KuberEva Book a call

Home/Platform evaluations

Platform evaluations

Every major CCaaS platform demos well. These notes are about what we look for underneath the demo — and what we've seen surface in month four.

Independence statement These are our own assessments. We are not a reseller or implementation partner for any vendor listed, we hold no referral agreements, and we receive no commission on any platform decision. All product and company names are trademarks of their respective owners. Where we cite analyst placement, it is attributed and sourced. Vendor capabilities change quickly — treat any specific observation as current to mid-2026 and verify it in your own proof of concept.

How we score a platform

Feature checklists don't discriminate between vendors, because every vendor ticks every box. These eleven dimensions do, because they're where platforms genuinely differ. We weight them with the client before scoring anything.

DimensionWhat we actually test
Routing modelWhat happens when your business rules don't match the vendor's assumed model. How far you can deviate before you're writing custom code.
Reporting granularityWhether historical data answers the question you'll actually ask, at the interval you need, without an export-and-rebuild step.
Real-time APIsEvent availability, latency, and — critically — rate limits at your concurrency, not at demo concurrency.
Telephony & carrier modelBring-your-own-carrier support, SIP trunk termination, SBC requirements, per-minute economics, and how number porting is handled.
Voice quality & media pathWhere media terminates, codec support, and whether you can measure quality independently of the vendor's own dashboard.
AI & automationWhether the native AI is genuinely usable, what it costs at volume, and whether you can bring your own model or are locked to theirs.
Integration depthCRM, WFM, and ticketing — depth of native integration versus what you'll be building and then maintaining.
Compliance postureDTMF masking, recording controls, data residency, E911 handling for remote agents.
Administration modelWhether your existing team can operate it, or whether you're now hiring for a specialism.
Commercial structureConcurrent versus named licensing, telephony charges, AI metering, and where the overage risk sits.
Exit costWhat it takes to leave. Data extraction, recording portability, and how much of your logic is trapped in vendor-specific constructs.
The eleven dimensions weighted with you, before any vendor is seen
Capability 01Routing model 02Reporting depth 03Real-time APIs 04Telephony model 05Voice quality Operations 06AI & automation 07Integration depth 08Compliance posture 09Administration Commercial 10Cost structure 11 Exit cost scored by nobody at selection · discovered by everybody at renewal
Ten dimensions tell you which platform fits today. The eleventh decides what your renewal negotiation looks like in year three — if your routing logic, recordings and history can't leave, it isn't a negotiation.

The platforms

Five vendors were named Leaders in the 2025 Gartner Magic Quadrant for Contact Center as a Service: Genesys, NiCE, Five9, Amazon Connect (AWS), and Talkdesk — which joined the Leaders quadrant that year. Leadership placement tells you a vendor is credible at scale. It does not tell you which one fits your routing model, and that is the decision that actually matters.

Genesys Cloud

2025 MQ Leader Enterprise scale

Where it's strong

  • Deep, mature routing engine — usually the most flexible option when business rules are genuinely complex.
  • Broad channel and workforce engagement coverage in a single estate.
  • Long enterprise track record; recognized as a Gartner CCaaS Leader for eleven consecutive years as of 2025.
  • Strong option where a large, heterogeneous contact center needs one platform to cover everything.

What to watch

  • Capability breadth comes with administrative surface area. Budget for the specialism, or plan for a partner.
  • Licensing tiers matter — capabilities assumed to be included are often a tier up. Model the real configuration.
  • Migrations from legacy Genesys estates are their own project; don't scope them as an upgrade.

We tend to shortlist it when routing complexity is the binding constraint and the organization has, or will fund, platform expertise in house.

NiCE CXone

2025 MQ Leader WFO depth

Where it's strong

  • Workforce optimization and quality management are first-class rather than bolted on — historically the strongest suite in this area.
  • Positioned highest for Ability to Execute and furthest for Completeness of Vision in the 2025 Gartner CCaaS Magic Quadrant.
  • Analytics and interaction analytics depth suits large, measurement-driven operations.

What to watch

  • The suite's value depends on adopting most of it. Using it as a routing engine alone leaves a lot of license on the table.
  • Verify that the WFM integration you're paying for covers your forecasting model, not just headline scheduling.
  • Commercial packaging is complex enough that a line-by-line model is worth the effort before signing.

We tend to shortlist it when workforce management and quality are strategic rather than incidental, and the operation is large enough to use the suite properly.

Five9

2025 MQ Leader Mid-market fit

Where it's strong

  • Pragmatic balance of capability and operability — frequently the fastest of the Leaders to get into production.
  • Strong outbound and blended capability, which matters more than vendors' inbound-focused demos suggest.
  • Named a Leader again in the 2025 Gartner CCaaS Magic Quadrant for both axes.
  • Administration is approachable enough that existing contact center staff can usually own it.

What to watch

  • At the extreme end of routing complexity you may hit the model's edges sooner than with Genesys.
  • Check real-time API limits against your concurrency before committing to a custom desktop or wallboard.
  • Confirm the telephony commercial model — bundled minutes versus bring-your-own-carrier changes the total materially.

