For a healthcare COO, technology is not an IT project you sponsor and forget. It is the layer that either removes friction from clinical work or quietly adds it. A well-run electronic health record saves a nurse a few clicks per patient; a badly-run one costs them, and across thousands of daily encounters that decides whether staff go home on time and patients are seen safely.

So the useful question is never "should we adopt more technology?" It is "does this system make the work easier, safer, and cheaper to deliver — and can we prove it?" Everything below is built around that test, across the four areas that carry the most operational weight: your record system (EHR/EMR), interoperability, telehealth and monitoring, and data security. Your task is to make sure each is measured against clinical throughput and patient outcomes, not a vendor's feature list. Technology is one input into flow, sitting inside the disciplines of a broader healthcare operations guide.

The COO's real job with health technology

The failure mode in healthcare tech is treating an installation as the finish line. The system goes live, the project team disbands, and eighteen months later clinicians have built workarounds — spreadsheets, sticky notes, verbal handoffs — because the software never fit the workflow. On paper the transformation happened; in practice the friction just moved.

Weak looks like a COO who reports on "systems implemented" and "training sessions delivered." Strong looks like one who reports on documentation time per encounter, order-to-result turnaround, and how many clinicians still route around the official system — activity versus whether the work got better. The practical move: for every technology you own, name the specific task it is meant to make faster or safer, and put a number on it before and after. If you cannot name the task, you cannot justify the spend.

Getting your EHR/EMR to earn its keep

The electronic health record is the operational spine of a hospital or clinic. An EMR is the digital chart within one organisation; an EHR is designed to travel across settings. Whichever you run, the gap between a strong and a weak deployment is rarely the product — it is configuration, training, and ongoing tuning.

A weak deployment ships the vendor's default templates, forces clinicians through screens designed for billing rather than care, and treats every complaint as a training problem. Alert fatigue sets in: so many pop-up warnings fire that staff dismiss them reflexively, including the ones that matter — and physicians finish notes at home.

A strong deployment treats go-live as the start of a tuning cycle. You watch where clinicians get stuck, retire alerts that fire too often to be useful, and streamline the handful of note types that make up most of the volume. A clinic might, for example, cut a routine visit note from twenty required fields to eight by asking which fields anyone actually reads downstream — and hand each clinician back most of an hour a day. That is an operations win disguised as an IT setting.

Interoperability: making your systems actually talk

Interoperability is whether your systems can exchange data in a form the receiving system understands and can use. It is where healthcare technology most often disappoints, because a hospital rarely runs one system — it runs a record system, labs, imaging, pharmacy, scheduling, and billing, often from different eras and vendors.

Weak interoperability means data moves as a PDF or a fax a human re-keys, or not at all: a referral arrives with an attached document nobody imports, so the specialist re-orders labs the patient already had. Strong interoperability means the referral, labs, and imaging arrive as structured data the clinician can act on without re-entry.

The real, well-established standards to insist on are HL7 FHIR for exchanging clinical data through modern APIs, and DICOM for medical imaging. When you evaluate a new system, the load-bearing question is not "does it have an API?" but "does it speak FHIR, and will it share the specific data our clinicians need at the moment they need it?" A system that exports data only in a nightly batch file is not interoperable in any way that helps at the point of care.

Telehealth and monitoring as operations, not novelties

Telehealth earned a permanent place in care delivery, but its value depends on how you route it. Treated as a bolt-on, a video visit is just a phone call with worse audio. Treated as an operations decision, it matches the right visit type to the right channel, cuts no-shows, and reaches patients who cannot easily travel.

Weak telehealth offers video for everything and measures success by visit count. Strong telehealth decides which encounters suit remote delivery — medication follow-ups, stable chronic-disease check-ins, some behavioural health — and which need to be in person, then builds scheduling rules that steer each to the right place. Remote monitoring extends the same logic to devices: a connected blood-pressure cuff or glucose monitor only creates value if someone owns the data and acts on out-of-range readings. A stream nobody reviews is worse than none — it implies a duty of care you are not meeting.

Data security is an operations problem, not just IT's

Healthcare holds some of the most sensitive and most targeted data there is, and a breach or ransomware lockout is not an IT incident — it is a clinical emergency. When systems go dark, care slows, elective procedures get cancelled, and staff revert to paper they may not remember how to run.

That is why security belongs in operations. HIPAA and the HITECH Act set the baseline legal duties; SOC 2 is the assurance standard to require of vendors who touch your data. Encryption at rest and in transit, multi-factor authentication, least-privilege access, and regular audits are table stakes, not achievements. The COO's specific contribution is making sure controls are designed into workflows rather than bolted on so awkwardly that staff route around them — which is how most real breaches begin.

