Choosing a workforce management system — key tips to help you through the process
Why this decision is harder than it looks
Choosing a workforce management system is one of the biggest technology decisions a contact centre operation will make, and one of the easiest to get wrong. The shortlist is dominated by polished vendors with confident demos. The internal stakeholders — planning, operations, IT, procurement, finance — each have a different view of what the system is for. The features that look impressive in a demo are often not the ones that determine whether the system delivers value in production. And the cost is significant: licence fees, implementation services, integration work, training, change management, and the opportunity cost of getting it wrong. This article walks through the planner-led approach that gives the project the best chance of landing well: starting with the planning problem rather than the product, framing build/buy/extend honestly, scoping requirements, running a structured shortlist, scoring against criteria that matter, and avoiding the traps. Pair it with our WFM vendor directory for a view of the platforms most operations will encounter.
Start with the planning problem, not the product
The first mistake most WFM selection projects make is starting with the vendors. The right starting point is your operation: what does it look like today, what does it need to look like in three years, and where does the current way of planning fall short? Without that picture, the demos will fill the vacuum — you’ll be sold features rather than buying a fit.
Spend the first two weeks writing a short document — five pages is plenty — that answers four questions. What is the operation? Sites, channels, queues, FTE, contract types, growth trajectory. What is the planning maturity? What gets forecast today, how accurately, by whom, in what tool, and what gets scheduled by which method. Where does it hurt? The two or three planning pain points that the new system is supposed to fix — forecast accuracy, intraday adherence, multi-skill scheduling, holiday handling, real-time visibility. What are the constraints? Budget, timeline, CCaaS platform, integration ecosystem, IT’s appetite for managed services. This document, agreed by the planning team and operations leadership, is the brief the vendors will respond to. Without it, the project drifts.
The build / buy / extend question
Before assuming the answer is a new WFM system, walk through the three honest options. Buy a new dedicated WFM platform. The default assumption, and usually the right one for operations above ~150 FTE with meaningful complexity. The benefit is breadth of capability; the cost is integration work, change management, and the licence. Extend the CCaaS platform’s WFM module. Genesys, Talkdesk, Five9, NICE CXone and others ship WFM inside their CCaaS platforms. For operations already standardised on a CCaaS, this is often the path of least resistance and the cheapest option. The trade-off is that the platform’s WFM is sometimes (not always) less deep than the standalone tools. Build or extend what you have. For small operations or those with very specific needs, a well-built Excel or Google Sheets approach with a Power BI dashboard layer may still be the right answer for another year or two. Don’t buy a WFM system because the industry expects you to; buy one because the cost-benefit case is real.
Scope requirements at the right level
The requirements document is the single most important artefact in the project. Most teams write it badly — either too vague (“the system must support multi-channel forecasting”) or as a thousand-line spreadsheet of every conceivable feature. Both extremes lose information. The level that works is capability-based, with clarity on the must-haves versus the nice-to-haves.
A workable structure has six capability areas, each with 6–12 specific requirements: forecasting, capacity planning, scheduling, real-time, reporting and analytics, and integrations. For each requirement, mark whether it’s a must-have (the system must deliver this to make the shortlist), a should-have (a strong preference but negotiable), or a nice-to-have (a tiebreaker). Resist the urge to mark everything must-have; if more than half your requirements are must-haves you haven’t prioritised them. Use real examples from your operation — not abstract phrases like “supports multi-skill”, but concrete cases like “agents have 1–3 skills, schedules must respect skill priorities, and routing decisions must be visible to the planning team”.
Run a structured shortlist
The shortlist should be three vendors, not five. Beyond three, the effort to do each demo and reference call properly exceeds the discrimination you gain. To choose the three, use the requirements document and the vendor directory — an hour’s desk research per vendor is enough to drop the ones that obviously don’t fit. Common rough-cut filters: scale fit (the vendor’s customer base broadly matches your operation), geographic fit (UK/EU presence if you need it, GDPR posture, language support), integration fit (does it speak to your CCaaS and HR platforms), and commercial fit (within budget order-of-magnitude).
For the three on the shortlist, run a structured demo. Send each vendor the same scripted scenario — not their generic demo, but your operation’s actual challenge. Ask them to forecast a sample week with your data, build a schedule for a specific shift pattern, handle a real-time scenario you’ve drafted. The point is to see how the system handles your problem, not how it handles their canned demo. The vendors that push back on this exercise are giving you useful information about how they’ll behave during implementation.
