Thirty-six answers, including the awkward ones.

Two things in one codebase. A multi-tenant laboratory information system covering patient registration, test ordering, specimen chain of custody, result entry and validation, Westgard quality control and the audited release of signed reports. And a public marketplace where a patient finds a laboratory and a laboratory applies to be listed.

They are together because a marketplace is worthless if the laboratory behind it cannot produce a trustworthy result.

No. It holds the laboratory's view of a patient, not their clinical history, and it is designed to send results to whichever system does hold that record. It is also not a revenue-cycle system, not an instrument driver, not a quality management system and not a diagnostic device.

The web application is the surface you can actually use. An Android application is built from the same codebase and has never been published to a store, so today it is not something you can install. There is no iOS application; that platform was removed from the project.

Earlier versions of this site listed iOS availability. That claim has been removed.

No. There is no demo tenant and no trial, so neither is offered anywhere on this site. A trial button leading to a form that collects your details and produces nothing is a pattern this site does not use.

Specimen collection does. Seventeen custody actions queue on the device with the network off and reconcile on reconnect, with conflicts surfaced for a person rather than resolved by whichever write arrived last. That covers home visits and basements, which is where the requirement comes from.

Result entry, validation and release are online operations.

Not without a recorded override. Release is an enforced gate rather than a button: the control run is bound to the patient results produced beside it, an out-of-control run blocks their release, and the refusal names the Westgard rule that refused rather than reporting a generic failure.

An override is possible, because a laboratory sometimes has a legitimate reason, and it is an audited act with a person's name and a stated reason attached. Which conditions are hard blocks and which are overridable is your configuration.

A correction is a new row at the next version carrying a pointer to the row it supersedes. The superseded row keeps its released status, because it was released, and that stays true.

The issued report is a snapshot and is not silently rewritten, and whoever received the original is told about the amendment. An amendment nobody is told about is worse than the original error. There is deliberately no status called amended, because a status flip destroys the value a clinician actually acted on.

From your laboratory. LabFlow resolves against the intervals you have verified for your own methods and population, selecting by age, sex and method, and it records which interval it used so the interpretation travels with the result.

It does not ship clinical reference intervals as defaults. A vendor that did would be making your validation decisions for you, in a way that is invisible until it is wrong.

It compares this result against this patient's previous one for the same test. A haemoglobin that moved four grams per decilitre since yesterday is either a real emergency or a mislabelled tube, and both need a person immediately.

It is not predictive and not machine learning. It is arithmetic against a value already in the record, which is exactly why it can be trusted in a clinical setting and exactly why it is not described as artificial intelligence.

It is a communication obligation rather than a colour. The result carries an escalation ledger: who was told, when, by which route, and whether they acknowledged it. An unacknowledged critical value blocks release, and that state is stated on screen rather than implied by an absence, because absence is not something a person notices at three in the morning.

LabFlow records the call. It does not place it. There is no automated outbound telephony and no auto-faxing.

Partly. Calibration dates, due and overdue dates and each calibration’s result are recorded on the equipment register, and a failed calibration takes the analyser out of service. Nothing schedules a calibration and nothing sends a reminder. Earlier versions of this site described quality control and calibration as though they were one feature.

By status, and it is labelled every single time the value is shown, including in exported messages. A preliminary result that looks identical to a final one is a patient-safety defect rather than a display preference. There are eight result statuses in total, and pending, in progress, entered, preliminary, validated, released, rejected and cancelled are all real and separately queryable states.

No, and no software is. Compliance is a property of an organisation and the processes it runs, and there is no certificate a product can hold.

LabFlow is HIPAA conscious: audit logging, least-privilege access, tenant isolation and recorded consent are designed in. The word compliant does not appear on this site, and it was removed from every page where it previously did.

Not today. No BAA has been drafted or reviewed, and advertising a legal commitment that does not exist is worse than offering nothing. It was previously listed on this site as available and that claim has been removed.

If a BAA is a requirement for you, it is a blocker today and it is better established here than in month three of an evaluation.

No. Tenant isolation is enforced at the query rather than decorated at the interface, and every list read has to prove its tenant from its own filters rather than trusting the caller to have been careful. A query that cannot prove it is treated as a defect even when it happens to return the right rows.

The patient, the laboratory that produced them, and the clinician who ordered them. A receptionist can register and book and never sees a result value, which is a capability grant checked where the work happens rather than a hidden menu item.

Sharing beyond that is a grant the patient makes, with a scope and an end, and it can be withdrawn. Consent is recorded rather than assumed.

Yes. Released reports download as HL7 v2 and FHIR files, formats another laboratory system reads, plus the audit trail, which is rows in your tenant rather than a product feature you could lose access to.

This matters more than usual here, because LabFlow is built by the LabFlow team, with no larger company behind it. An export that only LabFlow can read would make that risk much worse.

No. There is no data sale, there never will be, and the free tier is not funded by one. It is subsidised by the paid shapes above it, which is the only arrangement that survives contact with a clinical record — a tier subsidised by nobody is a tier that disappears, and one subsidised by your data costs more than money.

