metrics in a dashboard even though they do not have access to the underlying Salesforce object records. The company requires that CRM Analytics respect Salesforce row-level security for those objects. What should the consultant configure?

A consultant deploys an Einstein Discovery model in a production org, but the predictions appear systematically different from those seen in the sandbox where the model was built. The same scoring logic is used, but business outcomes are worse in production. Which is the most likely cause and the best next step?