Resolution Layer Case Study · Read & Diagnose
A Regional Multi-Franchise Dealer Group
A group-wide inventory analysis — seven rooftops, one storefront, zero manual setup
Before the findings
What's actually riding on one group's worth of vehicle data.
About this case study. This is an anonymized version of a client deliverable. The dealer group, its rooftops, its locations and its history have been removed; every finding, severity, count, field name and verbatim data value is reproduced exactly as it appeared in the original analysis. Rooftops are labeled by number and franchise brand only.
This group has stood for a certain kind of experience in its market for more than four decades, starting from a single luxury storefront in the early 1980s. That storefront has since grown into seven rooftops and thirteen franchises — Acura, Audi, Volkswagen, Infiniti, Porsche, Jaguar, Maserati, Alfa Romeo, Fiat, and the full Chrysler Dodge Jeep Ram lineup — all presented to shoppers under one banner on a single group domain. A world-class drive is a promise made one customer at a time, in the showroom and the service bay. What this analysis examines is whether the catalog infrastructure behind that promise is keeping pace with it online.
Every vehicle listing across all seven rooftops feeds the same handful of channels that decide whether a shopper ever sees it: the group's own site, the big three marketplaces — Autotrader, Cars.com, CarGurus — and, increasingly, a channel most dealer groups haven't built for at all. A group storefront makes that stakes higher, not lower: one inconsistent field convention repeated across seven stores is one problem multiplied seven times.
"Even with affordability top of mind, buyers know exactly what they need: efficiency, transparency and tools that actually help them easily navigate their vehicle purchase."
Lori Wittman, President, Retail Solutions, Cox Automotive — 2026 Car Buyer Journey Study press release, Jan. 13, 2026
We ran the group's live inventory feed — a single Dealer.com storefront that tags every vehicle to the rooftop it came from — through EKOM's analysis pipeline exactly as we would a new customer's catalog. One feed, one unattended pass: no hand-built schema, no manual setup, no one telling the system what to look for or which store was which. What came back wasn't noise. It was a concentrated set of pricing, completeness, and consistency errors — the majority of them clustered in identifiable segments and in two of the seven rooftops — every one pinned to the exact store, and the exact vehicles, it affects.
At a glance
A machine-surfaced triage signal, not a verified defect list — the vast majority are real; a small share may reflect intentional dealer choices. Every item below is worth a human look.
Findings by severity
| Severity | Count | What it means |
|---|
| CRIT | 50 | Wrong or missing data that actively breaks search, pricing, or trust |
| HIGH | 69 | Inconsistent or fragmented data that degrades filtering and syndication |
| MED | 20 | Smaller quality issues worth cleaning up |
| LOW | 4 | Minor polish |
| Total | 143 | 119 of them high-severity (Critical + High) |
Where to start — highest-impact fixes
Structural · all rooftops
No dedicated mileage field anywhere in the feed
Odometer readings for used vehicles appear only sporadically in free-text descriptions — used inventory is invisible to the most common shopper filter and can't feed valuation tools.
Structural · all rooftops
SUVs split across two filter buckets
The body-style field uses both "Sport Utility" and "SUV" for the same category, fragmenting SUV inventory into two distinct facets a shopper has to find separately.
Store 3 · Acura · 11 vehicles
New vehicles carry a $0 price sentinel
Unpublished prices are stored as msrp = "0" instead of a proper null — price filters and savings math treat them as zero-dollar cars.
Store 2 · Volkswagen · 15 vehicles
Unfilled template tokens published live
Placeholders like {YMM}, {SERIES}, and {EXTERIORCOLOR} appear verbatim in customer-facing descriptions — listings look technically broken.
