Decision latency: the growth tax that never appears on the P&L
When every change needs four people to agree, the cost is not the work. It is the weeks between knowing and shipping.
The delay is rarely in the build
Ask a team how long a change takes and they will estimate the development time. Measure it end to end and the picture inverts: a two-day build routinely takes five weeks because it waits for a decision, then for an agency's queue, then for someone to confirm it will not break something else. The build was never the constraint.
Fragmented ownership creates the wait
Growing brands accumulate vendors: a developer, a marketing agency, an email specialist, a fulfilment integrator. Each performs well inside its own scope. But no one owns the result, so anything crossing two scopes escalates to the founder—who becomes the integration layer between people who never speak to each other directly.
Measure lead time, not activity
Weekly status reports full of completed tasks can coexist with a business where nothing important ships. The useful measure is lead time: from the moment a change is agreed to be worth doing, to the moment it is live and measured. Track it for ten changes and the pattern of where time is lost becomes obvious without any new tooling.
Fix ownership before adding capacity
Hiring another developer into a system with unclear ownership adds throughput to a queue that was never the bottleneck. The first correction is one accountable owner per decision, deliverable and metric—then capacity, if the evidence still supports it.