What is a Vuser and how is it licensed

A Vuser is a virtual user, a unit of simulated load that LoadRunner generates to imitate the behaviour of a real person interacting with an application during a performance test. It is not a named human, not a seat, and not a server. Understanding what a Vuser actually is matters because almost every inflated LoadRunner finding rests on a misreading of this single unit, and a buyer who can define it precisely is already most of the way to defending the count.

When a performance test runs, LoadRunner spins up many Vusers at once, each executing a script that mimics a user logging in, navigating, submitting, and waiting, so the application is exercised under realistic concurrent load. The license is measured in concurrent Vusers: the number that can be running at the same instant in a test. That single fact, concurrency rather than total population, is the hinge on which a defensible Vuser count turns. Because the EULA makes compliance the sole responsibility of the licensee, the buyer must be able to state what the genuine concurrent Vuser requirement is, and the evidence that proves it lives in the buyer's own test execution records.

What a Vuser is, and what it is not

A Vuser is a script driven simulation of one user session running for the duration of a test. It is fundamentally different from the named or concurrent human users that govern a product like ALM or Quality Center, and conflating the two is a common source of confusion. A Vuser does not correspond to an employee, a login, or a license seat for a person. It is a measure of how much synthetic load the tool can drive at once. The distinction between simulated load and human seats is exactly why the named versus concurrent debate that runs through named versus concurrent user counting in ALM audits does not map cleanly onto LoadRunner, which is licensed by load capacity rather than by who is allowed to log in.

The other thing a Vuser is not is permanent. A test that drives ten thousand Vusers for fifteen minutes does not create a standing requirement for ten thousand Vusers; it creates a brief demand that ends when the test ends. The license question is how many Vusers the organization needs to run concurrently across its real testing programme, not the largest number that ever appeared in a single run.

How LoadRunner licenses Vusers

Concurrent capacity, not total population

The Vuser entitlement is a ceiling on simultaneous execution. Two tests that each use five thousand Vusers but never run at the same time do not require ten thousand Vusers of entitlement; they share the same capacity at different moments. An audit that sums the Vusers used across separate tests, rather than measuring the genuine peak of concurrent execution, overstates the requirement. The way that peak figures get misused is the subject of how Vuser counting inflates a LoadRunner finding.

Protocol bundles and Vuser types

Vusers are scripted against protocols, and different protocol families can carry different entitlement terms. The scope of what a given Vuser bundle covers is examined in LoadRunner protocol bundles and license scope, which matters because an audit can claim a protocol is outside the entitlement when the bundle in fact includes it.

Cloud and on demand variants

Vusers can be provisioned in cloud and on demand models that behave differently from a perpetual on premises entitlement, and the metering of those models has its own pitfalls, set out in LoadRunner Cloud and on demand Vuser models. Knowing which model governs a given deployment prevents an audit from applying the wrong measurement.

How the definition shapes a defense under the four Rs

Respond. OpenText gives seven days notice before an audit and the right to copy relevant records. We take over the single controlled channel and preserve the test execution data, because the meaning of the Vuser count depends entirely on what actually ran concurrently, and that evidence can be lost if testing continues unmanaged after a notice.

Reconstruct. We build the effective license position by reconstructing genuine concurrent Vuser execution from the records, establishing the real peak of simultaneous load rather than the sum of separate tests, before any vendor measurement script runs.

Rebut. We challenge every line that treats a Vuser as a person, sums non concurrent tests, or prices a one off peak as the standing entitlement. The finding falls to the genuine concurrent requirement the records support.

Resolve. We settle on the reconstructed Vuser count and, where it serves the buyer, convert forward into an OpenPass agreement that records exactly how Vusers are defined and measured, so the definition cannot drift in a future review.

An anonymised outcome

The reason the definition is worth this much attention is the remedy behind a finding. On noncompliance the licensee is deemed to have acquired licenses at then current list price, owes back maintenance and support, owes first year maintenance on the new licenses, and reimburses the cost OpenText incurs performing the audit, so each Vuser disqualified by a correct definition removes a fourfold charge. In a recent engagement an opening Vuser count treated separate, non overlapping tests as if they ran together; once the records showed the genuine concurrent peak, the count fell sharply and the finding followed it down. The same logic produced our anonymised banking ArcSight result, where separating burst from sustained reduced the finding 70 percent from $6.0M to $1.8M, because the unit of measurement, correctly defined, was far smaller than the opening reading.

Define the Vuser, then hold the count to it

The lasting lesson is that a Vuser is a unit of concurrent simulated load, and once that definition is fixed, most of the inflation in a LoadRunner finding has nowhere to hide. A buyer who can articulate what a Vuser is, that it is concurrency rather than population, and who reconstructs genuine simultaneous execution, holds the count to the real requirement. To see how the cost flows from the count, read how much does a LoadRunner Vuser finding usually cost, and to address peak charging directly, read can OpenText count peak Vusers against your license. For the full method see our ALM and LoadRunner audit defense track and our complete OpenText audit defense playbook for 2026. If a LoadRunner finding has misread the Vuser unit, open a case.