100+ Free Certified Pega Data Scientist Practice Questions

Pass your Pegasystems Certified Pega Data Scientist (CPDS) exam on the first try — instant access, no signup required.

✓ No registration✓ No credit card✓ No hidden fees✓ Start practicing immediately

Not publicly published Pass Rate

100+ Questions

100% Free

1 / 100

Question 1

Score: 0/0

Which Pega application is the primary home for the Data Scientist persona's day-to-day work configuring adaptive and predictive models?

App Studio

Admin Studio

Prediction Studio

Dev Studio

to track

2026 Statistics

Key Facts: Certified Pega Data Scientist Exam

Exam Questions

Pega Academy / Pearson VUE

90 min

Time Limit

Pega Academy

70%

Passing Score

Pega Academy

US$190

Exam Fee

Pearson VUE 2026 list price

PEGACPDS88V1

Current Version

CDH 8.8 / Infinity '25

~28%

Largest Domain

Adaptive Analytics

Pearson VUE

Delivery

Online proctored or test center

Data Scientist '25

Recommended Mission

Pega Academy

The current CPDS exam ships as PEGACPDS88V1 (Pega Customer Decision Hub 8.8 / Infinity '25 alignment) with 50 questions, a 90-minute time limit, a 70% passing score, and a US$190 Pearson VUE fee. Adaptive Analytics (around 28%) and Predictive Analytics (around 25%) carry the most weight, with sizable additional coverage of Prediction Patterns, Pega NLP, Pega Process AI, and AI for Customer Decision Hub. The exam expects you to act as a configurer rather than an algorithm author: you tune ADM predictors, import and validate PMML or H2O models, build decision strategies, and read AUC, lift, performance, and propensity diagnostics.

Sample Certified Pega Data Scientist Practice Questions

Try these sample questions to test your Certified Pega Data Scientist exam readiness. Each question includes a detailed explanation. Start the interactive quiz above for the full 100+ question experience with AI tutoring.

1Which Pega application is the primary home for the Data Scientist persona's day-to-day work configuring adaptive and predictive models?

A.App Studio

B.Admin Studio

C.Prediction Studio

D.Dev Studio

Explanation: Prediction Studio is the dedicated workspace for managing predictions, predictive models, adaptive models, and text models in Customer Decision Hub. It is where the Data Scientist persona spends most of their time.

2In Customer Decision Hub, what is the purpose of the Next-Best-Action (NBA) paradigm?

A.To send the same message to every customer at the same time

B.To determine the most relevant action for each customer in each interaction

C.To replace all human marketing decisions with hard-coded rules

D.To train an LLM that writes marketing copy automatically

Explanation: NBA in Pega CDH evaluates eligibility, applicability, suitability, and propensity for each candidate action and selects the action that maximizes value for the customer and the business in the current context.

3Which component of Customer Decision Hub stores every interaction and outcome that has been recorded for customers?

A.Visual Business Director

B.Interaction History

C.Decision Hub Portal

D.Strategy Framework

Explanation: Interaction History (IH) is the persistent log of who saw what, when, and how they responded. ADM and many strategies read from IH and from IH summary aggregates.

4What is the Adaptive Decision Manager (ADM) in Pega Customer Decision Hub?

A.A batch ETL component for moving data into Pega

B.A self-learning service that builds and updates propensity models in real time

C.A configuration UI for static if-then rules

D.A reporting dashboard for marketing managers

Explanation: ADM is the online learning service that creates and continuously updates Naive-Bayes-based propensity models for every Issue/Group/Action and context combination as new responses arrive.

5Which underlying algorithm does the standard Pega ADM model use to estimate propensity?

A.Random forest

B.Naive Bayes with frequency-based predictor binning

C.Deep neural network

D.Linear regression

Explanation: Each ADM model is a Naive Bayes classifier whose numeric and symbolic predictors are auto-binned to maximize predictive value across the active group of accepts vs rejects.

6What is the role of Predictors in an Adaptive Decision Manager (ADM) model?

A.They are the response outcomes the model learns to predict

B.They are the input features (customer, channel, action context) the model uses to estimate propensity

C.They are the email templates sent to customers

D.They are the visualization widgets in the ADM dashboard

Explanation: Predictors are the input attributes (numeric, symbolic, IH-derived) that ADM evaluates per response and uses to estimate the probability that a given context will accept a given action.

7Which two of the following are typical context keys for an ADM model in Customer Decision Hub?

A.Issue and Group

B.Page Title and Image URL

C.Database Schema and Table

D.OS Version and Browser Build

Explanation: ADM models are usually keyed on Issue, Group, Action (Proposition), Channel, and Direction. Issue and Group are the canonical first two keys that scope an adaptive model to a meaningful business segment.

8Two ADM metrics are reported in Prediction Studio: Performance and Success Rate. What does Performance measure?

A.How fast the model returns a propensity at runtime

B.The model's discriminative power, expressed as AUC scaled to 0-100

C.The percentage of customers who clicked any action this week

D.The number of predictors currently active

Explanation: ADM Performance is AUC (area under the ROC curve) scaled into the 50-100 range. It measures how well the model separates accepters from non-accepters and is independent of base rate.

9An ADM model is reporting Performance close to 50. What does that indicate?

A.Excellent discrimination — half of customers accepted

B.The model is no better than random at separating accepters from non-accepters

C.The model is overfitting badly

D.The training data has been deleted

Explanation: Performance is AUC on the 50-100 scale. A value of 50 corresponds to AUC 0.5, meaning the model is no better than chance. You should investigate predictors and data quality.

10What does Success Rate represent for an ADM model in Customer Decision Hub?

