What decision latency actually is
Decision latency is the interval between the moment a signal appears in your campaign data and the moment your system acts on it. In a human-managed process, that interval includes: someone noticing the signal, flagging it, the team reviewing it at the next available checkpoint, agreeing on a response, and implementing the change.
At its fastest — a responsive team, a clear signal, no approval friction — that process takes hours. In normal operating conditions, it takes days. In complex organizations with multiple approval layers, it can take a week.
The market doesn’t pause while that process runs.
How decision latency cost accumulates
A campaign starts showing CPA degradation at 11pm on a Tuesday. The signal is clear in the data. Nobody is watching. By Wednesday morning, the degradation has been running for eight hours. A competent analyst catches it at the 9am review, investigates through the morning, and implements a fix by early afternoon.
That’s roughly 14 hours of suboptimal spend. On a EUR 50k weekly campaign, 14 hours represents approximately EUR 4,200 of budget running against a problem the system already knew about.
That’s one incident. One signal. One campaign. Running every week, across multiple campaigns, the number compounds quickly — and quietly, because it never appears as a line item anywhere.
Why it never appears in the report
Decision latency cost is structurally invisible in standard reporting because the reports were designed inside the assumption that latency exists. The weekly or daily report shows you what happened across the full period. It averages across the hours when performance degraded and the hours when the fix was running. The issue and its resolution both disappear into the aggregate.
There is no column for ‘budget spent while a known problem was pending human action.’ That column has never existed because building it would require acknowledging the cost of the process itself — and the process was accepted as normal before anyone thought to measure its cost.
What the invoice looks like when you calculate it
The calculation is straightforward, even if it’s never been done. Take your average weekly campaign budget. Identify the average number of optimization interventions your team makes per week. Estimate the average interval between when the signal that prompted each intervention first appeared and when the intervention was implemented.
Multiply the hourly budget rate by the average latency interval across all interventions. That number is a conservative estimate of your decision latency cost per week — conservative because it only counts the interventions your team actually caught, not the ones that compounded undetected until they showed up as a performance problem in the following week’s review.
For most advertisers running campaigns at meaningful scale, the number is significant. It’s been invisible, but it’s been real.
What changes when decision latency approaches zero
When the system that identifies a signal is the same system that acts on it — without the coordination overhead of human review, without the scheduling friction of a checkpoint, without the approval loop between diagnosis and execution — the interval between signal and action shrinks to milliseconds.
The cost doesn’t get reduced. The condition that produces it gets eliminated.
The budget that was running overnight against a known issue gets redirected the moment the problem is identified. The reallocation window that stayed open for four days closes in seconds. The invoice that was never shown to you stops accumulating.
Mainkore operates at that speed. Across every campaign. Continuously. KPIs guaranteed by contract.