Two disciplines carry disproportionate weight. First, tested backups and a documented recovery plan, so an outage becomes hours of disruption rather than weeks — the muscle covered in any serious business continuity guide. Second, controls that assume people will make mistakes and limit the blast radius when they do. The deeper threat models and audit cadence live in a dedicated cybersecurity handbook; the COO owns whether those controls survive contact with real clinical work.

Budgeting and sequencing: what to fund, and when

Healthcare technology fails more often from bad sequencing than from bad products. Buying the exciting thing before the foundation is ready — an AI diagnostic layer on top of an EHR clinicians already hate — stacks new complexity on unresolved friction. The table below is a way to think about phasing, not a fixed formula; the right split depends on your starting maturity.

Optimisation and security are not a final phase — they run throughout. The most common budgeting mistake is treating training and tuning as a one-time launch cost rather than a standing line item; unmaintained systems decay into the workarounds this guide is trying to prevent.

Measuring whether the technology is working

A KPI dashboard for health technology should read like an operations report, not an IT one. Uptime and ticket counts are inputs; the outcomes are documentation time per encounter, order-to-result turnaround, no-show and readmission rates, patient wait times, and clinician-reported burden. If a system moves those the right way, it earned its place.

Pair those with adoption data. A system used by most clinicians as designed is worth more than a richer one a minority uses properly — value comes from consistent use, not installed capability. Instrumenting the workflow rather than just the server is the discipline at the heart of data-driven operations.

Choosing and managing technology partners

Vendor selection in healthcare rewards boring diligence over demos. The demo always works; the question is what happens in year two, when you need an integration built, a security question answered, or a bug fixed before it affects care.

Weak selection optimises for the slickest demo and the lowest sticker price. Strong selection weighs healthcare-specific track record, proven FHIR and DICOM support, security posture (ask for the SOC 2 report and read it), the real cost and timeline of integration, and the vendor's financial stability — because a vendor that folds leaves you stranded on an unsupported system holding patient data. Contract for the exit as carefully as the entry — how you get your data out, in what format, at what cost. The structured way to score and manage these relationships sits in a dedicated vendor management guide.

None of this succeeds without the people using it, which is why every deployment above is really a change project wearing a software badge. Clinical champions, honest talk about what gets harder before it gets easier, and involving frontline staff in selection do more for adoption than any feature — the playbook lives in these change management strategies.

Key takeaways

• Judge every health-tech decision by one test: does it remove friction from clinical work or add it — and can you prove it with a before-and-after number?
• The EHR is your operational spine. Value comes from configuration, tuning, and cutting alert fatigue, not from the product logo.
• Interoperability is the make-or-break layer. Demand HL7 FHIR and DICOM, and ask whether data arrives structured and in time to act on, not just "available."
• Route telehealth and monitoring by visit type, and give every data stream a named owner who acts on it.
• Treat security and backups as operations, not IT hygiene — a lockout is a clinical emergency. Design controls into workflows so staff don't route around them.
• Measure outcomes (documentation time, turnaround, readmissions, adoption), not activity (systems installed or sessions delivered).

Frequently asked questions

How do I know if our interoperability is actually working? Trace one real patient journey — a referral, a lab order, an imaging study — and count how many times a human re-keys or re-orders data that already existed elsewhere. Every manual re-entry is a gap. Working interoperability means the receiving clinician gets structured data they can act on immediately; a nightly batch file or a PDF nobody imports does not count. Is telehealth worth the operational overhead now that the initial surge has passed? Yes, when it's routed deliberately rather than offered for everything. The value comes from matching visit types to channels — remote for stable follow-ups and some behavioural health, in-person for anything needing examination — which cuts no-shows, frees exam rooms, and keeps hard-to-travel patients in care. Offered indiscriminately and measured only by volume, it adds cost without those gains. Where should data security sit — with IT or with operations? Both, but the COO owns the operational consequences. IT runs the technical controls; operations owns whether those controls survive contact with real clinical work, and what happens to care delivery when systems go down. A ransomware lockout cancels procedures and reverts staff to paper, so tested backups and a rehearsed recovery plan are operational readiness, not just an IT checkbox. How do I get clinicians to actually use new technology instead of working around it? Involve them in selection so the tool fits how they work, recruit respected clinical champions to lead adoption, and be honest that some things get harder before they get easier. Then measure adoption and act on the friction people report — the fastest way to lose credibility is to launch a system, disband the team, and ignore the workarounds that appear within months.