Back to articles

Contact Center Workforce Management: Process, KPIs, Tools

November 20, 2026
By
Cindy Graciella

Half your agents are at lunch when the 2:30 p.m. queue doubles. By the time someone reworks the staffing spreadsheet, the spike has passed, and so has your service level.

It's easy to blame the spreadsheet, but a faster tool won't fix a weak plan. Contact center workforce management (WFM) is how you build a better one: forecast demand, work out how many agents you need, schedule them, and adjust during the day to keep service, cost, and workload in balance.

I'll go through each step, show you when a spreadsheet stops being enough, and compare six WFM platforms for when it does.

The WFM planning cycle, step by step

WFM runs as a loop of five steps. Each one feeds the next, and what you learn in the last step improves your next forecast:

  1. Forecast how much work is coming.
  2. Calculate how many agents that work needs.
  3. Schedule shifts that cover those requirements.
  4. Adjust during the day when reality breaks from the plan.
  5. Review the results and decide what to change next cycle.

Step 1: Forecast contact volume and handle time

A forecast estimates how many contacts arrive in each interval and how long each one takes. Forecast by interval, not by day, because daily totals can match perfectly while individual half-hours are badly understaffed. That's why ICMI advises against judging forecast accuracy on daily, weekly, or monthly totals.

Start with interval history and seasonality, then adjust for known events like marketing launches, billing runs, and holidays.

Forecast volume and handle time separately. Handle time shifts on its own when a policy changes or a new hire class goes live, and a stale handle time will understaff the floor even if your volume forecast is spot on.

Channel type matters too. Voice and live chat need to be answered as they arrive. Email and tickets land in a backlog, so you're forecasting new arrivals, the existing backlog, and when each item is due.

Step 2: Turn forecasts into staffing requirements

Here's how it works for one 30-minute voice interval:

  1. Set the inputs. You're forecasting 100 calls, average handle time (AHT) is 180 seconds, and the goal is to answer 80% of calls within 20 seconds.
  2. Calculate workload. 100 × 180 / 1,800 seconds = 10 Erlangs, or five agent-hours of work in half an hour. That's average workload, not the headcount you need when calls bunch up.
  3. Apply a queueing model. Call Centre Helper's Erlang C worked example uses these same inputs and lands on 14 available agents. I ran the numbers myself, and 13 agents lands just under the goal at about 79.6%.
  4. Add shrinkage. With a 30% shrinkage allowance, you need ceiling(14 / 0.70) = 20 scheduled agents for that interval.

Keep in mind that Erlang C assumes random arrivals, one call per agent at a time, and no abandoned calls, so treat it as a starting point.

Shrinkage covers planned time away (breaks, meetings, training, and time off) and unplanned time (absence and lateness). Just don't count the same minutes twice. If shrinkage already includes breaks, don't add a separate break allowance on top.

Other channels need other models. Chat staffing depends on how many sessions each agent handles at once. Email and tickets are planned around clearing the backlog within response-time windows. And multi-skilled agents can't be counted in full by every queue they cover.

Step 3: Build schedules agents can actually work

A schedule turns interval requirements into real shifts. It has to respect shift rules, local labor laws and contracts, skill coverage for every open queue, and each agent's availability.

Within those limits, use split shifts for separate peaks, part-time shifts for the lunch rush, and staggered starts for opening coverage.

Give agents some control, too. Shift bidding, approved swaps, and self-chosen breaks (all within coverage rules) make schedules easier to live with. Publish schedules early so people can plan their lives. Predictable schedules help you keep good agents.

Step 4: Manage intraday changes and real-time adherence

During the day, compare actual arrivals, AHT, and available agents with the plan for the same interval. That tells you what went wrong:

  • More arrivals than forecast, with attendance on plan, means demand changed.
  • Arrivals on plan, with fewer agents available, means you lost staff.
  • AHT jumping across the whole floor at once usually means a system issue, so check that before blaming agents.

Real-time adherence tracks whether agents are where the schedule says. But adherence and service are separate. A late return from break may not matter in an overstaffed interval, and perfect adherence won't save you from a bad forecast.

