Founded in 2015, ReUp Education helps "stopped out students"-learners who have stopped making progress towards graduation-achieve their college completion goals. The company relies on a team of success coaches to engage with learners and help them reenroll. In 2019, ReUp developed an artificial intelligence (AI) algorithm to help coaches better tailor the support they provide. A disappointing 2021 experiment showing limited utility of the algorithm, however, causes ReUp's senior leaders to question the value of AI for such a personalized, nuanced task like success coaching.
In February 2003, P&G hosted two meetings-one with its largest woman- and minority-owned suppliers and one with its largest non-minority-owned suppliers. Attendees in each meeting heard the same message: P&G was keen to grow its commitment to inclusive supply chains, but felt hamstrung by the limited scale and scope of its existing woman- and minority-owned suppliers. It was up to the attendees to determine how to work together to meet P&G's needs. Otherwise, the consumer packaged goods giant would be forced to look outside its home city of Cincinnati, Ohio, for diverse suppliers at scale. This case tells the story of how Carl Satterwhite, president and co-owner of Infinity Services, a minority-owned furniture installation business, and J. Scott Robertson, president and owner of Globe Business Interiors (GBI), a non-minority-owned office furniture company, responded to P&G's call.
It was April 6th, 2020, and the management team at Drizly-an online alcohol marketplace where consumers could browse and purchase alcohol from local liquor retail stores via Drizly's app for immediate home delivery-were thrilled to see record-breaking sales from the demand surge brought on by the COVID-19 pandemic. For the 10th straight day, gross merchandise value (GMV) sold through Drizly's online alcohol marketplace had surpassed Drizly's GMV on New Year's Eve-typically Drizly's busiest day of the year. But their excitement didn't come without serious concerns. The average delivery time well-exceeded the 60-minute customer promise, the void rate-the percent of orders that were subsequently canceled by either the consumer or retailer prior to delivery-was experiencing a five-fold increase over pre-COVID-19 levels, and the number of customer complaints were on the rise and not being answered in a timely manner. What should Drizly do to ensure the long-term success of the company?
Michael Joyce, Vice President of Inventory Management at JOANN, championed an effort to develop and implement an inventory allocation analytics tool that used advanced analytics to predict in-season demand of seasonal items for each of JOANN's nearly 900 stores and optimize store-specific inventory allocation decisions guided by these predictions. After the analytics tool had been in use for a full season of products, Joyce was surprised to hear that not everyone was pleased with its outcomes. With all of the concerns raised, should Joyce halt the use of the analytics tool for the next season?
The note introduces a variety of methods to assess the accuracy of machine learning prediction models. The note begins by briefly introducing machine learning, overfitting, training versus test datasets, and cross validation. The following accuracy metrics and tools are then described: mean squared error (MSE), mean absolute deviation (MAD), Brier score, and cross-entropy, true/false positives/negatives, the confusion matrix, true positive rate (sensitivity or recall), false negative rate (Type II error rate), precision, true negative rate (specificity), false positive rate (Type I error rate), receiver operating characteristic curve (ROC) and area under the curve (AUC), and precision-recall curve.
Kate Wilson, retail analytics manager at Flashion-a fashion flash-sale site-is tasked with developing analytics to optimize pricing for first-exposure products on the site. Many in the industry have relied on years of experience and intuition to determine pricing, can Wilson provide new insights?