CASE STUDY

Predicting Demand Using Machine Learning

Client: Wag!

0 K+

Active users

0 +

Employees

At a glance

Challenge

Leveraging Machine Learning to improve the marketing efficiency of an on-demand dog walking and boarding app.

Result

Pariveda assisted the Wag! data scientists in assessing their data and identifying relevant features for Machine Learning experimentation in order to generate improved accuracy and correlation results.

Impact

Wag! now has a Machine Learning model that can reasonably predict whether dog walking demand will increase or decrease within the next seven days.

TECHNOLOGIES USED

Amazon SageMaker

Wag! is an American pet care company that offers a technology platform to connect pet owners with independent pet professionals.

With 400,000 Pet Caregivers offering services available in more than 5,300 cities nationwide, Wag! is used by regular citizens and celebrities alike. When presented with the opportunity to improve their marketing efficiency, Wag! could sit no longer and jumped at the chance.

Improving marketing efficiency for a nationwide pet care platform.

Wag! is an on-demand dog walking and boarding app available in 5,300 U.S. cities across all 50 states. The company recently sought to improve their marketing efficiency by more effectively targeting their marketing campaigns and other activities – specifically by leveraging Machine Learning to forecast nation-level dog walking demand and predict drops in demand, knowledge which, in turn, would inform their marketing efforts. Wag’s data scientists needed the ability to make more accurate predictions more quickly. In an effort to amplify these efforts, Pariveda and AWS partnered with the company to conduct a seven-week Machine Learning Jumpstart project.

How Pariveda assisted the Wag! data scientists in assessing their data and identifying relevant features for Machine Learning experimentation:

The Pariveda team developed a Machine Learning model that can reasonably predict whether dog walking demand will increase or decrease within the next seven days.

Pariveda also identified additional, high-probability data features to incorporate in future Machine Learning experimentation and defined next steps for productizing the current Machine Learning process.