Advertising is a separate question and the answer there is softer, so it is worth splitting them. LabFlow expects to carry adverts on marketing pages and non-clinical screens, and the Android app declares the advertising identifier permission for that reason. That is not the same thing as selling data: an advert placed beside a page is not a laboratory’s records leaving its boundary. No advert is loaded on a signed-in clinical route, and no clinical data reaches an advertising network on any plan. App permissions clause 05 and the privacy policy clause 04 are the binding versions.

Yes, as files, with no live feed. A released report downloads as an HL7 v2 message and a FHIR R4 bundle, and an HL7 v2 order file becomes a draft order a person confirms. The files come from the same records as the printed report, so the two cannot disagree.

No. LabFlow writes and reads files in the standards those products speak, which is not the same thing. No mapping has been validated against any of them, no connectathon has been attended and no vendor certification exists.

All four were previously named on this site as validated integrations. That claim has been removed, and it is the kind of claim that costs a laboratory six weeks when it turns out to be aspirational.

It is the network process that receives HL7 v2 messages over a socket on your own infrastructure. LabFlow has none: it exports and imports messages as files, with no live feed.

So moving a file between systems is yours to do, and a live feed is its own project. The features page says the same.

Not directly. LabFlow is not an instrument driver and does not ship middleware. It reads and writes HL7 files, not a live feed, for whatever already talks to your analysers, which in most laboratories is a middleware layer that exists and is staying.

Yes, with LOINC, alongside container and storage requirements. Orders reference the catalogue rather than free text, which is what lets a test ordered at one laboratory be recognised by a second one.

The catalogue itself is yours to configure. No national test menu is preloaded.

No price has been set, so none is published. Three plan shapes are described and a quote is given in conversation, reflecting engineering and infrastructure cost rather than a markup priced for a funding round.

Bring three numbers to that conversation and most of it is settled: tests per month, number of sites, and your retention obligation in years.

No. There is no payment gateway, no page takes card details, and no plan can be activated by a form. The Buy button on the pricing page is deliberately visible and deliberately disabled, with the reason written beside it. When purchase arrives it will be on the web only, and that button will start working.

Not today. A plan is recorded against every tenant and the check that would refuse a write over a limit is built, but it is switched off, so every limit described anywhere on this site is effectively unlimited right now.

Before it is switched on, anyone already running will be told in writing, with the numbers.

No. Tests processed, number of sites, retention and integration work are the drivers. Charging per user would price a laboratory for being adequately staffed, which is the opposite of what the product is for.

Partly. Invoices, payments and claim records are built: a laboratory raises invoices, records what arrived, and tracks each claim’s billed, allowed and denied amounts through appeal or resubmission, by hand. There is no 837P or 837I generation, no 835 remittance parsing, no ERA posting and no patient statements. Those were described on earlier versions of this site and none of them is built.

Whatever you use for revenue cycle today, you will still be using it. If revenue cycle is the requirement, LabFlow is the wrong product.

None. There are no laboratories in production, no pilot tenants and therefore no user count to publish. Earlier versions of this site said LabFlow served laboratories around the world and had hundreds of users across pilot tenants. Neither was true, and both have been removed.

Zero is the honest number and it is printed rather than omitted.

The LabFlow team. The same team writes the code, runs the deploy, reads the email and would sign a contract. A message reaches a real inbox rather than a ticket queue, which is a consequence of the size of the team rather than a service promise.

No, to both. One team, one time zone, no overnight rota and no escalation path past the team. Twenty-four seven phone support was previously advertised on this site and has been removed, along with a nightly backup guarantee that also did not exist.

A laboratory that runs overnight should weigh this seriously before anything else on this page.

Half of this answer is reassuring. Your data exports in formats other systems read, the audit trail is rows in your tenant rather than a feature, and nothing is locked behind a payment relationship that could lapse.

Half of it is not. There is no source escrow, nobody on call and no larger company behind the team. Continuity of the product is continuity of the LabFlow team's attention. The full version of this answer is on the about page, with both columns side by side.

No. The repository is private and written by the LabFlow team. There is no community, no contributor list and no licence granting you the source. This site previously described LabFlow as open source and built by a global community, and both statements were false.

If source availability matters to your procurement, it is a conversation rather than a checkbox, and no escrow arrangement exists today.

No. Delta checks are arithmetic against a previous result and Westgard evaluation is a published rule set. The code names them that way and so does this site.

They were previously described as AI-powered predictive analytics. In a clinical system that is the worst place to be vague, because it misleads a reviewer about how the number in front of them was produced.

No. On-premise deployment and a dedicated VPC are not offered at any plan shape or any price. Both were previously advertised. If your procurement requires the software to run inside your own boundary, that is a blocker today.

The audit trail, result versioning and the capability model are built. The validation documentation package that a Part 11 assessment expects is planned work and is not shipped, and calling it done would be the easiest and worst claim on this site.

An intact audit trail makes evidence easier to produce. It does not pass an assessment, and no vendor can honestly promise that it will.