Blog
Dealer survey estimate vs satellite fleet count: how the two methods differA side-by-side look at how dealer survey estimates and satellite-derived fleet counts build a regional installed-base number, and where they disagree.
Field notes: the border town where two sample regions both claimed the same quarryA reconciliation note on what happens when two adjacent sample regions both count the same boundary-straddling quarry site.
How to set a region boundary for a satellite-based installed-base estimatePicking the boundary for a satellite installed-base estimate decides the math before you count a single machine. Here's how to set it.
What is fleet attrition, and why it changes your installed-base numberFleet attrition shrinks your installed-base denominator every year. Here's how retirement curves and write-off rates throw off a market-share figure.
Field notes: the quarter our regional total didn't match the boardA strategy analyst's field notes on defending an installed-base number the board didn't like, and why the denominator was never the problem.
What is a denominator problem in market share analysis?A denominator problem means your market-share math rests on a weak bottom number. Here's why TAM isn't installed base, and what makes a defensible one.
How many sites do you need to sample for a regional installed-base estimate?Sample size for a regional fleet estimate isn't one number. Here's how strategy teams think through stratification, coverage, and error.
What is sampling error in a satellite-based fleet estimate?Sampling error and occlusion error often get lumped together as one vague "imagery problem." Here's what each one is, and why the difference matters.
Field notes: the week a dealer survey and our pixel count landed worlds apartWhy a dealer-reported fleet number and a satellite-derived count can disagree, and what the gap tells you about each method.
How to pick a reference region for validating a satellite fleet countWhat makes a good ground-truth region for checking a satellite-based installed-base count before you trust it at regional scale.
How to scale a sample-region machine count to a national totalA stratified method for turning a site-level machine count into a defensible national fleet estimate, without the errors a flat multiplier introduces.
What is survivorship bias in equipment fleet sizing?Survivorship bias inflates installed-base estimates by undercounting decommissioned machines. Here's where it creeps in and how to catch it.
Adjusting for occluded equipment in satellite fleet countsHow to correct a raw satellite machine count for equipment hidden under cover, canopy, or stacking before it goes into a market-share denominator.
What is coverage correction in satellite-based fleet counting?Coverage correction explained: how satellite-based fleet counting adjusts for cloud cover and occlusion to build a defensible regional count.
How to estimate OEM market share when there's no registration dataNo registration data in your region? Here's how OEM analysts build a defensible market share denominator without relying on dealer surveys.
SOM vs installed base: two different numbers OEM teams confuseSOM is a forecast. Installed base is a count. Here's how OEM strategy teams tell them apart and why the denominator matters.
Shipment survival curve vs. site-count fleet model: which one to trustShipment survival curves and site-count fleet models answer different questions. Here's how each one builds a regional installed-base denominator.
What a parc study measures in heavy equipmentA plain explanation of what a parc study is, how equipment parc gets counted, and why the fleet population number is so hard to pin down.