About

Built to be checked, not just believed.

Mijares Group is a research and advisory firm. We build the analysis behind decisions that cost money, write the operating standards that follow, and measure whether either one held. Three divisions, one rule: name the source or do not make the claim.

Why the firm exists

The people who decide rarely live with the decision.

And the people who live with it rarely get a vote. Most advice arrives from somewhere above the work, confident, unsourced, and gone before anyone finds out whether it held. Mijares Group was built the other way round: the analysis is done to a standard a client can audit, the recommendation names what it rests on, and the failures get reported alongside the findings.

Much of the experience behind that came out of emergency medical services, an unforgiving place to learn a business: 24/7 operations, regulated to the hour, capital-heavy, thin margins, and staffed by people who leave if you get the culture wrong. Nothing about it is theoretical. What transfers is not the sector, it is the discipline of having to live with your own call.

What the firm does

Three divisions, one standard of proof.

01

Analytics

Demand, unit economics, competitive position and where the money is actually going. The analysis behind an expansion, a closure, a price or a site, with the uncertainty given as a number rather than a hedge.

02

Advisory

Organizational transformation on current best practice: culture, workforce and management development, operating standards with a measure against each one, and the change work that gets them adopted rather than filed.

03

Learning

Instruction and practice tools built from the federal practice guides, the published norms and the intervention literature, with every target read from the source report rather than quoted from a summary.

Who is behind it

Leo Mijares.

Operator, leader, executive, clinician and researcher. Executive director of an emergency services organization carried back to profitability with overtime cut 30 percent. Business development manager for a multi-site region. Manager of a training center running seven programs for four thousand students a year. US Army combat veteran and a service-connected disabled veteran. Pursuing a Doctorate of Business Administration at Purdue University Global, researching how public data can be turned into decisions a board can actually defend.

Two decades in the field across a dozen organizations, roughly eight of them spent running one rather than working in it. Most people see a single company and assume that is how things are done. Seeing a dozen tells you what actually transfers.

Running one

Accountable for the whole number

Executive director of a 24/7 emergency services organization running roughly 14,000 service events a year. Full accountability for the number: prepared and administered the operating and capital budgets, reorganized staffing and scheduling back to profitability without dropping a shift of coverage, and took 30 percent out of overtime.

  • The full personnel function: hiring, training, performance, discipline, termination
  • Licensing, the policy manuals, the metric set, and the QA and QI program
  • Investigations of incidents, personnel matters and complaints, to resolution
Growing one

Revenue, and where it comes from

Business development manager for a five-site region covering roughly 1,500 square miles. Built the referral relationships the volume actually comes from, and targeted the outreach with ZIP-code-level payer analysis rather than by calling on whoever was friendliest.

  • A standing review cadence with hospital leadership to close service issues
  • Relationships held across hospitals, EMS, dispatch, fire and local government
  • Growth strategy aimed by payer mix, not by territory habit

We retracted our own finding after a permutation test showed it was an artifact.

Why that matters

The standard is not negotiable when it is inconvenient.

A spatial-clustering result came out of the siting model that would have made a good slide. A permutation test said it was noise, so it was withdrawn and the withdrawal was published alongside the work. Nobody asked for that and no client would have known.

That is the whole proposition. Anyone can say they are evidence-led. The test is what they do when the evidence costs them something, and the only honest way to answer it is to show a case where it did.

Military service

US Army, 1998 to 2013.

Sergeant. Honorable discharge from both periods of active duty. Health care specialist and paramedic, Operation Iraqi Freedom, where he directed a staff of thirty running a 24-hour infirmary for more than two hundred detainees at Khadamiyah Prison in northern Baghdad, trained more than 1,400 coalition soldiers in combat lifesaving, and built mass casualty and incident command structures for ground evacuation in the Kuwait theater.

Later, lead instructor for the US Air Force Air Advisor Course, where he wrote the Combat Lifesaver curriculum and trained more than a thousand deploying air advisors. Service-connected disabled veteran.

Education and credentials

Still in school, on purpose.

01

Doctorate of Business Administration

Pursuing, at Purdue University Global, expected 2028. Applied research project in turning public data into decisions a board can defend, using design science method. The field case is rural air medical siting.

02

Master of Science

Management and Leadership, project management concentration, Purdue University Global, conferred October 2025. Bachelor of Science in Organizational Management, April 2024.

03

Clinical licensure

Nationally Registered Paramedic. Paramedic licensure across ten states. American Heart Association instructor for BLS, ACLS and PALS. Previously Flight Paramedic-Certified and a Certified Ambulance Coder.

04

Technical

Negative binomial regression, cross-validation and out-of-sample validation design. Public records and registry acquisition. Payer machine-readable file extraction at terabyte scale. Python and Power BI.

Working with us

One person, and that is the point.

There is no analyst two levels down whose work nobody has checked. The person who builds the model is the person who flew the aircraft, ran the payroll and will sit in the room when the number is challenged. If that is not the right shape for your problem, we will say so rather than take the work.