Resolution Layer Case Study · Read & Diagnose
Agricultural equipment.
A catalog read across the used and new agricultural machine inventory of a multi-branch equipment dealer — tractors, combines, headers, planters and implements, where one listing is one physical machine and a buyer decides on its hours, its warranty and its price before anything else.
This is a real EKOM catalog analysis, with the client's identity removed. The client is a multi-branch agricultural equipment dealer selling new and used machines through a public inventory search. EKOM read every unit in its agricultural inventory, raised 22 findings, and reports seven here. Every extent was recomputed twice by independent code paths; the automated findings' own counts ran low and are not used.
What makes an equipment inventory different from a product catalog is that one row is one physical machine. It has its own stock number, its own serial, its own meter reading and its own warranty, and a buyer comparing two of them is comparing those numbers — not a feature list. The hours decide the price. The warranty decides what the buyer is buying. The title decides what a search engine thinks the machine is. When a field on a machine's record contradicts another field on the same machine, there is no sibling to check it against; the record simply says two things.
A product page that is wrong misdescribes a product. An inventory record that is wrong misdescribes a specific machine on a specific lot — and the number it gets wrong is usually the one the price is negotiated on.
Stock numbers, serials, makes, models, branch locations and the dealer's own category vocabulary have been removed or generalized here, because on a dealer inventory a make, a model and a location together identify the dealer. Counts are exact, and the field values quoted are the ones the record carried, except where a value is rounded so that it does not point at one machine.
What's inside
- At a glance — the severity split, where the findings concentrate, and how the read was done.
- What a buyer runs into first — planters and headers carrying engine hours, and a placeholder warranty date on hundreds of units.
- Titles and codes that name the wrong machine — harrows titled as rock pickers and sprayers, a cutter coded as a tire.
- Prices and new units — one-cent machines, units with no price anywhere, and new machines carrying working hours.
- Where it traces back, and what's next — the four mechanisms most of these findings read as, and the order to fix in.
Nothing was supplied by the dealer. Everything below was read from what its public website publishes and the search behind it, and every figure was recomputed twice before it was written down.
7
Findings reported
(22 raised)
315
Units with a warranty
that "expired" in 1975
Where the 7 concentrate
Theme
What it is
Findings
Hours on machines with no engine
Planters and headers carrying engine hours; implements reading tens of thousands
1
A placeholder warranty date and a flag that says No
315 units expiring in 1975; every unit flagged as having no warranty
1
Titles that contradict the listing
36 units whose title and name disagree
1
Classification codes that contradict the machine
A cutter coded as a tire, a tiller as a sprayer
1
Placeholder prices
One-cent machines; units with no price anywhere
1
New machines carrying working hours
17 units flagged new with more than 1,000 hours
1
Smaller items
Question-mark serials, blank flags, brand and make spelled two ways
1
How EKOM read this inventory
1
Read
The dealer's own public site search, category by category — nothing supplied, no integration.
One row is one physical machine: its own stock number, serial, meter and warranty.
→
2
Profile
Profile the inventory's fields, then test each machine against them.
Hours, warranty, price and classification codes sit in separate fields — which is what makes a disagreement detectable.
→
3
Analyze
Surface defects, rate severity, and bind each one to the machines and values behind it.
7 findings, 4 possible causes.
Machine-surfaced signals from a cold read, with every figure recomputed — not a dealer-verified defect list. Some will prove intentional, which is why each is tied to the values behind it. Read point-in-time; an inventory turns over daily.
What a buyer runs into first.
The lead finding is about hours — on a used machine the number a buyer reads first, and on an implement with no engine the number that should not be there at all.
Critical · Hours that cannot be engine hours
A planter reading over 40,000 hours.
22 of the dealer's 26 planters carry an engine-hours value, though a planter is a towed implement with no engine; five read from some 12,500 up to over 40,000. On the draper, corn and flex headers, 74 of 109 carry engine hours, among them a recent-model header near 18,000. The dealer's own classification codes say these are implements, so the hours field is the one that is wrong.
Field
What the record holds
Verdict
Equipment type
planter — a towed implement, classified as one
agree
Engine
none
agree
Engine hours
a value over 40,000
cannot be true
Also in the hours field
Combine hours on tractors. One hour on a header.
29 tractors, balers, planters and seeders carry separator hours, which belong to combines. And 106 units read exactly "1 Hour": 75 are used, and 63 are equipment with no engine — headers, disks, tires, grain carts and planters. The large values on planters and balers read as acre or bale counts stored in an hours field.
A buyer who sorts or filters by hours is using the one field that is supposed to be arithmetic. One impossible value near the top of that sort says the sort cannot be trusted, and that doubt does not stay on the planter.
A placeholder where a warranty should be.
A warranty is a date and an hour limit. On hundreds of units the date appears to be the ERP's empty value, published as if it were real.
Critical · A placeholder date standing in for no warranty
315 units carry a warranty that expired on 1 January 1975.
Each carries the same date in its extended-warranty field, and 35 carry it in the basic-warranty field too. The same date on every one of them fits an empty value stored as the earliest date the system will hold; it is the analysis's reading, not something the record states.
Critical · A flag that says No across the category
"Has warranty" reads No on every unit — including 255 whose warranty has not expired.
Those 255 units carry a basic or extended warranty that has not yet expired, and the flag says No. Only six units across the dealer's whole inventory say Yes. The field is built as a facet, so any warranty filter or feed built on it would find almost nothing.
Field
What the record holds
Verdict
Warranty expiry
a date after today
covered
Has warranty
No
contradicts
The two halves have different fixes. The 1975 date is derivable from the record: an empty value can be read as "no date" and published as blank. The flag needs the dealer's own warranty records, because only they say which units are covered.
Titles and codes that name the wrong machine.