Here are the main levers, and when each one makes sense:

Signal Lever Tradeoff
Workload sustained above plan, attendance adequate Approved overtime Premium cost and fatigue risk
Available agents below plan in one queue Reassign qualified agents while keeping the donor queue's minimums Less buffer in the donor queue
A clear near-term coverage gap Move a break, keeping break rules intact Breaks get less predictable
Sustained excess capacity, with service healthy Training, coaching, or voluntary time off Less buffer if volume picks up
01
Workload sustained above plan, attendance adequate
Lever
Approved overtime
Tradeoff
Premium cost and fatigue risk
02
Available agents below plan in one queue
Lever
Reassign qualified agents while keeping the donor queue's minimums
Tradeoff
Less buffer in the donor queue
03
A clear near-term coverage gap
Lever
Move a break, keeping break rules intact
Tradeoff
Breaks get less predictable
04
Sustained excess capacity, with service healthy
Lever
Training, coaching, or voluntary time off
Tradeoff
Less buffer if volume picks up

Step 5: Review forecast accuracy, service, and labor use

The review closes the loop and decides what changes in your next plan. I'd track five things:

  • Forecast accuracy: compare forecast and actual volume and AHT by interval. Track both signed error (to spot bias) and absolute error (to see how far off you are).
  • Service level: contacts answered within your target wait time, divided by all contacts offered, including abandoned ones. For email and tickets, track replies sent before their deadline.
  • Occupancy: handling time divided by handling plus waiting time. Pushing it too high leaves no room for bursts and burns agents out.
  • Adherence and shrinkage: whether agents followed the schedule, and whether your shrinkage allowance was realistic.
  • Hours against plan: staffed hours versus planned hours, which ties service results to labor cost.

Each pattern points back to a step. Persistent forecast error means fixing the forecast. Coverage gaps despite an accurate forecast mean fixing schedule design. And a steady gap between planned and actual time away means adjusting your shrinkage assumption.

What AI can automate in workforce planning

A lot of what gets sold as "AI" in WFM is really rule-based automation, like auto-approving time off when coverage allows. That's useful, but it's workflow automation, not AI.

AI-assisted forecasting and scheduling goes further. It uses predictive models to pick forecasting methods or propose schedules within your rules. Planners should always be able to see why it made a recommendation.

Here's where I'd use it:

  • What-if scenarios, like a volume surge or a delayed hire class.
  • Intraday adjustments, as long as every automated change follows your staffing, skill, and break rules, and exceptions go back to a planner.
  • Replanning after AI deflection. When an AI voice agent, AI receptionist, or answering service handles contacts for you, don't just subtract them from your old forecast. The contacts left for humans are often harder and take longer.

Planners still own service goals, operating rules, and final approval on major changes.

Set WFM policies before the pressure hits

Tools only get you so far. The steadiest WFM programs also run on decisions made before things get hectic:

  1. Plan holiday hot spots early. Publish the rules for popular days off (rotation, lottery, or seniority) before requests open.
  2. Keep your SOPs current. Write down who approves overtime, who can move agents between queues, and when to offer voluntary time off.
  3. Get WFM into budget talks early. Planners can show what a service goal costs in hours before the target is locked.
  4. Protect development time. Schedule training and coaching as planned activities, and rebook anything you postpone.

Do you need dedicated WFM software yet?

Not necessarily. Everything above can run on spreadsheets if your operation is simple enough. Ask yourself five questions:

  1. Channels: do you run one live channel, or voice, chat, and email that each need a different staffing model?
  2. Agents and skills: how many agents and skill combinations does each schedule have to cover?
  3. Sites and time zones: does your coverage span locations that plan separately?
  4. Rebuild effort: how many hours a week go into rebuilding schedules after absences and time-off changes?
  5. Reforecast speed: when the day breaks from plan, can your team adjust while it still matters?

Your answers will point you to one of three options:

  • Stay on spreadsheets if planning is accurate enough and rebuilds are manageable.
  • Use the WFM built into your contact center platform if it covers your channels and skills.
  • Look at a dedicated WFM product if you still need things like concurrency modeling, multi-site scheduling, or real-time adherence.

Team size alone doesn't decide this. A small multilingual team covering voice, chat, and email can be harder to plan for than a much bigger single-queue voice operation.

6 best contact center WFM platforms compared

If your answers point to software, the next question is where it comes from. The right platform depends on whether you're buying it with a contact center platform, a help desk, or a broader engagement suite. Here's how my six picks compare:

Platform

How WFM is sold

Forecasting approach

Pricing

Best for

Nextiva

Add-on to Contact Center Professional; included in Premium

Forecasting from historical data, with intraday management

Essential from $75 per agent/month; Professional and Premium quoted by sales

SMB and mid-market teams that want phones, contact center, and WFM from one vendor

NiCE CXone WFM

Part of the CXone platform (separate from NiCE's IEX WFM)

Autoselect forecasting for volume and AHT per skill; ML needs 8+ weeks of history

From $110 per agent/month (Omnichannel Suite platform, not standalone WFM)

Large enterprise contact centers

Verint (Verint WFM, Calabrio ONE)