We tend to shortlist it when the operation is mid-market to upper mid-market, outbound matters, and time-to-production is a real constraint.

Amazon Connect

2025 MQ Leader Build-your-own

Where it's strong

  • Consumption pricing with no seat licenses — genuinely different economics for seasonal or highly variable volume.
  • Deepest customization ceiling of the Leaders if you have engineering capacity; the platform is effectively a set of building blocks.
  • Native access to the wider AWS estate for data, analytics, and model hosting.
  • Recognized as a Leader for a third consecutive year in the 2025 Gartner CCaaS Magic Quadrant.

What to watch

  • "No license cost" is not "no cost." Engineering effort is the real spend, and it is ongoing, not one-off.
  • Out-of-the-box supervisor and WFM experience is thinner than the suite vendors. Budget for what you'll build or buy alongside.
  • Without disciplined ownership, flow sprawl becomes unmaintainable quickly.
  • Model consumption costs against a realistic busy-hour profile, not an average.

We tend to shortlist it when the organization already runs on AWS, has real engineering capacity, and has variable volume that punishes seat-based licensing.

Talkdesk

2025 MQ Leader Vertical packaging

Where it's strong

  • Joined the Leaders quadrant in the 2025 Gartner CCaaS Magic Quadrant.
  • Industry-specific packaging — healthcare, financial services, retail — can genuinely shorten configuration time where the vertical fits.
  • Modern administration experience; generally quick for teams to become productive in.

What to watch

  • Vertical accelerators help when your process matches the template and add friction when it doesn't. Test against your actual workflow.
  • Validate integration depth for your specific CRM version rather than the marketed integration in general.
  • As a newer Leader, weight reference calls from organizations of your size and complexity more heavily than usual.

We tend to shortlist it when a client sits squarely in a supported vertical and wants configuration speed without a large platform team.

Cisco (Webex Contact Center)

Incumbent path

Where it's strong

  • Natural continuity for large existing Cisco voice estates — UCM, CUBE, and established network operations.
  • Strong story where contact center and enterprise collaboration are deliberately converged.
  • Familiar operational model for network teams already fluent in Cisco tooling.

What to watch

  • Assess the cloud product on its own merits. Continuity with your on-premises estate is a real benefit but it is not a feature comparison.
  • Migration paths from legacy UCCE/UCCX vary considerably by starting configuration — get yours assessed specifically.
  • Confirm which capabilities are native versus dependent on adjacent Cisco products you'd also be licensing.

We tend to shortlist it when there's a substantial Cisco voice estate, a capable network team, and convergence with collaboration is an actual objective rather than an assumption.

Microsoft (Dynamics 365 Contact Center)

Ecosystem play

Where it's strong

  • Compelling where the organization is deeply committed to Microsoft — Teams, Dynamics, Azure identity, and the surrounding data estate.
  • Agent experience benefits from tooling people already use daily.
  • Commercial leverage is often available inside an existing enterprise agreement.

What to watch

  • Contact center maturity is younger than the established CCaaS specialists. Test complex routing and historical reporting hard.
  • Be precise about which capabilities are native versus assembled from adjacent Microsoft products.
  • Voice architecture — carrier connectivity, media path, and E911 for remote agents — deserves specific scrutiny.

We tend to shortlist it when Microsoft alignment is a strategic constraint and contact center requirements are moderate rather than extreme.

UCaaS-adjacent options (8x8, Zoom, RingCentral)

Blended estates

Where they're strong

  • Single-vendor simplicity where UCaaS and contact center are bought together, with one commercial relationship and one support path.
  • Attractive economics for smaller contact centers inside a larger telephony estate.
  • Genuinely good fit where a modest contact center sits attached to a big UC deployment — a common shape that the specialist vendors serve expensively.

What to watch

  • Contact center depth is generally below the CCaaS specialists. Establish your genuine complexity ceiling before shortlisting.
  • Reporting and WFM are the usual gaps. Test them against your real requirements, not the demo dataset.
  • Bundling is a real benefit and also a lock-in mechanism. Score exit cost carefully.

We tend to shortlist these when the contact center is a component of a wider telephony decision rather than the center of it.

Four things we tell every client before an RFP

[ 01 ]

Weight before you score

Agree the criteria weights before you see any vendor. Weights set afterwards get tuned, consciously or not, toward whoever gave the best demo.

[ 02 ]

Make them build, not describe

A proof of concept should require the vendor to implement your three hardest routing scenarios. Anyone can describe a capability. Building it surfaces the caveats.

[ 03 ]

Model the busy hour, not the average

Consumption pricing, API rate limits, and concurrency licensing all behave differently at peak. Averages hide exactly the costs that hurt.

[ 04 ]

Write down why

Record the reasoning, not just the score. In three years the people who made the decision will have moved on, and the reasoning is what lets their successors revisit it honestly.

Running a selection?

We'll build the weighted scorecard and the proof-of-concept scenarios — and we don't get paid by whoever wins.

Book a call