Score against what matters
A weighted scoring matrix protects the decision from being swayed by the loudest voice in the room or the most charismatic salesperson. Seven dimensions consistently matter:
Forecasting fit. Does it handle your channels, intervals, seasonality, holidays, special events? Can it absorb business intelligence on top of statistical models? How accurate has it been in operations like yours? Reference calls are key here — the vendor’s claims are less informative than another customer’s experience.
Scheduling fit. Does it support your contract types, shift patterns, skill structure, multi-site allocation? Can it self-roster if you need to, or run optimisation if you don’t? Does the scheduler understand your operational constraints or only its own?
Real-time and intraday. Adherence, exceptions, what-if re-forecasting. The intraday function is where the day-to-day battle is won or lost, and it’s often the area vendors demo least well. Push for specifics.
Integrations. Will it talk to your ACD/CCaaS, your HR system, your QA platform, your payroll? Native connectors are far cheaper than custom development. A WFM that integrates poorly with the rest of your stack will bleed value for years.
Total cost of ownership. Not just the licence, but implementation services, the integration build, the training and change management, the internal time to run the project. Five-year TCO is usually the right horizon. The cheapest licence is rarely the cheapest system.
Vendor stability and roadmap. Is the vendor financially solid, are they investing in the product, are their long-standing customers still happy? A vendor that’s coasting will deliver less value year on year, even if today’s product is good.
Implementation realism. Talk to the implementation team, not just the sales team. Ask for the realistic timeline (not the optimistic one), the typical resource requirement from your side, the typical post-go-live stabilisation period. Vendors that quote 8 weeks for an enterprise implementation are either lying or they’re shipping a vanilla deployment with none of your customisation.
Reference calls done well
Every vendor will give you reference calls. Make them count. Ask the vendor for two references that look like your operation (size, sector, complexity), and ideally one that’s been live for at least two years. On each call, ask the same five questions. What did the implementation actually take? — both the timeline and the internal resource. What does the system do well and where does it fall short? — honest answers, not marketing. How responsive is the vendor when things go wrong? — tells you more than any SLA document. Has the vendor delivered on the roadmap promises they made you? — tells you whether to trust the roadmap promises being made to you. If you were choosing again today, would you pick the same vendor? — the most useful question of all, and the answer is rarely a clean yes.
Common traps
Five patterns recur in WFM selection projects that go wrong.
Vendor demo theatre. The flashy AI-driven forecast in the demo is built on synthetic data and tuned to look impressive. The real test is whether it can forecast your data in your operation. Insist on the scripted scenario.
Feature-list scoring without weighting. A 200-feature spreadsheet treats “supports CSV export” the same as “supports multi-skill optimised scheduling”. Weight the requirements ruthlessly: 80% of the value is in 20% of the features.
The cheapest-on-paper trap. A WFM that’s 30% cheaper at the licence line but takes twice as long to implement, integrates worse, and gets less use is more expensive in year one and far more expensive over five. TCO, not licence cost.
Ignoring the implementation team. The sales team that wins the deal hands over to an implementation team you haven’t met. Insist on meeting them before signing, and ideally putting their names in the contract.
Treating it as an IT project. A WFM implementation that’s run by IT, with the planning team as a consulted stakeholder, delivers a technical system that doesn’t fit how planning works. The planning team must own the requirements, the demos, the scoring, and the go-live readiness. IT supports; it does not lead.
The decision moment
When the demos are done and the references are in, the temptation is to decide by gut and rationalise the score afterwards. Resist it. Sit down with the scoring matrix, the reference notes, and the implementation conversations, and run through each shortlisted vendor methodically. If the leading vendor on the matrix is not the gut favourite, it’s worth a second pass to understand why — sometimes the gut is picking up on something the matrix missed (cultural fit, vendor responsiveness, intangible product polish) and that’s a legitimate adjustment to make. Sometimes the gut is just responding to charisma, and that’s not.
Document the decision and the reasoning, even briefly. In three years when the system is in production, you’ll want to remember why you chose it; in three years when you’re considering whether to replace it, you’ll want to remember what you traded off.
Conclusion
Choosing a WFM system is a planning problem disguised as a procurement decision. The operations that get it right treat it as a planning problem — start with the operation, scope requirements honestly, run a structured shortlist, score against what matters, and stay alert to the traps. The investment in doing this well is small relative to the cost of the system and the cost of getting it wrong. The discipline is in the preparation, not the negotiation. For the platforms most operations will encounter on the shortlist, see our WFM vendor directory.
Pair this with what is a WFM system, the benefits of investing in a WFM system, and building a credible planning function.