Workforce engineering built on exact math, not guesswork.
The complete open WFM toolkit for contact centers and operations teams: interval forecasting, Erlang C capacity planning, shift scheduling, real-time intraday analysis, and scenario simulation.
Five tools. One unified math engine.
Erlangly connects the entire workforce planning lifecycle from historical time series forecasting all the way to real-time intraday adherence.
1. Time Series Forecasting
Model multi-period contact volume with weighted moving averages, trend projections, and day-of-week seasonality curves. Built to process 100k+ historical intervals off-thread via Web Workers.
2. Erlang C Capacity
Calculate exact net productive agents and gross scheduled staffing needed to hit target SLAs and occupancy ceilings. Supports single-interval modeling and full multi-day bulk CSV uploads.
3. Scheduling & FTE
Bridge interval volume into required FTE headcount accounting for standard work hours and part-time mixes, then allocate shifts across multi-hour patterns to eliminate coverage gaps.
4. Real-Time & VTO
Simulate intraday shift progression against actual call spikes, detect adherence breaches, and compute safe Voluntary Time Off (VTO) releases with strict SLA guardrails.
5. What-If Simulator
Simulate multi-month staffing scenarios under variable volume growth, AHT changes, attrition rates, and new-hire ramp delays. Compare up to 3 scenarios side-by-side.
Saved Plans & Persistence
Save named, timestamped plans across any tool to reload anytime. Zero login required for same-session calculations; Supabase account persistence for cross-device storage.
Why Erlang C Matters in WFM
A common operational mistake in contact centers is naive division: taking total call workload (e.g. 300 calls × 180s = 54,000 work seconds in a 30-min window) and dividing by 1,800 seconds to conclude that 30 agents are sufficient.
Because customer arrivals follow a Poisson distribution with random variance, having exactly 30 agents creates an infinite queue buildup. Erlang C calculates the extra agent buffer needed to absorb statistical peaks and guarantee target wait times.