Workforce forecasting is most useful when it helps operations act earlier. By the time vacancy counts spike, the leading indicators that could have warned the business were usually visible already.
This guide outlines simple forecasting models and the signals employers can use to improve labour planning.
Need better workforce planning support? Learn more about managed skilled workforce solutions.
Key takeaways
- Good forecasting combines demand signals, labour availability and operational constraints.
- Leading indicators matter more than lagging vacancy counts alone.
- A simple forecasting rhythm is usually more valuable than a complex model no one uses.
Useful leading indicators
- sales pipeline or production schedule shifts
- shutdown or project mobilisation dates
- planned overtime trends
- absence and attrition movement
- training throughput and supervisor capacity
Simple forecasting models
Baseline model
Use historic demand, seasonality and known operational peaks to set a base case.
Scenario model
Build low, base and high-demand scenarios so procurement and operations can prepare supplier capacity earlier.
Constraint model
Layer in what could limit output: induction slots, supervisor ratios, transport, site access or training bottlenecks.
How to make forecasting practical
- Review a small set of indicators weekly or fortnightly.
- Translate forecast changes into approved actions.
- Share the same view across procurement, HR and operations.
- Compare forecast versus actual and refine assumptions each cycle.
Building the forecasting rhythm
The model matters less than the cadence. Most forecasting programs improve significantly when they move from ad hoc to a consistent review rhythm:
- Weekly: review the operational dashboard — current fill rate, open orders, early attrition, overtime levels. Flag any signals that are moving outside acceptable range.
- Monthly: review the demand plan against the labour plan. Update assumptions based on what changed. Confirm the supply pipeline is tracking against the next 30–60 day need.
- Quarterly: review structural assumptions — has the mix of demand changed? Are there seasonal patterns not previously captured? Do supplier capacity levels still match the plan?
Common forecasting failures
- Only tracking vacancies: vacancies are a lagging signal. By the time a vacancy is raised, the lead time to fill is already eating into delivery capacity. Monitor demand before requisitions are raised.
- Siloed planning: operations forecast volumes, procurement manages suppliers, HR manages headcount — but they’re not connected. Misaligned assumptions produce plans that look coherent on paper and break in execution.
- Over-reliance on last year: seasonal patterns are useful baselines, but operational changes, market shifts and attrition trends mean last year’s curve may no longer fit. Validate assumptions each cycle rather than rolling them forward unchanged.
When your forecast misses — and what to do next
Forecasts will be wrong. What matters is whether the miss was a model failure — wrong assumptions built in, such as assuming 15% attrition when actual was 25% — or an execution failure, where the model was right but input data arrived late or was incorrect. Distinguishing these determines whether you fix the model or fix the data process. Treating every miss as a model problem leads to constant overhauls that rarely help.
Run a brief retrospective after each cycle — 30 minutes, focused on what the model predicted versus what happened and why. Update one assumption at a time rather than overhauling everything at once. Teams that do this consistently get meaningfully more accurate within 3–4 cycles. The discipline of reviewing the miss matters more than the sophistication of the model itself.
Related reading
For a closely related guide, read Workforce Planning Template: 90-Day Hiring Plan.
Related services
FAQ
Do you need complex software to forecast well?
No. Many employers improve significantly with cleaner assumptions, better operating rhythms and clearer ownership.
What is the biggest forecasting mistake?
Treating demand planning and labour planning as separate processes. They should inform each other directly.
How do you know when your forecast assumptions are wrong?
When actual outcomes consistently diverge from forecast in the same direction, your assumption is wrong — not your luck. Track forecast versus actual at each cycle and explicitly review why the gap occurred. Common culprits: underestimated lead times, overstated internal redeployment potential, or attrition rates higher than assumed.
Who should be involved in the forecasting process?
At minimum: operations (demand inputs), HR or workforce (headcount planning), and procurement or labour supply (supplier capacity). Finance should have visibility for budget alignment. The forecasting process is weakest when it’s owned by a single function without inputs from the others.
Next step
If you want stronger workforce forecasting and mobilisation planning, explore managed skilled workforce solutions.
General information only: this article provides general information and is not legal advice.
