TeleTracking

Most of this platform would function without a model, and the company says so in its own framing: more than thirty years of operational expertise integrated with advanced artificial intelligence. The expertise came first and the models arrived recently.

Strip out the artificial intelligence and what remains is a working business. Bed management, transport dispatch, transfer coordination, referral routing, perioperative workflow and real time location tracking are workflow automation and logistics, and they manage a reported 200,000 beds today. The artificial intelligence product, Decision IQ, launched in November 2025 as one module among seven and advises rather than executes.

Membership was decided with this grade rather than the reject filter, following the Aetion precedent, and neither the horizontal nor the services filter applies to a healthcare only software company.

Recording the grading judgement openly because it is close. A D is defensible on the reasoning above. Resolved upward to C because the model does carry the consequential decision in the product the company is now leading with, the analytics layer is a genuine learned component rather than reporting, and this is the first incumbent operations platform graded on this axis in the lane, so the least disruptive resolution is the one that keeps it in band with comparable platforms rather than setting a new floor.

The boundary is unambiguous and nothing else about it is published. Decision IQ is decision support: it gives forward looking visibility so teams can anticipate where throughput problems will occur and coordinate across departments, and people act on it. Nothing is executed autonomously, no bed is assigned and no discharge is triggered without a human, and the company does not claim otherwise.

Clarity about a fully advisory posture is worth something, particularly in a domain where an autonomous system moving patients would be alarming.

Everything that would let a buyer assess the advice is missing. No accuracy, precision or forecast error measurement is published for the throughput predictions. No confidence expression, no statement of how far ahead the model forecasts reliably, no false alert rate, and no description of what a user sees when the model is uncertain. For an operations tool the practical failure mode is alert fatigue, where staff stop acting on predictions that do not materialise, and nothing published addresses it. Ask for forecast accuracy at the horizons the product operates over, and what the false positive rate looks like in production.

Nothing about the models is published. Decision IQ is described entirely by what it accomplishes, replacing retrospective data with proactive system wide visibility, and never by how. No model class, architecture, technique, training data, version or method is named anywhere.

The strongest statement located is that the company integrates the most advanced artificial intelligence technology in the industry, which is an adjective and a superlative with no comparator, and this index records rather than credits that form of claim.

Inputs are partially inferable from the platform description, since the system ingests admission, discharge and transfer events, real time location data and enterprise operational data, but the company does not state which of those feed the predictive layer.

No accuracy figure of any kind appears for any product. Graded D because there is no technical disclosure to assess, not because what exists is thin. A single paragraph describing the modelling approach and one published forecast accuracy figure would move this materially.

No party in the chain is named. No foundation model provider, model class or version, no cloud platform, no sub processor list, and no statement on whether customer operational data contributes to model development.

The move to a hosted model makes the omission more consequential rather than less. The company completed a transition from installed software to a hosted platform during 2024, explicitly eliminating customer hardware and hosting investment, which means operational and patient movement data that previously stayed inside the hospital now sits with the vendor. Who provides that infrastructure is unstated.

One question is specific to this product category. A capacity forecasting model improves with volume and variety of operational data, and this vendor sits across a reported 200,000 beds internationally. Whether models are trained per customer or pooled across the installed base is the central supply chain question here, it has competitive as well as privacy implications for a hospital, and it is not addressed in either direction. Ask whether models are tenant isolated or pooled, and for the sub processor register.

Enormous installed base, and essentially nothing measured about the artificial intelligence.

The scale claims are substantial and consistent across sources: more than 200,000 beds managed globally, millions of patient stays annually, more than three decades of operation, customers across North America and Europe including the United Kingdom and Canada, and independent technology tracking placing the deployed footprint in the hundreds of organisations. That is a deeper operational track record than any recent entrant in this lane can show.

For Decision IQ specifically the evidence is one named early adopter, University of Louisville Health, with an attributed quote from a named chief nurse executive describing a forward looking view of operations. That is a real reference with a real person behind it and it contains no outcome figure. Other customer references on company material describe transformed patient flow and significant operational efficiency gains without numbers.

No third party research assessment, buyer survey score, peer reviewed publication or independently audited measurement was located for the artificial intelligence product. Graded C on installed base and one named reference. Ask University of Louisville Health what changed quantitatively.

