Some of our weekly numbers fluctuate wildly from week to week based on timing, leading to constant red-to-green whiplash that derails our Level 10 Meeting. Should we use four-week or thirteen-week rolling averages on our Scorecard to smooth out the noise, or does that defeat the purpose of weekly data?

Question

Tyler Smith · Accepted Answer

Using rolling averages on a weekly Scorecard defeats the purpose of the tool. The Scorecard exists to give your leadership team an early warning radar. It must show what actually happened during the prior seven days, raw and unvarnished.

When you apply a four-week or thirteen-week rolling average, you create a lagging indicator disguised as a leading metric. A catastrophic operational drop this week gets diluted by the three prior good weeks, allowing a burning fire to hide in plain sight until cash or delivery is compromised.

To handle natural weekly volatility without masking real problems, apply these rules:
- Set targets that reflect realistic weekly floors: If sales closes swing between zero and four every month, do not set the goal at one per week and panic when it hits zero. Set an activity goal that is truly weekly, such as qualified proposals submitted, while keeping revenue closes as a monthly or quarterly tracking metric.
- Look at the thirteen-week trend line: Enter the actual weekly raw number every single week. When reviewing the Scorecard, look at the pattern across thirteen weeks. A single red week in a volatile metric is normal variation. Three consecutive red weeks establish a trend that requires IDS.
- Separate pulse metrics from batch metrics: If an operational output only happens biweekly, track the weekly inputs that feed it rather than averaging the output.

Keep your weekly data raw. Smoothing numbers creates artificial comfort and robs your team of the urgency required to fix broken processes early.