Genesys Cloud CX vs VICIdial: the license model is the decision

Compare the license model, not just the sticker price

Fast answer: Genesys Cloud CX publishes four monthly per-user tiers -- CX 1 at $75, CX 2 at $115, CX 3 at $155, and CX 4 at $240, all in US dollars and billed annually, as published on genesys.com on August 5, 2026 -- but the number that actually decides your bill is which of three license models you buy under: named, concurrent (priced at a premium over named), or Hourly Interacting, a usage-based option billed for actual interaction time. Genesys states plainly that named and concurrent models may not be mixed on one account. VICIdial has no seat license of any kind: it is self-hosted, open source software, and what limits an operator is server capacity, not a per-person or per-seat fee.

This is a licensing-model comparison, not a verdict. Genesys Cloud CX is a cloud contact-center-as-a-service (CCaaS) platform -- CCaaS being the general term for contact-center software sold as a hosted subscription rather than software you install yourself -- and CX in its name is short for customer experience. This article does not rank the two products, state a VICIdial total-cost figure, or claim what any Genesys tier includes beyond its published pricing and license terms; it compares how each vendor's licensing structure responds to a real staffing pattern, specifically a shift-based floor where headcount and simultaneous logins are two different numbers.

In plain language: an agent is the person logged into a contact center seat taking calls, whether that seat is a VICIdial agent screen or a Genesys Cloud CX desktop; a campaign is one calling project -- which leads get dialed, by which agents, under which settings; a carrier is the phone company that carries a call to the public telephone network; a channel is one simultaneous call path a system can carry; a disposition is the outcome code an agent saves after a call, such as SALE or NOT INTERESTED; and dial level is the predictive-dialing setting that controls how many lines a system dials per available agent, a VICIdial concept with no direct equivalent in a licensed-seat CCaaS platform.

• Read every Genesys figure from genesys.com/pricing yourself; this article dates every number to August 5, 2026.
• Treat named-versus-concurrent as one account-wide choice, not a per-seat option.
• Measure your own distinct-agent and peak-concurrency numbers before either vendor conversation.

What Genesys actually publishes: four tiers and three license models

Genesys's own pricing page lists four Genesys Cloud CX tiers, all in US dollars, per user, per month, billed annually, as published on August 5, 2026: CX 1 at $75, CX 2 at $115, CX 3 at $155, and CX 4 at $240. Every tier also offers a “Get custom pricing” option instead of the listed rate, and the page states the same pricing in seven other currencies: Canadian dollars (CAD), British pounds (GBP), euros (EUR), Japanese yen (JPY), Australian dollars (AUD), New Zealand dollars (NZD), and South African rand (ZAR).

None of that is the interesting part. What sets Genesys apart from the other platforms in this comparison set is that it names three distinct ways to buy a seat on any of those four tiers: a named license, assigned permanently to one person; a concurrent license, drawn from a shared pool for the length of a session and priced at a premium over the named rate; and Hourly Interacting, a usage-based option that charges for actual interaction time rather than for holding a seat at all. Genesys states, without qualification, that “named and concurrent models may not be mixed” -- you commit to one or the other for the account, not a blend chosen seat by seat.

This article does not restate what each CX tier includes feature-for-feature, because that would mean asserting content from a page this research did not fetch. What it can compare, from the pricing page itself, is the shape of the decision every Genesys buyer has to make before a single feature matters: which of the three license models fits how your people actually work.

• Reconfirm price, currency, and license-model wording against the live Genesys page before a purchase.
• Do not assume tier contents beyond price and license model from this article; that was not fetched here.
• Note that Hourly Interacting has no published rate, unlike the four named/concurrent tiers.

Named vs. concurrent: the question a dialer operator actually has to answer

A named license tracks a person. Once assigned, it stays assigned to that person whether they are logged in, on a break, or off shift entirely; the next person who needs a seat needs a license of their own. A concurrent license tracks a session instead: it comes out of a shared pool the moment someone logs in and goes back into the pool the moment they log out, so the pool only has to be as large as the busiest moment across every shift, not as large as the roster.