Two separate product lines, sold inside engagement suites

AI-powered automation; multichannel forecasting

Custom quote

Enterprises wanting WFM inside an engagement suite

Genesys Cloud CX

Included in CX 3; WEM add-on for CX 1 and CX 2

Automatic Best Method picks from 20+ methods, including ensembles

CX 3 at $155 per user/month, billed annually

Contact centers running Genesys as their CCaaS

Assembled

Standalone WFM

Out-of-the-box contact volume forecasting

$25 to $75 per agent/month, plus a platform fee

Digital-first support teams

Zendesk WFM

Add-on, or bundled with QA in the WEM bundle

AI-powered forecasting from historical ticket data

$25 per agent/month; $50 with QA in the WEM bundle (billed annually)

Support teams already on Zendesk

1. Nextiva

Nextiva's

‍

Nextiva combines business phones, a contact center, and workforce management in one platform. Nextiva Workforce Scheduling plugs straight into Nextiva Contact Center, so forecasts and adherence run on live interaction data instead of exports. It's the one I'd look at first if you're a small or mid-sized team that wants everything from one vendor.

Standout features:

  • Forecasting, automated scheduling, real-time adherence, intraday management, and reporting
  • An agent app for shift bidding, swaps, and time-off requests
  • Pairs with Nextiva's AI quality management and agent assist for a full WFO setup

Limitations: WFM is an add-on at Professional and only included at Premium, so budget for one of those tiers.

Pricing: Essential starts at $75 per agent per month. Professional and Premium are quoted by sales.

Best for: SMB and mid-market teams that want communications, contact center, and WFM with one vendor.

2. NiCE CXone WFM

NiCE's homepage

‍

NiCE CXone is an enterprise contact center suite, and CXone WFM is its built-in workforce management module. It's the depth pick for large contact centers juggling lots of skills and channels. Just note that NiCE's older IEX WFM is a separate product, so make sure you're demoing the right one.

Standout features:

  • Autoselect forecasting, applied separately to volume and AHT for each skill
  • Mobile shift bidding and self-service

Limitations: Its machine-learning forecasting needs at least eight weeks of history, which hurts on new queues and migrations. It can also be more than a smaller team needs.

Pricing: According to NiCE's pricing page, the CXone Omnichannel Suite starts at $110 per agent per month for the platform, not standalone WFM, billed monthly in arrears.

Best for: Large enterprise contact centers with complex multi-skill, multichannel demand.

3. Verint

Verint's homepage

‍

Verint sells workforce engagement software, and since merging with Calabrio, the combined company operates under the Verint name. Calabrio products are staying in the portfolio, so Verint WFM and Calabrio ONE are still separate product lines. Both suit enterprises that want WFM bundled with quality and analytics.

Standout features:

  • AI-powered forecasting and scheduling, with multichannel support in Verint WFM Essentials
  • TimeFlex Bot for agent schedule changes, now also available inside Calabrio WFM
  • Calabrio ONE combines analytics, WFM, and quality management in one suite

Limitations: You'll need to figure out which product line fits, and features in one demo may not exist in the other product.

Pricing: Custom quote.

Best for: Enterprises that want WFM tied to quality and analytics in one engagement suite.

4. Genesys Cloud CX

Genesys'

‍

Genesys Cloud CX is a cloud contact center platform with WFM built in. Because forecasts and schedules draw directly on its routing data, there's no integration to maintain. That's the main reason I'd pick it over a standalone WFM product if you're already on Genesys.

Standout features:

  • Automatic Best Method, which picks from more than 20 forecasting methods, including ensemble forecasts
  • Time-off and shift-trade self-service on desktop and mobile
  • Activity plans that schedule training, meetings, and compliance briefings

Limitations: It's strongest when Genesys is already your contact center platform, and WFM access depends on your tier.

Pricing: WFM is included in Cloud CX 3 at $155 per user per month, billed annually. CX 1 and CX 2 customers can add a WEM license just for the users who need it.

Best for: Omnichannel cloud contact centers already on Genesys, or moving to it.

5. Assembled

Assembled

‍

Assembled is a standalone WFM platform built for customer support teams. It plugs into the help desk and phone tools you already use, so you don't have to switch contact center platforms to get better scheduling. That makes it my pick for support teams that just want WFM.

Standout features:

  • Out-of-the-box volume forecasting and automated schedules that place lunches and breaks
  • Agent-initiated shift swaps, with admin rules by queue, channel, team, or skill
  • Billing based on who's actually scheduled each month

Limitations: The platform fee makes it harder to compare against all-in per-agent quotes, so ask for the full annual cost.

