Within two-month study (Oct and Nov, 2022), I passed the PCDC exam and the PCDS exam via self-learning from the official Pega Academy website. I got high scores (91 and 96 out of 100) for the two exams respectively. I would like to share the high levels of those exams. Hope it may help you on the journey. (I also passed the Pega Certified Senior System Architect (PCSSA) at Sept 2022.)

1.         Here are the two score reports and certificates as evidences.  ( to make you have confidence on this blog).

2.         The learning schedules

Normally, I read the mission documents and implement the challenges from the official Pega Academy for 1 hour on weekdays and 2 hours on weekends. For each document, I will make a deep understanding on why it said that. Sometimes, it will give you a good hint on the whole view of the two exams. For example, in one of the documents, it mentions that “Predictions are strategies that add best practices to predictive models”. The statement looks simple, but it covers a lot of info. I will explain it later. After that, I will make notes for each document.  For each exam, I may have around 8-page notes. It helps you to recall what you learn when you review the note.

It is better to pass the PCDC before you take the PCDS exam, as PCDS covers how to use models in the strategies, which are the core of PCDC.

This article is focusing on the high-level understanding of the PCDC and PCDS. For detail level, such as how to create strategy and define an engagement policy, I will leave it to the official Pega Academy website.

3.         The PCDC high level view

It is very important to have a high-level understanding of the PCDC. If you understand that, the PCDC is not hard. From the official website, it can be described with the 2 statements below:

The strategy framework is applied to all relevant Actions and Treatments after you define a Trigger in the Next-Best-Action Designer Channels tab.

Each Trigger generates a strategy that first imports the Actions from the appropriate level of the business structure and then applies the Eligibility, Applicability, and Suitability rules.

When you go to the Channels tab, you will find out there are 2 trigger types: Real-time Container and Schedule. The former is sitting there and waiting for customers to trigger passively (such as bank websites); the latter is triggered actively (such as marketing email campaigns) by companies to provide information to customers. Each trigger will define the appropriate issue/group according to business request. When triggered, it will go to a funnel as the diagram below.

(Diagram from https://www.linkedin.com/pulse/how-does-pega-marketingdecisioning-strategy-work-nanjundan-chinnasamy).

Here is another statement from the Pega Academy. “A real-time container is a placeholder for content in an external real-time channel.” It defines the relationship between real-time container and real-time channel. You can image channel as a string, and the container is the kite which is pulled (aka triggered) by the string. When kids play kites in the sky, actually, they play the string only. That is the same when customers get an offer on a bank website. Customers are actually getting the offer from the real-time container (for example, Web-Top-Offer container).

The PCDC AI/Propensity part is related to PCDS. I will describe it in the below chapter.

4.         The PCDS high level view

The PCDS does not focus on the theory level of data scientist. If you have ever studied the data scientist course/blog, you will find out that PCDS just touches the surface of data scientist only. For example, ROC/AUC is mentioned in the PCDS, without further explanation on why/how they work.

In my opinion, if you understand the official statement, “Predictions are strategies that add best practices to predictive model”, you will get the full picture of the PCDS. You don’t need to understand any mathematical formula to pass the exam.

In the statement, it has two parts, Predictions are Strategy; Predictions add best practices to Predictive model.

Let’s start with the latter, Predictions add best practices to Predictive model. It describes the relationship between Predictions and Predictive model. The two terms are the different types of Pega rules. If you have the Object-Oriented concept, (maybe, in my opinion), you may treat them as an encapsulation relationship. Predictive model rule/class is encapsulated inside the Prediction rule/class. You can replace the predictive model (such as from scorecard to real complicated model) without impacting other rules (such as engagement strategy) when they refer to the Prediction rule.

When the Prediction rule encapsulates the Predictive model rule, it adds the best practices, such as the control group to measure lift performance and the response time setting to create a response alternative result.

Further, Predictive model rule encapsulates different types of actual data-science mathematical models, such as decision tree model, regression model, Bayes model. You can treat Pega Predictive model as a placeholder of those mathematical models.

Now let’s go back to the former part, “Predictions are strategies”. It reveals the actual implementation of Predictions, which are strategy rules. If you have passed the PCDC, you will know how to create the strategies from scratch. So, you can create a Prediction from scratch without the Pega wizards, where you can define 3 prediction types, CDH, Case management and Text Analysis. But I did not implement the prediction from scratch by myself. Maybe you can try. Why do I mention it? I got it from xxx.

For the ROC/AUC, it is mentioned a lot in the PCDS. For example, it used to select Predictor and Model. The Pega official statement to understand the term is “The receiver operating characteristic (ROC) curve represents the predictive ability of a model by plotting the true positive rate (sensitivity) against the false positive rate (1 – specificity or inverted specificity)”. The statement is not clear enough. I use the Wikipedia statement, “The ROC curve is created by plotting the true positive rate (TPR) against the false positive rate (FPR) at various threshold settings.”, as a supplement to the Pega.  To understand the statement, you need to have the concepts about True Positive, False Positive, True Negative, False Negative and Threshold(cut-off) setting. The diagram (copied from Amazon SageMaker website) describes those relationships. You may use it as a hint to further investigation if you like (as PCDS does not focus on this part, so I will not describe further here. I may write it in another article later).

Summary

It is just a high-level view on PCDC and PCDS. Lots of low-level exam problems require you to read the Pega Academy thoroughly and carefully.

I have attended/passed 4 exams (CSA, SSA, PCDC, PCDS). Sometimes, I feel that those exams are SAT tests, which require a little bit of IQ to understand the problems(question and answer options). I may write some when I have time.