Impact by rooftop
Grouped by the store tag the feed carries on every vehicle — how many findings touch each rooftop, and how many of its vehicles carry at least one issue. Catalog-wide structural issues (below) affect every rooftop and are counted separately.
| Rooftop | Findings | Vehicles Affected | Critical |
|---|
| Store 1 — Chrysler Dodge Jeep Ram | 45 | 76 | 22 |
| Store 2 — Volkswagen | 34 | 83 | 13 |
| Store 3 — Acura | 16 | 33 | 3 |
| Store 4 — Audi | 6 | 26 | 2 |
| Store 5 — Infiniti | 12 | 28 | 5 |
| Store 6 — Porsche | 7 | 12 | 3 |
| Store 7 — Fiat / Maserati / Alfa Romeo / Jaguar | 9 | 24 | 3 |
| All rooftops | 143* | 282 | 50* |
Reading this table. Two rooftops — Store 1 (Chrysler Dodge Jeep Ram) and Store 2 (Volkswagen) — account for the majority of findings and criticals. Stores 3 and 4 are lighter but share the same structural gaps. * Group totals for findings and criticals reflect the catalog as a whole; per-rooftop counts include shared catalog-wide issues, so they overlap rather than sum cleanly. The vehicles-affected column does sum to the group total of 282.
Where the problems concentrate
By condition, the findings cluster predictably: new-vehicle listings carry most of the pricing and template-token errors, used listings are where the missing-mileage gap hits hardest, and several used vehicles should likely be reclassified as certified.
Fix once, every rooftop benefits
Twenty-three of the findings are schema or formatting problems in the feed itself. Because all seven rooftops publish through the same storefront, each of these is fixed once and clears across the whole group.
| Finding | Sev. | Field |
|---|
| Search & Filter Integrity |
| No dedicated mileage field; odometer only in free-text descriptions | CRIT | mileage |
| Body style split between "Sport Utility" and "SUV" for the same category | CRIT | body_style |
| Drivetrain uses inconsistent values & capitalization (4x4 / AWD / All-Wheel Drive) | HIGH | drivetrain |
| Fuel type uses both "Gas" and "Gasoline," splitting the gasoline bucket | HIGH | fuel_type |
| Interior color abbreviated / misspelled ("Blk", "Titan Blk"), fragmenting color filters | HIGH | interior_color |
| Pricing Representation |
| MSRP stored in three formats — raw integers, dollar strings, and sentinel zeros | HIGH | msrp |
| Dealer price uses a currency string while MSRP on the same record uses a bare integer | HIGH | dealer_price |
| Display & Syndication |
| Image URL lists duplicated end-to-end, doubling payload and breaking the carousel | HIGH | image_urls |
| A branding/logo image appended to every vehicle, risking it as the featured photo | HIGH | image_urls |
| Character-encoding garble ("harman/kardon®" → "harman/kardon®") in descriptions | HIGH | dealer_description |
| Structural — compound fields |
| Engine field combines several kinds of information in one column | HIGH | engine |
| Transmission field combines several kinds of information in one column | HIGH | transmission |
Note. This is a representative cut of the 23 catalog-wide structural findings, ordered by shopper impact. The full machine-readable list, with the remaining formatting and completeness items, accompanies this report.
Standout critical findings
The most severe issues, concentrated where the report expects them — the newer, higher-volume rooftops and the new-vehicle segment. Some pricing conventions may be intentional under the group's fixed-price program — worth confirming before treating as defects.
| Finding | Sev. | Rooftop | Scope |
|---|
| Pricing & Trust |
| New vehicles carry msrp = "0" as an unpublished-price sentinel | CRIT | Store 3 · Acura | 11 vehicles |
| New vehicles with both MSRP and dealer price empty — entirely unpriceable | CRIT | Store 3 · Acura | 10 vehicles |
| All New-condition items have an empty dealer-price string, no consistent sentinel | CRIT | Store 2 · VW | 10 vehicles |
| Dealer price exceeds MSRP by exactly $699, contradicting stated below-market pricing | CRIT | Store 1 · CDJR | 4 vehicles |
| Broken Copy on Live Pages |
| Unfilled template tokens ({YMM}, {SERIES}, {EXTERIORCOLOR}) published verbatim | CRIT | Store 2 · VW | 15 vehicles |
| Used/certified descriptions render literal CRM tokens ({excited}, {features}) | CRIT | Stores 5 & 4 | 12 · 9 vehicles |
| Classification & Completeness |
| Used vehicles missing a mileage value entirely, invisible to odometer filters | CRIT | Store 1 · CDJR | 8 vehicles |
| Grand Wagoneer L long-wheelbase encoded inconsistently across model/trim/absent | CRIT | Store 1 · CDJR | 6 vehicles |
| "Used" vehicles on /certified/ URLs referencing CPO terms — should be Certified | CRIT | Store 5 · Infiniti | 6 vehicles |
| eTorque mild-hybrids classified as fuel type "Other" instead of "Gasoline" | CRIT | Store 1 · CDJR | 5 vehicles |
| Raw paint code ("41") stored as exterior color instead of the color name | CRIT | Store 1 · CDJR | 2 vehicles |
When the listing is wrong
A distinct class of finding, and the highest-stakes one — listings that don't just filter poorly, they state something that isn't true. This is trust and advertised-price exposure, not search hygiene.
