Meal & Break Exception Application

Purpose

HR was previously identifying and following up on meal and break issues manually (loosely aided with an Excel macro). This was a largely unsustainable process that didn't capture issues as proactively as desired and required a lot of back-and-forth emails with managers. Working together with HR and Legal, we formalized Farmers' meal and break policy into the core logic needed to determine when an exception had occurred. The application I built analyzes employee timecards according to said logic for meal and break exceptions and supports the multiple workflows needed to resolve them. It was directly used by employees, managers, and HR consultants and tied into our existing waiver and timecard-correction processes.

My role

Served as the primary developer and point of contact for most of the project, from the initial requirements gathering through rollout and ongoing support. Delegated selected tasks to teammates and other teams as needed, and worked directly with HR on requirements, acceptance testing, and weekly sprints.

Operational scope

Scanned non-exempt employee timecards across Farmers for applicable state and local policy exceptions. Supported escalations, meal waivers (existing and incoming), and near misses, with metrics to identify recurring problems.

Implementation

• Used Quickbase (a low-code PaaS) as a frontend, customized with HTML, CSS, and JavaScript; Python and SQLite supported backend processing and business logic on a team-owned server
• Directed managers via email to the app to investigate exceptions and record resolutions in the application
• Supported multiple remediation paths for employees, managers, and HR, including waivers and timecard corrections
• Worked with HR on requirements and acceptance testing as features were introduced over many phased rollouts
• Support for a great number of edge cases (such as inter-state travel and missing in/out punches)
• The system was designed to be pessimistic, preferring false positives over false negatives, but allowing easy manual review of all exceptions for such cases

Outcome

Reduced ongoing meal and break exceptions by more than 60% compared to the baseline prior to the process starting. Exception handling went from a cumbersome process that took considerable effort to a largely automatic system. The process became quite good at catching issues over years of ongoing refinements to how exceptions were handled.