LodeGuard
Vectoring to the lode. Multivariate geochemistry with the math done right.
LodeGuard turns the multi-element geochemistry data sitting unused in your project database into a defensible vectoring analysis — without requiring you to have a statistician on staff. Drop in your CSV, get back what the data actually says about where your mineralization is, with a methodology memo your QP will actually sign off on.
The Problem
The Most Valuable Data on Your Project Is the Data You're Not Analyzing
Every junior holding a multi-element soil or rock survey knows this in the back of their mind.
The familiar pattern
You spent six figures on the survey. The lab returned 30 to 50 elements per sample. You built single-element threshold maps for gold and copper, picked the top three anomalies, and stopped. The other 28 elements went into a CSV that lives in a project folder that nobody opens. Somewhere in those 28 elements are the pathfinder signals that point toward the deposit you haven't found yet — but pulling them out requires statistical work nobody has time to do correctly.
The gap isn't methodological — multivariate geochemistry and compositional-data analysis have been textbook practice since Aitchison's 1986 work on the statistical analysis of compositional data. The gap is workflow. Doing it right requires log-ratio transformations, dimensionality reduction with the right conventions, clustering with defensible parameter choices, and a methodology disclosure your QP can sign their name to. Doing it wrong produces statistically biased outputs that any reviewer with training will see through.
Most teams skip the analysis entirely rather than do it wrong. LodeGuard exists so they don't have to.
What You Get
A Defensible Analysis, a Memo Your QP Will Sign
The math, done correctly. Every multivariate operation runs on log-ratio-transformed data — the textbook starting point that off-the-shelf GIS tools quietly skip. No raw-values shortcuts, no statistically biased outputs.
Defensibility built in. Every analysis ships with a methodology disclosure memo — what transformation, what components, what cross-validation, what assumptions. Drop it into your technical report and the QP can sign.
From multi-element noise to vectoring signal. Drop in your CSV, get back a ranked sample-by-sample map of where the data is pointing. Components, clusters, classifications — all interpretable, all spatially mapped.
No statistician required. The methodology is built in correctly so you don't have to know it. The product picks defensible defaults; the memo discloses what was picked; your QP audits the disclosure, not your code.
Who It's Built For
Exploration Teams Sitting on Underused Geochemistry
LodeGuard is built for the people whose property database has more analytical depth than the current workflow is using.
Project Geologists
Pull defensible vectoring out of the data you already have. Stop apologizing for the 28 elements that didn't make it into the targeting story.
Exploration Managers
Walk into the next board meeting with a real multivariate analysis on the property — not a single-element threshold map dressed up as targeting.
Qualified Persons
Sign off on multivariate vectoring in the next technical report without having to audit the underlying Python. The methodology memo is the audit trail.
Consultancies
Produce defensible multivariate analyses for your junior clients in a fraction of the time. The methodology memo template is consultant-ready.
The Numbers
What This Is Worth
In the Family
LodeGuard + AssayGuard + VistaGuard
The chain, working together
AssayGuard protects the integrity of your input data — per-batch QA/QC review in minutes, not quarters.
LodeGuard turns the validated data into vectoring signal — defensible multivariate analysis with a methodology memo.
VistaGuard turns the analysis into spatial output — maps, strip logs, sections, polygons, all reproducible and disclosed.
Use any one alone, or chain them for end-to-end coverage from lab return to QP-ready spatial deliverable. AssayGuard → · VistaGuard →
Why Us
Built by Geologists Who Read the Methodology Papers
Most exploration software is built by software companies who learn the geology second. LodeGuard is built the other way around — by people who have logged core, read Aitchison, and sat through the conversation where a reviewer asks "what log-ratio transformation did you use?" and the answer is silence.
Everything in LodeGuard is designed to be defensible to a working geochemist and readable by a working geologist. Methodology is disclosed, not buried. Defaults are conservative and cited. The methodology memo is structured so a QP can review it in five minutes and a reviewer can audit it in fifteen.
Limited Pilot Cohort Opening Soon
LodeGuard is currently in active build — the second product in our junior-explorer SaaS catalog, following AssayGuard. We're opening a small pilot cohort for the first version — exploration teams with completed or in-progress multi-element soil/rock geochemistry surveys who want to be the first to use it, shape its direction, and lock in early-adopter pricing.
Pilot cohort is limited. We're picking customers who'll get the most out of it and who'll give us the most useful feedback. Send us a note about your survey and your questions.
More on the Way
LodeGuard Is the Second of Several
LodeGuard is the second product in a planned catalog of focused tools for the junior exploration industry — each addressing a chronic workflow gap, each designed to be defensible to a working geologist. Join the early-access list to hear about the next releases first.