Advertised-price discrepancy
One Jeep Compass Latitude advertises a price of $30,484 in its description while its structured price field reads $31,231 — a live contradiction between what the shopper reads and what the system quotes. Under advertised-price rules, that gap is the kind of thing a dealership can be held to.
Store 1 · CDJR · 2 vehicles
Described as a V6 — actually an inline-4
Two 2026 Grand Cherokee descriptions name a V6 the vehicles don't have. A factually wrong spec a buyer can act on long before anyone in the showroom can correct it.
Store 2 · Volkswagen · 1 vehicle
Advertised to seat seven — it seats five
A 2026 Atlas Cross Sport listing claims seating for up to seven. It's a 5-passenger, two-row vehicle — a promise the car can't keep, and a familiar way a deal falls apart at delivery.
Store 2 · Volkswagen · 1 vehicle
A moonroof that's there and not there
A 2025 Golf R's features claim both a panoramic moonroof and that the moonroof was deleted for a solid roof — contradictory attributes on the same car that no shopper can resolve.
Store 1 · CDJR · 3 vehicles
An out-of-region incentive on a local lot
Several Compass listings surface a region-specific manufacturer bonus-cash program from a different sales region entirely — an eligibility and advertised-price risk the moment a shopper tries to claim it.
Cross-rooftop integrity
The pass also caught vehicles sitting under the wrong rooftop — a used Audi S5 and a Toyota 4Runner both assigned to Store 3 (Acura), with the Audi carrying an impossible "3L i-3" engine designation. On a seven-rooftop storefront, a car in the wrong showroom is a finding only a group-wide pass can see.
Why these rank highest. A filter gap costs a click. A listing that states a wrong price, a wrong engine, or a wrong seat count costs trust — and, where price or eligibility is involved, can create real advertised-price exposure. These are worth a same-day human look, ahead of anything else in this report.
What this means
already
44%
of car shoppers use an AI-powered search tool when researching a vehicle, and their trust in the information those tools return is higher than most dealers assume. A price field a shopper can't parse, a used car with no mileage, or a fuel type collapsed into "Other" is invisible to that shopper twice over — once to the AI tool, and once to the person reading its answer.
Cars.com, AI in Car Shopping Consumer Survey, Nov. 2025
This group's reputation was built the same way for more than four decades — by a team that knows its inventory, knows its customers, and gets the details right in the showroom. These findings are a sign that the same attention to detail needs to carry through to the digital side of the business, where more and more of the buying journey now begins. And they're concentrated: almost entirely in two of the seven rooftops and in the new-vehicle segment, where the majority of findings and the most severe pricing and completeness errors live — a focused extension of the craft the group already has, not a group-wide rebuild.
The exposure is specific. Eleven vehicles at the Acura rooftop priced at a literal $0, and ten more entirely unpriceable, are simply invisible to every shopper who filters or sorts by price — on a luxury rooftop where price is rarely the barrier, but a broken listing always is. Fifteen Volkswagen listings at Store 2 publishing raw {YMM} and {SERIES} tokens read as broken to any shopper who lands on them. And with no mileage field anywhere in the feed, the entire used inventory across all seven stores is hard to surface through the single most common filter a used-car shopper uses.
This is what one unattended pass surfaces with no access, no schema, and no one pointing at the problem.
The same pipeline runs the same way on the next rooftop, and the next platform.