Pricing: Core is $25, Pro is $45, and Enterprise is $75 per agent per month, plus a platform fee.

Best for: Digital-first support teams that want transparent pricing.

6. Zendesk WFM

Zendesk

‍

Zendesk WFM is the workforce management add-on built into Zendesk's help desk. Agents see their schedules and request changes without leaving Zendesk, and managers forecast from the same ticket data. (If you're still deciding on a help desk, here's how HubSpot and Zendesk compare.)

Standout features:

  • AI-powered forecasting from your ticket history, up to a year ahead
  • Automatic schedules built down to the minute
  • Time-off requests and shift trades right inside Zendesk

Limitations: It's less useful if you work across several platforms. Zendesk also requires WEM add-ons for every agent in the account, even if only some need scheduling.

Pricing: WFM on its own is $25 per agent per month, or $50 in the WEM bundle with QA. Both are billed annually.

Best for: Support teams already standardized on Zendesk.

How to choose contact center WFM software

Test each product on your shortlist against your own channels, skills, and service goals, not the vendor's demo script. Here's what I'd check, and what should knock a product off your list:

Criterion What to verify Deal-breaker
Planning depth Forecasting methods for each demand type, intraday reforecasting, and what-if scenarios One voice formula applied to chat and backlog work
Data in Telephony history, agent states, routing skills, ticket data, and HR availability A required source is missing, or state data arrives too late for adherence
Data out Schedules, time-off approvals, and time records flowing back to HR or payroll Manual re-keying of approved changes
Packaging Standalone WFM, WFM inside a CCaaS platform, or a full engagement suite WFM is only available through a migration you don't want
Pilot results Your real history and scheduling constraints, compared with your current approach Worse forecast error or coverage than today

Planning depth

What to verify

Forecasting methods for each demand type, intraday reforecasting, and what-if scenarios

Deal-breaker

One voice formula applied to chat and backlog work

Data in

What to verify

Telephony history, agent states, routing skills, ticket data, and HR availability

Deal-breaker

A required source is missing, or state data arrives too late for adherence

Data out

What to verify

Schedules, time-off approvals, and time records flowing back to HR or payroll

Deal-breaker

Manual re-keying of approved changes

Packaging

What to verify

Standalone WFM, WFM inside a CCaaS platform, or a full engagement suite

Deal-breaker

WFM is only available through a migration you don't want

Pilot results

What to verify

Your real history and scheduling constraints, compared with your current approach

Deal-breaker

Worse forecast error or coverage than today

Keep the WFM decision separate from any bigger platform switch. Buying WFM with a new contact center platform makes sense if you're moving anyway, but it shouldn't force a migration on a team that just needs better scheduling.

When comparing prices, add up platform fees, seat minimums, implementation, and optional modules. And run a pilot on your own data before a full rollout.

Next steps for contact center workforce management

Good contact center workforce management comes down to running the loop well: forecast by interval, staff with the right model for each channel, build schedules people can live with, adjust during the day, and review what to change.

Software makes that loop faster once your channels, skills, or sites outgrow a spreadsheet. Use the five questions above to decide whether you're there. If you are, shortlist two or three platforms and pilot them on your own history. If you want phones, contact center, and WFM from one vendor, Nextiva is where I'd start.

Have questions

What is the 80/20 rule in call centers?

It's a service level target of answering 80% of calls within 20 seconds. It's not a universal standard, so set your target based on what your callers need and what your budget allows.

‍

What's the difference between workforce management and workforce optimization?

Workforce optimization (WFO) is the bigger umbrella. It adds quality management, performance management, and analytics around WFM.

How is contact center management different from workforce management?

Contact center management covers the whole operation, including technology, hiring, quality, and customer experience. WFM is the capacity-planning part of it.

Do WFM software and a contact center CRM do the same job?

No. A CRM, like Salesforce or HubSpot, holds customer records and interaction history. WFM uses contact data to plan staffing.

How often should a contact center reforecast?

There's no universal schedule. Reforecast when interval error stays outside a tolerance you set, and refresh long-range forecasts each time you build schedules.

By clicking “Accept All Cookies”, you agree to the storing of cookies on your device to enhance site navigation, analyze site usage, and assist in our marketing efforts. View our Privacy Policy for more information.

Cindy Graciella

Cindy is an Outreach Manager and SEO Specialist at ONSAAS who helps SaaS companies grow through strategic link building and SEO. Outside of work, she loves spending time in nature, especially hiking in the mountains.

On this page:

No items found.

Related Articles

By clicking “Accept”, you agree to the storing of cookies on your device to enhance site navigation, analyze site usage, and assist in our marketing efforts. View our Privacy Policy for more information.