The new-hire ramp-up curve in planning

Workforce economics · ~7 minute read

The fiction that an FTE on day one is an FTE

The capacity plan says the operation hires twenty agents in March. The schedule on the floor for April assumes those twenty agents are working full shifts at full productivity. Neither assumption is honest. New hires complete classroom training, then spend weeks in nesting under supervision, then operate at reduced productivity for months as they learn the systems, the products, and the customers. Treating them as fully ramped from day one over-estimates capacity, under-estimates cost, and turns the planning conversation into a defensive one when the schedule consistently fails to deliver. This article walks through what the ramp-up curve actually looks like, why most operations under-model it, and how to bake a realistic curve into the plan.

What the curve actually looks like

A typical new-hire ramp-up curve has four phases. Training (weeks 1–4): the agent is in classroom or remote training, taking no live contacts. Productivity is zero, but salary is full and there is also a trainer cost being incurred alongside. Nesting (weeks 5–8): the agent takes live contacts with support, handling perhaps a third of the call volume of a tenured agent and taking longer on each one. Net productivity in this phase is typically 30–50% of tenured. Early tenure (weeks 9–16): the agent handles a normal call volume but with longer AHT, more transfers, and more help requests. Productivity rises from around 60% to 85% across this phase. Full productivity (after week 16): the agent matches the operation’s tenured productivity, though quality metrics may continue to improve for another quarter.

These numbers vary by operation. A simple customer-service queue with a single product can ramp in eight weeks; a complex regulated queue can take six months. The shape of the curve matters more than the specific numbers: training at zero productivity, nesting at partial, early tenure at most-of-the-way, and full only after material time on the floor.

Why most operations under-model the curve

Three reasons consistently appear. The first is convenience: modelling each new hire’s productivity individually is more work than counting them as full FTE from start date, and the planning team takes the shortcut. The second is incentives: when the recruitment KPI is “agents in seat,” the political pressure runs against modelling them as anything less than full from day one. The third is invisibility: the under-productivity is real but distributed across many agents over many weeks, and it never produces a single visible event — just a steady undershoot on capacity that everyone blames on something else.

The cost is large and consistent. For a 200-FTE operation hiring 60 agents a year (30% attrition), the cumulative under-productivity across the ramp-up curve typically equates to four to seven FTE-years of lost capacity if not budgeted for. At a fully loaded cost of £28k per FTE, that is well over £100,000 a year of capacity that the plan assumes exists and the operation does not deliver. The same number, modelled honestly, becomes a budgeted reality that the planning team and operations can work with.

The productivity dimensions that matter

New-hire productivity is not a single number. Three dimensions move together but at different rates. Volume per hour — how many contacts the agent handles per available hour — usually recovers fastest, sometimes reaching 90% within twelve weeks. AHT — the time per contact — recovers more slowly because shorter AHT requires both system familiarity and customer-handling instinct. Quality — CSAT, first-call resolution, error rate, complaint volume — usually recovers slowest of all, sometimes still moving twelve months after start. Operations that report “productivity” as a single percentage hide the differential, and the differential matters. A new agent who is doing 80% of the call volume but handling each call 30% longer is producing the same number of contacts but consuming more agent-hours than the plan assumes.

How to bake the curve into the plan

The mechanic is straightforward and depressingly often skipped. For each cohort of new hires, model the expected productivity by week of tenure. Apply that productivity profile to the agent count over time. Sum the result by week into “productive FTE” rather than “FTE in seat.” Schedule against productive FTE, not headcount.

In practice, this means the capacity plan looks like a stacked area chart with several layers: tenured agents at 100% productivity, year-one agents at perhaps 90%, recent hires at the ramp-up curve. The total productive FTE is the sum of the layers. Adding ten new hires in March increases the recent-hire layer by ten; that layer contributes its weighted productivity to the total. By week 20, the new hires have moved into the year-one layer and contribute at the year-one rate. The plan is more complex than “agents in seat” but it is honest, and it produces forecasts that match reality.

Mentoring and nesting design

The ramp-up curve is influenced by how the operation designs the nesting and mentoring phases. Two design choices matter. The first is the mentor ratio: how many new hires share a mentor or buddy. Smaller groups (one mentor per two or three new hires) ramp faster but cost more in mentor time; larger groups are cheaper but ramp slower. The second is the contact-mix progression: whether new hires start with the simplest call types and work up, or take a representative mix from the start. The progressive approach ramps confidence faster and reduces day-one quality damage; the representative approach builds breadth faster and reduces the “cliff” when the agent has to handle harder contacts later. Either can work; what fails is no deliberate design at all.

Tracking individual ramp curves

The aggregate ramp-up curve is the baseline; individual agents vary around it. Operations that track each new hire’s productivity curve weekly — volume, AHT, quality scores — spot ramp-up problems early enough to intervene. An agent whose curve is two weeks behind cohort norms at week six is rarely accidental; it usually signals a training gap, a system-access issue, a confidence problem, or a fit issue. Catching it at week six is much cheaper than discovering it at week twelve. The tracking also feeds back into the planning model: the cohort average becomes a measured number rather than a guess, and the model improves over hiring cycles.

Common mistakes

Four patterns recur. Counting new hires as full FTE from start date, the most common and most expensive mistake. Treating ramp-up as a single productivity number rather than a multi-dimensional curve. Failing to track individual ramp curves, missing the agents who need support. And designing the nesting and mentoring once and never reviewing the design, even as new products, channels, and routing rules change what the curve really looks like.

Conclusion

The ramp-up curve is one of the under-modelled inputs to a contact centre’s planning. It looks small — a few percentage points of productivity for a few weeks per new hire — and it adds up to several FTE-years of capacity that the plan either accounts for or quietly misses. The operations that model the curve honestly produce forecasts that match the floor, have better conversations with finance about hiring lead times, and build new hires more thoughtfully than the operations that treat them as instantly productive. The mathematics is simple; the discipline is harder. Most planning teams that take the discipline seriously look back twelve months later and wonder how they ever did without it.

The cost-of-attrition calculator includes a ramp-up productivity loss component. Pair this with the true cost of attrition for the wider workforce economics context.