CDSP (DSP-110) case studies

CDSP (DSP-110) · Put your knowledge to work in context.

The Certified Data Science Practitioner (CDSP) is a CertNexus credential that validates practical, vendor-neutral knowledge of the full data science lifecycle rather than a single tool or platform. CertNexus describes its certifications as built around transferable concepts and skills that are not limited to one technology provider, and CDSP is one of the select CertNexus credentials accredited by the ANSI National Accreditation Board (ANAB) under the ISO/IEC 17024 standard. The CDSP (DSP-110) case studies are 5 cases and about 3,808 words.

Churn Model Request at a Meal-Kit Subscription Service

BrightBox is a meal-kit subscription company with 48,000 active subscribers and 7.2% monthly churn. The VP of Marketing has asked your two-person analytics team to build a churn-propensity model in five weeks so a retention discount can be targeted before Q4.

814 words, about 4 min 2

Customer Data Merge Before a Wealth Propensity Model

Harborline Bank has 1.4 million customer records split between its legacy core system and the CRM of a credit union it acquired 14 months ago. Marketing wants a wealth-product propensity model in ten weeks.

780 words, about 4 min 3

Readmission Model with Good Discrimination but Poor Calibration

Meridian Regional Hospital has 220 beds and a 30-day readmission rate of 8.7%. A vendor-supplied readmission model reports AUC 0.81 on your 2022 validation set, but the calibration slope is 0.68: patients scored at 0.30 risk are observed to be readmitted about 41% of the time.

713 words, about 4 min 4

Checkout A/B Test: Significant Conversion, Flat Revenue

Lumen Outfitters ran a 50/50 A/B test of a redesigned checkout flow: 180,000 sessions over nine days, including a holiday weekend. Conversion was 2.10% in control and 2.36% in treatment, with a two-sided p-value of 0.04.

740 words, about 4 min 5

Demand Forecast Presentation to a Skeptical COO

Northwind Supply forecasts weekly demand for 14 distribution centers. The new model has a mean absolute percentage error of 14.2% versus 22.6% for the current naive seasonal baseline, a result the data team considers a clear win.

761 words, about 4 min