I've been thinking about WFM forecasting lately (shocker, I know). One thing I've noticed across many organisations is how we treat forecasting errors. Rather than seeing them as valuable data points, they're often brushed aside. The real gold is in those errors... patterns, biases, forgotten events that should be fed back into your process. Another issue? Forecast update frequency. When you're only refreshing forecasts every 4-6 weeks, you're missing countless opportunities to incorporate new information and improve accuracy.
In case you missed it Philip Stubbs's full episode:
Some thoughts to consider:
Measure your forecast error systematically and look for patterns; that Saturday bias might be telling you something important about your process.
Implement more frequent forecast reviews - even quick weekly adjustments can dramatically improve accuracy compared to monthly updates.
Create formal feedback loops between operations and planning teams - the contextual knowledge that exists in the heads of your frontline managers is invaluable.
Consider how real-time data can inform forecast adjustments - without real-time decisions, that data is just expensive noise.
Your forecasting process may not be the most exciting topic at executive meetings, but it underpins everything from customer experience to operational efficiency and cost management.
What's your approach to forecast error analysis? How frequently do your teams update forecasts? Always keen to hear different perspectives on this.
Share this post