Two themes where the listing and the record behind it name different things.
High · Page titles that contradict the listing
Harrows titled as rock pickers and sprayers. A wheel tractor titled as a track tractor.
On 36 units the page title and the listing name disagree, and they differ in kind: 11 name the wrong type of equipment, 2 name a different maker, 5 carry a different or missing year, and the remainder are wording differences. Five harrows are titled as rock pickers, though their classification codes and their own options say harrow; two more harrows are titled as sprayers and filed only under sprayers; a vertical tiller is titled as a header trailer; a wheel tractor is titled and filed as a track tractor. A title is what a search engine and a shared link show first.
High · Classification codes that contradict the machine
A cutter coded as a tire. A tiller coded as a sprayer.
A rotary cutter carries the group and base codes for tires. A vertical tiller carries the base code for sprayers, though its group code says tillage. An air seeder is named as a compact-tractor attachment and coded to the compact-equipment product line. These codes sit in the public record next to the title: a report or a feed that groups by them puts a cutter in the tire count.
In each case the correct value is already in the record: the code, the name or the options say what the machine is, and the title or the other code is the odd one out. That is why most of these corrections are derivations, not research.
Prices that mean nothing, and new machines with hours.
Two themes where a field does not mean what a buyer would read it to mean.
High · Placeholder prices
Machines priced at one cent. Machines with no price anywhere.
Three used machines are priced at one cent in the record: two receivers and a planter. Four more used units carry no price at all. What a buyer sees for these — a one-cent price, a call-for-price or a blank — was not confirmed, because the pages draw the price in the browser; the values here are what the record holds.
Medium · Lead · New machines carrying working hours
17 units flagged new carry more than 1,000 engine hours.
18 units carry more than 500, and 78 new units carry some hours. They are mostly utility and small-frame tractors. Dealers sometimes class demonstrator or rental units as new, so where the line sits is the dealer's call, and the report holds it as one. The 31 new units at exactly one hour are mostly compact tractors, which is plausible delivery time.
Smaller items
Item
What the record shows
Units
A question mark for a serial number
A placeholder where an identity should be
7
A passenger-car description on an ag tire
"Sedans and compact SUVs", on a tire sized for a combine
1
"Available to buy online" left blank
11 of them show an add-to-cart button
62
Brand and make spelled two ways
Mostly spellings; one unit names two different companies
29
Photos filed under a mistyped stock number
The photo is probably the right machine; at risk if photos are matched by file name
5
Seven findings sound like seven problems. Most of them point to four possible upstream mechanisms — and a mechanism, once confirmed, can be fixed once and stopped from recurring. Each tell below is a pattern in the data; the mechanism is EKOM's reading of it.
One hours field for every kind of meter
The tell: the large values on planters and balers read as acre or bale counts, and the dealer's own classification codes say these are implements. A count has no field of its own, so it lands in the one that is there.
An empty date published as a date
The tell: the same date sits on every one of the 315 units, and the warranty flag reads No beside it. It reads as the system's empty value passing straight through to the record.
Titles built apart from names and codes
The tell: on the harrows the classification code and the options both say harrow, and the title says something else; on the cutter and the tiller it is the code that is wrong and the name that is right. Two sources, two answers, no check between them.
Placeholders reaching the live record
The tell: a question-mark serial, a one-cent price and a passenger-car description on an ag tire are each a stand-in that was never replaced.
Correcting 315 warranty dates by hand corrects 315 units. Treating the empty date as empty at the point it is published corrects those and every unit listed after them — and the same check, applied to the hours field, is what would have stopped a bale count from becoming hours.
In McKinsey's 2024 survey of 3,942 business-to-business decision makers, 39 percent were willing to spend more than $500,000 in a single order through self-service digital commerce or remote online connections. The listing is what that buyer decides from.
A dealer sells expertise as much as iron. The buyer on the phone is answered by someone who knows which planter came back from rent, and whether a baler's meter counts hours or bales. Online, the fields carry that alone — across every branch, for the farmer three states away who reads the listing before calling, for the next dealership down the road quoting from it, and for the marketplace or assistant that copies it.
What this class of defect costs is not a missed click. A planter reading over 40,000 hours is a listing a buyer cannot trust. A warranty field that reads 1975, and a flag that says No on units whose warranty has not expired, tell no one anything. A harrow titled as a rock picker is a machine the dealer has, behind a listing that keeps the right buyer from it.
A second clock runs underneath. The buyer comparing tractors or planters increasingly starts with an assistant rather than a search box, and an assistant reads the fields: hours, warranty, price. On this inventory it finds a planter with tens of thousands of hours and a warranty that ended in 1975 — wrong answers, and they do not correct themselves.
Three moves, in order. The first needs no new data at all — and the corrections that look most urgent deliberately wait for the dealer's answer.
This pass read and diagnosed. The same structural understanding powers what follows — turning a diagnosed inventory into one a buyer can choose a machine from with confidence.
1 · Apply the corrections the record already answers for itself
Publish the empty warranty date as blank. Retitle the units whose title disagrees with a name and a code that agree. Blank the placeholder serials. Approved as patterns, not machine by machine.
2 · Settle the questions that gate the rest — starting with hours
Which hour figures are real, and where a count belongs, is a question for the dealer's inventory system. Which units are under warranty is a question for its warranty records. Whether a 1,000-hour demonstrator is new is a policy call. Each is answered once per pattern, not once per machine.
3 · Hold the line at intake
Keep ongoing intelligence where a machine enters the inventory, so the next bale count filed as hours, or the next empty date published as a real one, is caught the day the listing goes live rather than found later by a buyer sorting by hours.
This is how EKOM moves an inventory from insight to impact —
and keeps every machine described as itself as the inventory turns over.