A.The fraction of model invocations that returned without an error

B.The proportion of positive outcomes (accepts) among all responses captured for that model

C.The number of predictors with high importance

D.The latency of the propensity calculation

Explanation: Success Rate is the empirical positive-response rate for the model — accepts divided by total responses. It is a base rate measure and should be interpreted alongside Performance.

About the Certified Pega Data Scientist Exam

The Certified Pega Data Scientist (CPDS) exam validates your ability to configure adaptive and predictive AI inside Pega Customer Decision Hub. The blueprint focuses on Prediction Studio, Adaptive Decision Manager (ADM), Next-Best-Action arbitration, decision strategies, predictions, scorecards, PMML and H2O model import, response and outcome capture, and Pega NLP for text-driven case work.

Assessment

50 multiple-choice and scenario questions delivered through Pearson VUE

Time Limit

90 minutes

Passing Score

70%

Exam Fee

US$190 (latest published Pearson VUE list price; some Pega Academy postings reference US$175) (Pegasystems)

Certified Pega Data Scientist Exam Content Outline

~6%

AI for Customer Decision Hub

How AI fits into Pega Customer Decision Hub, the Next-Best-Action paradigm, and the data scientist persona role within an NBA delivery team.

~16%

Decision Strategies

Strategy framework, components like Group By, Filter, Set Property, Switch, Champion-Challenger and Sub Strategy, plus integrating predictions and adaptive models into arbitration.

~25%

Predictive Analytics

Predictive model lifecycle in Prediction Studio, model templates, scorecards, importing PMML and H2O models, training data preparation, and validating performance with AUC, lift, gini and confusion matrix.

~28%

Adaptive Analytics

Adaptive Decision Manager service, ADM model context keys, predictors, Naive Bayes binning, online learning, propensity vs performance, and ADM monitoring dashboards.

~22%

Prediction Patterns

Prediction types and lifecycle (catalog, candidate, shadow, active), response strategies, outcome capture, IH summary, churn or proactive retention, and propensity-based arbitration.

~8%

Pega NLP / Text Analytics

Text categorization, sentiment, intent classification, entity extraction, training NLP models in Prediction Studio, and using Pega Email Bot for case routing.

~5%

Process AI and Governance

Pega Process AI for case-level predictions, Visual Business Director, ethical bias monitoring, transparency thresholds, and notification templates for model issues.

How to Pass the Certified Pega Data Scientist Exam

What You Need to Know

Passing score: 70%
Assessment: 50 multiple-choice and scenario questions delivered through Pearson VUE
Time limit: 90 minutes
Exam fee: US$190 (latest published Pearson VUE list price; some Pega Academy postings reference US$175)

Keys to Passing

Complete 500+ practice questions
Score 80%+ consistently before scheduling
Focus on highest-weighted sections
Use our AI tutor for tough concepts

Certified Pega Data Scientist Study Tips from Top Performers

1Spend roughly half of your study time in Prediction Studio and on Adaptive Decision Manager because Adaptive Analytics plus Predictive Analytics together approach 53% of the exam.

2Build at least one ADM model end to end: define context keys, add predictors, run a data flow with simulated responses, then read the predictor importance and AUC graphs.

3Memorize the prediction lifecycle states (candidate, shadow, active) and what happens to outcome capture and arbitration at each step.

4Drill the difference between performance (AUC) and propensity, because exam scenarios will contrast the two and a wrong answer almost always confuses them.

5Practice importing a PMML or H2O model into Prediction Studio and mapping its inputs to Pega data model fields, since import scenarios appear in Predictive Analytics questions.

6Memorize the strategy components (Group By, Filter, Set Property, Switch, Sub Strategy, Proposition Data, Decision Analytics) and which one is correct for combining predictions with arbitration.

7Skim Pega NLP topics in the last week: text categorization, sentiment, intent, entity extraction, and Email Bot use cases each yield a small but reliable cluster of questions.

Frequently Asked Questions

How many questions are on the Pega CPDS exam and how long do I have?

The CPDS exam ships as 50 questions in 90 minutes. You need at least 70% to pass. Questions mix straight multiple-choice with scenario-style items where you read a small business situation and pick the correct configuration in Customer Decision Hub or Prediction Studio.

What is the current exam version code?

The current production version is PEGACPDS88V1, aligned with Pega Customer Decision Hub 8.8 / Infinity '25. Older 8.x versions (such as PEGAPCDS86V1) have been retired. Always confirm the active version on Pega Academy before scheduling because Pega refreshes its exam codes when major Infinity releases land.

How much does the exam cost?

The most-cited Pearson VUE price in 2026 is US$190 per attempt, with some Pega Academy postings still showing US$175. Each retake requires the full fee; Pega does not provide free retake vouchers and does not bundle the exam into a Pega Academy mission.

Are there prerequisites?

Pega Academy lists the Data Scientist '25 mission as the recommended preparation path, plus passing the Pega Infinity '25 platform exam if you want a current Infinity-aligned credential. There is no formal mandatory training, but candidates without hands-on Prediction Studio time tend to fail.

Which topics matter most?

Adaptive Analytics is the largest single block at roughly 28%, followed by Predictive Analytics at about 25% and Prediction Patterns near 22%. Decision Strategies is 16%. NLP, Process AI, and AI for Customer Decision Hub split the remaining roughly 19%. Plan to spend most of your time in Prediction Studio and on ADM tuning.

Do I need to know data science math?

You do not need to derive algorithms, but you do need to read AUC, lift, gini, confusion matrix, propensity, and performance numbers and choose the right action. CPDS is a configurer exam: you select the right ADM predictor, scorecard, response strategy, or NLP model rather than write training code.