That distinction is close to meaningless for an office running one eight-hour shift with little turnover, because named headcount and peak concurrency end up close to the same number anyway. It is not close to meaningless for a contact center running around the clock across time zones, or a business process outsourcer (BPO) with far more staff on the roster than seats filled at any one moment -- exactly the staffing pattern a predictive dialer is usually built for. There, named licensing prices the roster; concurrent licensing prices the busiest shift.

Genesys prices concurrent above the named rate per seat, and does not mix the two models on one account, so this is a single either/or choice for the whole floor, not something you can tune shift by shift. Work the two numbers out for your own operation before either conversation: how many distinct people log in across a representative month, and what is the highest number of them logged in at the same moment. The gap between those two numbers is what a concurrent license is worth to you.

• Compute your own distinct-agent and peak-concurrency numbers before naming a preferred license model.
• Remember Genesys does not mix named and concurrent on one account.
• Treat the concurrent premium as real but unpublished; get a current quote rather than assuming a multiplier.

WORKED EXAMPLE -- ONE 24-HOUR, THREE-SHIFT FLOOR Employees on payroll who take at least one call somewhere in the month: 90Highest number of them logged in at the same moment, any shift, measured not assumed: 40 Named-license math  A named license is assigned to a person, not a login slot, so it stays assigned whether  that person is on shift, on break, or off the clock entirely.  90 people who log in somewhere across the month need 90 named licenses, not 40, because  the license tracks the person, not the moment they are logged in. Concurrent-license math  A concurrent license is drawn from a shared pool for the length of one session and  returned to the pool at logout, so the pool only has to cover the busiest moment across  all three shifts, not the whole roster.  90 people sharing a 40-seat concurrent pool works, provided every shift's peak was  actually measured -- and Genesys prices a concurrent seat at a premium over the named  rate for exactly that flexibility. The crossover  Named licensing suits a floor where most of the 90 are logged in most of the time, so  the named count and the concurrent peak sit close together.  Concurrent licensing suits a floor where the roster is much larger than any single  shift's peak -- the normal shape of a three-shift operation with real turnover.  Named and concurrent models cannot be mixed on one Genesys account, so this is one  either/or choice for the whole floor, not a shift-by-shift one. 

The third option: paying for interaction time instead of a seat

Genesys's third model, Hourly Interacting, does not charge for a seat at all; it charges for actual interaction time, in the same way a metered utility charges for what you used rather than for a connection you keep whether or not you use it. Genesys's own pricing page does not publish an Hourly Interacting rate, so no figure can be stated here; what its page does state is that the option exists on every tier alongside named and concurrent.

An hourly, usage-based model is the closest thing in Genesys's own published structure to how VICIdial actually behaves once it is running: you do not pay a company for the right to have a person logged in, you pay for what a channel is actually doing right now. The difference is that VICIdial's version of that idea is not a subscription meter at all -- it is a server you own, and what caps your usage is hardware capacity rather than an hourly invoice.

VICIdial's side: no seat license, a hardware ceiling instead

VICIdial charges nothing per seat, per name, or per concurrent session, because there is no license to buy: it is self-hosted, open source software, and every agent who logs in does so against capacity the operator already owns. There is no named-versus-concurrent decision to make on the VICIdial side, because there is no license model of any kind -- the trade a VICIdial operator makes instead is owning the server, the carrier relationship, patching, backups, security, and upgrades in exchange for not paying anyone per seat.

That does not mean VICIdial capacity is unlimited; it means the ceiling moves from a licensing model to a hardware one. A single VICIdial box's Asterisk media configuration opens a fixed range of RTP ports for call audio, and that range, not the agent headcount, is what actually caps how many calls can be in progress on that box at once. The width of that range is a per-install setting, not a published constant -- check rtpstart and rtpend in /etc/asterisk/rtp.conf on your own server before assuming a number for it.