No stewardship material was located. There is no retention schedule, encryption statement, access control description, data ownership or deletion position, data minimisation commitment, or statement on whether customer data contributes to model development. As on the privacy axis, this is graded conservatively because the pass was targeted and a vendor of this tenure plainly documents these controls somewhere.

The data held is broader than most records in this index and is not primarily clinical. Admission, discharge and transfer events, bed assignments, transport requests, transfer decisions and real time physical location traces describe where identifiable people were, when, and who moved them. That is a movement history for patients and, where staff badges are tracked, for employees.

The hosted transition completed in 2024 moved that data from hospital infrastructure to the vendor, which is a material change in custody that the announcement framed as removing hardware cost rather than as a data handling change. Nothing published describes what changed about retention, access or segregation when it happened. Ask what the hosted model retains, for how long, and who inside the vendor can query location history.

No health privacy material was located in a dedicated pass: no compliance statement, control enumeration, de identification position, or business associate agreement posture, template or execution requirement.

Graded conservatively rather than punitively, on the same basis applied to Optum Integrity One. A vendor operating for three decades across a reported 200,000 beds, including public health systems in Europe, has necessarily satisfied privacy review at every one of those institutions and executes agreements as routine practice. What the pass established is that none of it is published where a prospective buyer would look, and the search was targeted rather than exhaustive.

One dimension here has no counterpart elsewhere in this index. The platform incorporates real time location technology, which tracks the physical position of patients, equipment and in many deployments staff. Location data about identifiable people is regulated differently from clinical documentation, staff tracking raises employment and works council questions particularly in European deployments, and nothing published addresses either. Ask what location data is retained, for how long, about whom, and how European deployments handle it.

No credential of any kind was located. A dedicated retrieval pass targeting certifications and compliance material found no controls report, no information security certification, no health security framework certification and no cloud authorisation. There is no trust center, no security page, no report availability process, no penetration testing disclosure and no vulnerability disclosure policy.

The strongest security statement located is that the hosted platform offers industry leading security, an adjective published in a product launch announcement with nothing behind it.

Graded D rather than the conservative C applied to Optum, and the distinction is deliberate. Optum's estate is vast enough that a product scoped pass is genuinely partial. This is a focused company with a single corporate property and a defined product line, so a targeted certification search returning nothing is materially stronger evidence of non publication than it would be for a conglomerate.

The absence is almost certainly a publication failure rather than a controls failure, since public health systems in Europe and large United States systems do not deploy platforms handling patient movement data without security review. That does not change the grade, because this index measures what a buyer can verify before contacting sales, and this is the easiest grade on the record to move.

No device pathway applies and none is claimed. Bed management, transport dispatch and transfer coordination are operational logistics rather than clinical determinations, so the absence of a clearance is correct.

The boundary is less clean than it looks and the company does not address where it sits. Throughput optimisation produces recommendations about discharge sequencing, admission acceptance and transfer prioritisation, and each of those interacts with clinical judgement about whether a patient is ready to leave, whether a facility can safely accept, and which transfer is most urgent. A model that advises on flow is advising on decisions with patient safety consequences even though every individual decision remains a human one.

Nothing published describes clinical governance over the recommendations: whether clinicians review the logic, what the product does when operational efficiency and clinical caution point in opposite directions, or whether any clinical safety case exists for the forecasting layer. Competitors in operational artificial intelligence face the same question and it is unusually pointed for a product explicitly designed to accelerate discharge. Graded C because the regulatory classification is correct and the clinical governance question is untouched.

Nothing is published. No bias or fairness testing, no model validation methodology, no monitoring output, no distribution reporting, no drift detection and no external audit was located for any predictive component.

The equity question in capacity artificial intelligence is real, specific and different from the one this index usually records, so it belongs on the file. A system that prioritises transfers, sequences discharges and allocates beds is deciding whose care moves first. Optimising throughput systematically favours shorter, more predictable stays, which correlates with patient complexity, comorbidity burden, social circumstances affecting discharge destination, and insurance status affecting placement options. A model trained on historical flow will reproduce whatever the historical pattern was, including any part of it driven by which patients were easy to place rather than which were clinically ready.

None of that is addressed. There is no statement that fairness was considered, no breakdown of recommendations by patient population, payer or facility, and no disclosure of what the optimisation objective actually is. Graded D because a product whose core function is allocating scarce clinical capacity publishes nothing about how it allocates. Ask what Decision IQ optimises for, and whether recommendation distributions have been examined by patient population.

