Inside the Cascade Model: How Zaro's Three-Tier Testing Actually Works
Put two blood testing panels side by side and the instinct is almost automatic: pick the one with more markers on it. More data, more coverage, better decision, right? Except that's not actually the question worth asking. The more useful question isn't "which panel has more," it's "what is each panel actually capable of telling me." Those aren't the same thing, and the gap between them is the entire idea behind Zaro's Surface, Signal, and Source tiers.
The question your panel is actually answering
Surface is built to answer one question: has something already gone wrong? It's a foundational snapshot: a Comprehensive Metabolic Panel, a full CBC with differential, HbA1c, fasting insulin, a complete lipid panel with hs-CRP, TSH, a full urinalysis, and ApoB, the cardiovascular risk marker most annual physicals skip entirely. These are almost all threshold markers: a value is either inside a normal range or it isn't, and by the time one of them crosses that line, the underlying process behind it has usually been running for a while.
That's where Signal does something structurally different. It keeps everything in Surface and adds Lp(a), NMR lipid particle fractionation, a full hormone panel, and a nutrient panel covering B12, iron, and magnesium. The particle-count markers are the interesting part: a standard LDL number is really an estimate of total cholesterol carried inside LDL particles, but it says nothing about how many particles there are or how large they are. Two people can have identical LDL numbers while one of them is carrying mostly small, dense particles that squeeze into artery walls more easily and mostly large, buoyant ones that don't. NMR fractionation is what actually resolves that difference, which is exactly why a normal-looking LDL number on Surface doesn't automatically mean Signal would come back clean too. Signal isn't asking whether something's already broken, it's asking whether something's quietly forming underneath a number that still looks fine.
Source pushes one layer deeper again: what's causing this. It keeps everything in Signal and adds TPO and thyroglobulin antibodies, fibrinogen, a broader autoimmune screen (ANA with reflex and ESR), and OmegaCheck. ANA and ESR work the same way autoimmune markers generally do: systemic autoimmune activity can be measurable in antibodies and inflammatory markers long before it's measurable in symptoms or organ-specific tests. The thyroid antibody panel works the same way for the thyroid specifically: autoimmune attack on thyroid tissue can be measurable in antibodies long before it's measurable in thyroid hormone levels themselves. Source is also the only tier that unlocks Biological Age (PhenoAge) and Heart Age (AHA PREVENT), two independent estimates of how your biology is aging relative to your actual birthdate.
Why "cascade" and not just "tiers"
A tier system implies you're paying more for more of the same thing, a bigger bucket of markers. A cascade implies depth: each level sits on top of the last and answers a question the one before it structurally can't. Surface tells you what's already crossed a line. Signal tells you what's forming behind a number that hasn't crossed it yet. Source tells you why. You don't need all three to get value from any one of them; you need the layer that matches the question you actually have right now.
What happens after the blood draw
A panel result by itself is just a number sitting in an app. Say Signal comes back and flags a low vitamin D reading. Zaro's Recommendation Engine ties that result directly to a specific action: a dosage recommendation sized to how deficient the result actually was, not a generic "take more vitamin D" note. That's step one, and it's where most testing platforms stop, a report handed over with the assumption that you'll act on it.
The harder problem is what happens next, because most people don't fail from lack of good advice, they fail from not following through on it consistently. That's what Zaro's Corrective Action Plan Engine, its adaptive notification engine, is actually built for. Instead of sending the same generic reminder on a loop, it checks in on that specific vitamin D recommendation, and if you say you haven't started yet, it asks why: out of stock, forgot, unsure of the dose. Whatever the actual blocker turns out to be, it adapts the next message around solving that blocker specifically, rather than repeating the same nudge that already wasn't working.
The loop, not the snapshot
Test, analyze, recommend, act, improve, then test again. That loop is the actual product. The cascade model determines how deep your test goes; the two engines determine what happens with the result afterward. Together they're built around one premise: a panel result only matters if it changes what you do next.
See pricing for all three tiers or explore how the process works end to end.
This article is for educational purposes and is not a substitute for professional medical advice. Always talk to a physician about your specific results and health history.