That per-box ceiling is the same shape as a Genesys concurrent pool, just measured in open ports instead of purchased licenses: both cap how many people can be doing something at the same instant, independent of how many people are on the roster. The guide titled Scaling VICIdial: designing a multi-server cluster for 500+ agents works through what happens once a single box's ceiling is not enough -- splitting database, dialer, and Asterisk roles across multiple servers rather than raising one number on one box -- which is the self-hosted answer to the same problem a Genesys buyer solves by paying for more concurrent licenses.

• Confirm your own rtpstart/rtpend and firewall range on your own server; this article does not assume a number for either.
• Raise rtpstart, rtpend, and the firewall rule together; changing one alone produces one-way audio.
• Treat the multi-server cluster guide as the next step once one box's ceiling is the real constraint.

Run the numbers before either conversation: distinct agents vs. peak concurrency

The named-versus-concurrent decision, and the VICIdial capacity-versus-headcount question right alongside it, both come down to the same two measurements: how many distinct people log in over a representative period, and the highest number of them logged in at the same instant. Guessing either number in a vendor conversation is how an operator ends up licensed for the wrong shape of staffing.

vicidial_log alone can answer both questions without assuming which column marks a session's start or end. Counting distinct agents over a rolling 30-day window approximates the roster; counting distinct agents active within the same clock hour approximates the busiest shift -- an approximation, not an exact simultaneous-session count, but close enough to tell a named shape from a concurrent one, the same two numbers you would ask a Genesys sales engineer to model against named or concurrent pricing. Both statements below are scoped to one fixture campaign for a safe capture on a shared lab; point them at your own real campaign on your own install to see your real roster.

• Run both statements against a read-only account, not a production write account.
• Scope both queries to the same rolling 30-day window before comparing the two numbers.
• On your own install, replace the fixture campaign filter with your real one before trusting either count.

SELECT COUNT(DISTINCT user) AS distinct_agents_last_30_daysFROM vicidial_logWHERE campaign_id = '<CAMPAIGN_ID>'  AND call_date >= DATE_SUB(NOW(), INTERVAL 30 DAY); SELECT DATE_FORMAT(call_date,'%Y-%m-%d %H:00') AS hour_bucket, COUNT(DISTINCT user) AS agents_active_this_hourFROM vicidial_logWHERE campaign_id = '<CAMPAIGN_ID>'  AND call_date >= DATE_SUB(NOW(), INTERVAL 30 DAY)GROUP BY hour_bucketORDER BY agents_active_this_hour DESCLIMIT 5;

What to weigh next, without a vendor verdict

This article states no price for Genesys concurrent or Hourly Interacting licensing, because Genesys does not publish one, and it states no VICIdial total-cost figure, because that number depends on hosting, carrier rates, and staffing that vary by operator. What it can leave you with is the actual shape of the decision: a named-license model prices your roster, a concurrent model prices your busiest shift, and VICIdial's self-hosted model prices neither -- it prices a server, sized against a hardware ceiling instead of a headcount.

If you are weighing what it costs to own that server yourself, the guide titled VICIdial VPS requirements and cost: separate a lab floor from production capacity works through the same sizing question a VICIdial operator faces before any licensing conversation. If owning the box is the part you would rather not do, the guide titled Managed VICIdial hosting vs self-hosting: a decision framework, not a vendor ranking sets out the same trade-off Genesys's Hourly Interacting option gestures at -- paying someone else to carry the operational load -- without turning it into a marketing comparison.

Neither path is stated here as cheaper, more reliable, or better supported than the other. Measure your own distinct-agent count and peak concurrency first; that pair of numbers is what makes a named-versus-concurrent quote, or a self-hosted capacity plan, mean something instead of a guess.

• Decide whether you are pricing a roster (named), a peak shift (concurrent), or a server (VICIdial) before requesting a quote.
• Read the linked hosting and capacity guides before assuming self-hosting is automatically cheaper.
• Revisit this comparison whenever your shift pattern changes materially, not just once at signup.