No performance figure is published, so there is no stated level against which a shortfall could be measured and nothing to hold the vendor to. No forecast accuracy, precision, error rate or confidence measure appears for the predictive product, and no operational outcome figure is attached to the named early adopter.

No service level agreement, warranty, indemnity or remediation commitment was located. No pilot, benchmark or validation offer of the kind several competitors publish was found, so a buyer has no published route to establishing forecast quality on their own data before committing.

The recourse question in this category is unusual and worth stating precisely. The exposure created by a wrong capacity forecast is not a claim denial but an operational one: a unit staffed for demand that did not arrive, or a discharge sequence that left the emergency department boarding. Those costs land on the hospital and are difficult to attribute to any single recommendation, which makes a contractual performance commitment more valuable here than in domains where errors are individually traceable, and none exists.

One pre emptive note for future passes: further scale or installed base figures cannot move this grade. Bed counts measure adoption, not forecast quality. Only a published accuracy measurement or a contractual commitment on one will change it.

Integration is foundational to the product and is nowhere specified. Bed management cannot function without admission, discharge and transfer feeds from the record system, and the platform demonstrably consumes them across a reported 200,000 beds, so the capability is established by deployment rather than by claim.

What is absent is every detail. No record system is named, no interface standard is described, no messaging mechanism is specified, no vendor marketplace or programme listing was located, and nothing states whether the platform writes back or only reads.

The integration surface is unusually wide and equally undocumented. Beyond the record system it spans real time location infrastructure, which is a hardware ecosystem of tags, sensors and gateways from third parties, transport and environmental services systems, and an external referral portal connecting non affiliated organisations. Each is a distinct integration class and none is described.

Graded C for capability evidenced by scale with no published specification, consistent with how the same absence was treated for SmarterDx and Infinx. Ask which record systems, through what standards, and which location technologies are supported.

The deployment model itself is published, which is more than most records here manage, and residency is untouched.

What is stated is a genuine architectural fact rather than a slogan. During 2024 the company completed a transition of the Operations IQ Platform from installed software to a hosted model, launching the hosted version of its capacity product in October and describing the change as eliminating customer hardware and hosting investment while enabling immediate feature updates. A buyer therefore knows the delivery model and knows it recently changed, which matters for an existing customer evaluating migration.

Nothing states where it runs. No cloud provider, region, tenancy model, residency commitment or customer controlled option was located.

Residency is a live question here rather than a formality because the customer base spans North America and Europe. European hospitals operate under data protection rules with specific requirements on transfers outside the region, and a platform holding patient movement and location data for United Kingdom and European deployments has to answer where that data sits. Nothing published does. Ask for regions by geography, the tenancy model, and whether European deployments are hosted in region.

Nothing about cost is published. A dedicated pass located no pricing page, no unit of charge, no range, no implementation or onboarding fee position, no minimum commitment, no pilot terms, no return calculator and no percentage saving figure.

The one economic statement located is directional and concerns the delivery model rather than price: the hosted platform eliminates costly hardware and other hosting investments that the installed version required. That is a genuine cost structure change for an existing customer and it carries no figure and says nothing about subscription cost.

Two questions matter more here than for a point solution. The platform is modular across at least six named products, and nothing indicates whether they are licensed separately or as a suite, which determines what adding the artificial intelligence module costs an existing customer. And a bed based footprint suggests capacity linked pricing, which is unconfirmed. Ask for the pricing basis, whether Decision IQ is an upgrade or a separate licence, and what the migration from installed to hosted costs.

The broadest operational footprint of any record built recently, and one of very few in this index that reaches beyond the United States.

Coverage spans the acute care patient journey rather than a single workflow: referral intake, inter facility and intra facility transfer, bed assignment and capacity management, perioperative flow, transport, discharge and enterprise analytics. The company describes customers as rural, urban, national and international, and the deployed base covers North America and Europe with United Kingdom and Canadian customers specifically identified.

That international presence is worth flagging because almost every other record in this index carries a note that nothing addresses regimes outside the United States. Patient flow is less nationally bounded than coding is, which is why this vendor can operate across jurisdictions where a coding engine cannot.

Held at B rather than A because coverage is described by workflow rather than by care setting depth, nothing distinguishes performance in academic medical centres from community or rural hospitals despite both being claimed, and no specialty level detail exists for the perioperative module. Ask for reference sites by hospital size and country.