How We Build Software, to One Standard

Department management keeps technology legible and steerable for the business, a department with goals and metrics, not a black box with a budget.

• The business learns about technology problems when they blow up: there is no view of the system’s state until it’s an incident.
• Everything hinges on one vendor or two people: a departure, a dispute, or a sick day, and the business is hostage.
• The business can’t tell good engineering work from bad, so it can’t set expectations, and spend leaks with no way to see where.

Project management makes delivery predictable: the right work, sized to real capacity, finished in the order that matters.

• The plan never reflected the team’s actual availability: it’s fiction, and commitments start slipping on day one.
• Sprints end with everything started and nothing shipped: there was no single goal, so effort scattered across half-finished work.
• Scope changes mid-sprint and deadlines slide, and every slipped date breaks something downstream: a launch, a campaign, a promise to a client.

Infrastructure is what holds when something fails: environments, backups, access, and security, owned by the business rather than the vendor.

• Backups exist but a restore has never been tested, and the repos, the tracker, the docs aren’t backed up at all. One failure is unrecoverable loss.
• Data sits exposed to the public internet: a breach waiting for its discovery date, with the legal and reputational bill attached.
• The agency owns the servers, the accounts, and the code: when the relationship ends, so does your access to your own product.