The questions to ask a WFM vendor before signing
Most WFM selections are decided on the wrong information
The standard WFM buying process is two demos, a reference call, a procurement negotiation, and a signature. The demos show the platform at its best on curated data. The reference call is with a customer the vendor has briefed in advance. The procurement negotiation is about price, not capability. By the time the contract is signed, the buyer hasn’t asked half the questions that will matter in year two, and the answers they have are calibrated to look good in a sales process rather than to predict operational reality.
This article is the buyer’s side of the conversation. Twelve specific questions to put to any WFM vendor before signing — the ones that surface where the platform actually struggles, where the implementation will hurt, and where the year-two cost lives. The vendors who give clear honest answers are the ones worth shortlisting; the vendors who deflect them are telling you something more important than the answer itself.
The twelve questions
1. “Show me the forecast for a queue with 20 contacts an hour and a queue with 2,000. What does the platform do differently for each?” Tests whether the forecasting engine understands the Poisson floor (see why understanding Poisson matters for planners). Strong platforms handle small-volume forecasting with different methods and surface confidence intervals automatically. Weak platforms produce a single point estimate either way and don’t mention the variance. The answer here predicts how much manual override your team will be doing.
2. “How does the schedule engine handle multi-skill queues with overlapping skills?” Tests whether multi-skill scheduling (see multi-skill scheduling) is properly solved or papered over. Ask to see the platform schedule a real-looking 4-skill operation. Watch what happens when skills overlap and contention exists. Strong platforms solve this with documented algorithms; weak platforms hand the problem back to the planner as “manual adjustment.”
3. “What real-time interventions can the platform suggest, and what’s the evidence the suggestions are right?” Tests whether real-time AI is genuine or marketing. Ask for documentation of the intervention logic. Ask about the false-positive rate. Ask whether the suggestions have been measured against operational outcomes. Strong vendors share the maths; weak ones say “our AI handles it.”
4. “Walk me through a typical 6-month implementation from kick-off to first production schedule.” Tests realism. Get the actual timeline, the resources required, the integration work, the data-migration steps, the parallel-run period, the cutover plan. Vendors who quote “8–12 weeks” for a complex multi-site implementation are setting up an over-running project. Realistic timelines are 4–9 months for non-trivial operations.
5. “How much of the implementation effort is your team and how much is mine?” Tests where the work actually lands. Most WFM implementations require substantial buyer-side effort — data cleansing, business-rule translation, change management, training. Vendors that downplay the buyer’s effort are setting up a relationship where blame for delays will be contested. Ask for a specific RACI by phase.
6. “What does year-two cost look like, including the things not in this proposal?” Tests for the hidden cost. Year-one pricing is usually keenly discounted; year two often jumps 30–100% as discount unwinds, modules get added, and professional services bills land. Ask for the full three-year cost including support, professional services, additional modules they know you’ll need, and the price-uplift mechanism. A vendor who can’t quote year two transparently is hiding something.
7. “What integrations are in scope, and which ones have you actually built before vs ones you’d build new for us?” Tests integration risk. Vendors will quote integration with your ACD/CRM/HR system as routine. Often it’s the third or fourth time they’ve done it — routine. Sometimes it’s the first time — novel and risky. Ask which it is for each integration. Novel integrations should be costed and risked accordingly.
8. “Who from your team owns my account after the implementation team rolls off, and what’s their workload?” Tests post-implementation reality. Pre-sale, you get the A-team. Post-implementation, you get a customer success manager who has 40 other accounts. Ask specifically: who, by name, how often you’ll meet, what their authority is, and what happens when you escalate. The honest answer determines how much of the platform’s value you’ll actually realise.
9. “What happens to my data if I leave?” Tests exit. The platform holds years of forecasting data, schedule history, configuration, custom reports. The contract may have provisions about data extraction, format, and timing. Ask for them specifically. Vendors who answer vaguely have made the exit deliberately friction-laden. Knowing this before signing changes the negotiation.
10. “What’s your AI roadmap for the next 24 months, and what are customers like us paying for that, if anything?” Tests future direction. Vendors are all investing in gen-AI; the question is whether the AI improvements are bundled in the existing contract or sold as new modules. Vendors who plan to charge separately for AI are setting up a renewal that won’t be linear. Get the roadmap on paper.
11. “Can I have a six-month break clause if implementation goes badly?” Tests confidence. Vendors confident in their implementation will offer reasonable contract flexibility — not full exit, but reduced commitment if specific milestones are missed. Vendors who refuse to discuss any break clause are signalling they don’t believe in their own implementation timeline. The answer is itself the answer.
12. “Give me three references — one happy, one mid-implementation, and one who chose you and regretted it.” Tests honesty. Every vendor will give you happy customers. The vendor who gives you a mid-implementation customer is letting you see real, current pain rather than rear-view-mirror polish. The vendor who gives you a regretful customer is being honest about who they’re a poor fit for — and that honesty is the strongest positive signal in the entire process. Vendors who refuse the third reference are telling you they don’t want you to know who their unhappy customers are.
How to use these questions
Three practical notes on running this conversation.
Ask them in writing. The vendor’s sales lead may answer well in the room and forget it later. Get the answers in writing — an email, a follow-up document, a marked-up RFP response. The answers form part of the contract conversation.
Talk to the implementation lead, not the sales lead. The sales lead’s job is to win the deal. The implementation lead’s job is to deliver. They give different answers to the same questions. Always have at least one conversation with the implementation lead, ideally without the sales lead present.
Score the deflections. A vendor who deflects three or more of the twelve questions is telling you something important about their post-sign behaviour. Count the deflections; weight them in the decision. A vendor that scored slightly worse on capability but answered all twelve honestly is usually a better long-term partner than one that scored slightly better on capability but ducked the hard ones.
The questions vendors love and you should be wary of
Three questions buyers commonly ask that vendors prepare for and that don’t actually surface what matters.
“What’s your roadmap?” Every vendor has a slick roadmap presentation. It’s rehearsed, polished, and a 50/50 bet on actually being delivered. Useful as colour; not decisive.
“Who are your other customers?” The logo slide. Vendors deploy this to imply credibility by association. It tells you nothing about whether the platform works for an operation like yours.
“Can you do a custom demo?” The custom demo is curated, the data is sample, the path through the platform is rehearsed. It shows the platform working when conditions are perfect. Useful as a feature walkthrough; misleading as evidence of operational fit.
Replace these questions with the twelve above and the conversation changes shape.
The honest summary
WFM selection is a 5-year commitment costing six or seven figures. The information asymmetry between vendor and buyer is enormous — the vendor has run this conversation hundreds of times; the buyer is running it once. The twelve questions level the field. They’re not exhaustive, but they surface the things that go wrong most often after signing — weak forecasting on small queues, multi-skill scheduling that doesn’t solve, professional services bills, year-two cost jumps, post-implementation support drift, and exit friction. Ask them. Get the answers in writing. Weight the deflections. The vendor that earns your business should be the one that answers honestly, not the one that demos best.
Conclusion
Twelve questions, twelve answers worth getting in writing. The vendors who answer them clearly are the ones worth shortlisting; the vendors who deflect them are giving you a more important answer than the one you asked for. Most WFM disasters are predictable from these conversations — the buyer just doesn’t have the conversations until after the contract is signed. Don’t be that buyer.
Pair with choosing a WFM system, what is a WFM system, WFM system benefits, the WFM vendor directory, and what planners get wrong about gen-AI.