A medical college buys a simulation platform. It is demonstrated impressively, procured after a long process, and launched with an announcement.
Eighteen months later, a minority of one batch uses it regularly, faculty are unsure how it relates to the curriculum, and the renewal discussion is uncomfortable because nobody can say what it changed.
This is the common outcome, and the causes are almost never technical. They are decided in the months before deployment, in four questions that procurement processes rarely ask.
Question one: what is it replacing or relieving?
A tool introduced alongside everything else is additional work, and additional work in a medical curriculum is absorbed by the students with the most time — who are usually the ones who need it least.
The question worth answering before purchase is what specifically gets easier. Fewer hours spent on generic remediation. A reduction in the faculty time consumed by supervised practice that does not require judgement. A gap in a rotation where students currently have unstructured time.
If the honest answer is "nothing gets easier, this is extra", adoption will follow the pattern above regardless of how good the tool is. That relationship — that adoption is determined before training begins — holds in clinical education exactly as it does in commercial deployments, as described in training staff on a system they didn't ask for.
Question two: where does it sit in the curriculum?
Optional resources are used by volunteers. That is not a criticism of students; it is a fact about a timetable with no free space in it.
The deployments that work attach the tool to something that already exists: a defined number of encounters before a rotation begins, a structured component of a posting, a required element of formative assessment.
The distinction that matters is between required and credit-bearing. Requirement produces use. Making it a graded component produces use and a set of second-order effects — anxiety, optimisation toward the score, and pressure on the tool to be defensible as an examination. Many institutions are better served by a requirement without a grade, at least initially.
Question three: who owns it internally?
Every deployment that survives has a named person whose responsibility it is. Every one that fails was owned by a committee.
The role is not technical. It is deciding which batches use it when, answering the questions faculty ask, noticing when usage drops, and being the person who knows what the data says. Without it, the tool is nobody's problem, and nobody's problem is what a renewal cycle quietly ends.
This should be identified before purchase, with time allocated. A person given the responsibility on top of a full load is not an owner.
Question four: what will you look at, and when?
Institutions frequently deploy without deciding what evidence would show it working — which makes the renewal conversation a matter of impressions.
Worth deciding in advance:
Usage. How many students, how many encounters, how distributed across the cohort. A high total driven by a keen minority is a different situation from broad moderate use, and the average hides which one you have.
Dimensional patterns. Where the cohort is consistently weak, which is the teaching signal — the argument in what faculty can see that exam scores hide.
Faculty time. Whether supervision effort shifted toward the cases that need judgement.
Student experience, collected honestly rather than through a satisfaction form administered by the department that bought it.
What is not worth promising in advance is an examination-result improvement attributable to the tool. Cohorts differ, curricula change, and a claim of that kind is not defensible from institutional data — a caution that applies to any vendor offering it.
Rolling out in the right order
A phased rollout is not caution; it is how the first batch's problems get fixed before they reach the fourth.
One batch first, with the owner present and paying attention. Expect the first fortnight to surface practical problems — device access, scheduling, students unclear what is expected.
Fix quickly. Small irritations resolved fast establish that the tool is responsive. Deferred, they teach students it is not, and that impression is hard to reverse.
Then widen. The second batch benefits from the first batch's lessons, and the owner is no longer improvising.
Faculty before students, or at least alongside. Faculty who first encounter it through a student's question are not in a position to support it.
The infrastructure that has to be checked
Unglamorous and it decides more deployments than pedagogy.
Devices — whose, how many, whether students are expected to use their own. Connectivity in the places where practice will actually happen, which for voice-based encounters is a real constraint rather than a formality. Whether a quiet space exists for a student to conduct a spoken consultation. Institutional data handling for student records.
The last one is worth settling early rather than at signature, and the general position is in security and data ownership in client systems.
What we build, stated plainly
SYNTAX is Truffaire's clinical simulation platform: 500 cases across five specialties and 43 sub-specialties, delivered as voice-based encounters where information must be elicited rather than presented, with structured review across nine dimensions.
We have an obvious interest in institutions deploying it, and the honest statement of what it does is that it changes the volume of practice a student can accumulate and produces structured evidence about how they reason. It does not replace clinical rotations, faculty teaching or supervised patient contact, and any deployment premised on that will disappoint.
The educational reasoning behind the design is in how clinical reasoning is actually assessed and, for the assessment framework it supports, competency-based medical education in practice.
Frequently asked questions
Should it be graded?
Not initially, in most institutions. Required without being graded produces use without the distortions that a mark introduces. Grading is a decision to take after a year of observed use.
How many encounters should a student complete?
Enough that a dimensional pattern is legible rather than anecdotal, which means a sustained cadence across a term rather than a burst before an examination.
What if faculty are resistant?
Ask what specifically they object to. Objections that survive a demonstration are usually accurate — most often that it adds work without removing any, which is a deployment design problem rather than an attitude.
Can it be used for assessment of record?
That is an institutional and regulatory decision, and it raises invigilation and identity questions that formative use does not. Most institutions should establish formative value first.
What does it cost to run beyond the licence?
The owner's time, device provision where students do not have their own, and faculty time to interpret the data. Budgeting the licence alone is the commonest under-estimate.
Where to start
Before evaluating any platform, write down what currently happens in the hours it would occupy, and who will own it.
If neither answer is available, the deployment is not ready — and that is worth discovering before procurement rather than at renewal.
If you are evaluating simulation for a college or teaching hospital, get in touch.