dealer-performance-intelligence

name	dealer-performance-intelligence
description	Industry cohort benchmarking and performance intelligence. Triggers: "performance report", "benchmark my group", "how do I compare to industry", "competitive analysis", "performance intelligence", "where am I winning", "where can I improve", "industry benchmarking", "dealer performance report", "how do we stack up", "competitive strengths", "improvement opportunities", benchmarking against 400+ dealer group industry cohort and named public peers.
version	0.1.0

Date anchor: Today's date comes from the # currentDate system context. Compute ALL relative dates from it. Never use training-data dates.

get_sold_summary parameter safety:

• Always set inventory_type explicitly (New or Used) — omitting it defaults to New, returning zero results for used-vehicle queries
• Always set limit: 5000 — the default (1000) silently truncates when (months × states × ranking combos) exceeds 1000 rows
• For volume totals, use ranking_dimensions: dealership_group_name (or the single relevant dimension) — never use the default make,model,body_type which creates ~150K rows for national 3-month queries
• Use separate calls for totals vs breakdowns — don't combine in one call

Dealer Performance Intelligence Report

Benchmark your dealer group's operational performance against the full ~400 US dealer group industry cohort. Identifies competitive strengths (where you outperform) and improvement opportunities (where targeted focus could drive gains), with named comparisons to the 8 publicly traded dealer groups.

Architecture: This skill uses a multi-agent Wave pattern. Wave 1 runs the cohort benchmarking agent and lot scanner in parallel. Wave 2 assembles the report.

Dealer Group Profile (Load First)

Load the marketcheck-profile.md project memory file. If missing, prompt /onboarding and stop.

Extract: dealer_group.group_name, dealer_group.locations[] (all location details), preferences.default_inventory_type (default: "used"), dealer_group.franchise_brands, dealer_type.

Target group identification: The group name from the profile is used to search for this dealer in the MarketCheck cohort data. If the profile contains a dealer_group_name_mc field (the exact name as it appears in MarketCheck), use that. Otherwise try group_name.

Confirm: "Generating Performance Intelligence Report for [group name] | Locations: [N] | Brands: [list]"

Named Public Peer Groups

These 8 publicly traded groups are always included as named comparison points:

SAH  → Sonic Automotive        | KMX  → CarMax
CVNA → Carvana                 | GPI  → Group 1 Automotive
PAG  → Penske Automotive Group | LAD  → Lithia Motors
AN   → AutoNation              | ABG  → Asbury Automotive Group

Workflow

Step 1 — Compute Date Ranges

From # currentDate, compute:

• Current month: most recent fully completed month
• Prior year same month: same month one year ago
• Q1 of trailing year: Jan 1 – Mar 31, ~12 months ago
• Q4 of trailing year: Oct 1 – Dec 31 of most recent completed Q4

Wave 1 — Launch Simultaneously

Agent A: cohort-benchmarking-agent

Use the Agent tool to spawn the marketcheck-cowork-plugin:cohort-benchmarking-agent with:

Benchmark these target groups against the full industry cohort: Target groups: [dealer group name from profile], AutoNation, Lithia Motors, Penske Automotive Group, Sonic Automotive, Group 1 Automotive, Asbury Automotive Group, CarMax, Carvana

Date ranges:

• current_month_from: [date] | current_month_to: [date]
• prior_year_month_from: [date] | prior_year_month_to: [date]
• q1_from: [date] | q1_to: [date]
• q4_from: [date] | q4_to: [date]

Important: All get_sold_summary calls MUST set inventory_type explicitly (Used or New per profile default) and limit: 5000.

Return: quintile thresholds, per-group KPI values, quintile assignments, and composite scores.

Agent B: lot-scanner (facets + stats mode)

For each location in the profile, use the Agent tool to spawn dealership-group:lot-scanner with:

Pull lot composition for dealer_id=[location's dealer_id], country=[location's country], mode=facets_only. Use rows=0 with facets=make|0|10|1,body_type|0|10|1 and stats=price,dom,miles. Also pull car_type=new and car_type=used separately to get new/used split. Return: total units, new count, used count, avg price, avg DOM (used vs new), avg mileage (used), top makes, body type mix. Location label: [location name].

Wave 2 — Assemble Report (After Wave 1 Completes)

Using the cohort benchmarking results + lot scanner results:

Section 1: Performance Benchmarks at a Glance

Create a summary table of key KPIs with industry context:

KPI                    | [Group Name] | Industry P20 | Median | Industry P80 | Position
───────────────────────|──────────────|──────────────|────────|──────────────|──────────
Used Vehicle DOM       | [val] days   | [val]        | [val]  | [val]        | [quintile or %ile]
YoY Unit Volume Growth | [val]%       | [val]%       | [val]% | [val]%       | [position]
DOM Trend (Q1→Q4)      | [val] days   | [val]        | [val]  | [val]        | [position]
New Vehicle DOM        | [val] days   | --           | ~[val] | --           | [context]
New Price vs. MSRP     | [val]%       | [val]%       | [val]% | [val]%       | [position]

Highlight the 2-3 strongest KPIs as "Where [Group] is winning" and the 1-2 weakest as "Where there is opportunity to improve."

Section 2: Competitive Strengths

For each KPI where the target group outperforms the industry median (P50) or outperforms named public peers:

Write a focused paragraph covering:

1. The metric and its value — with specific comparison to industry median and named peers
2. Why it matters — operational and financial significance (floor plan costs, depreciation risk, volume trajectory)
3. Named peer comparison — compare directly to Carvana, CarMax, Sonic, etc. where relevant

Example structure: "Used vehicle velocity — faster than Carvana. [Group]'s used vehicles sell in an average of [X] days — faster than Carvana ([Y] days) and CarMax ([Z] days)..."

Include a horizontal bar comparison showing the target group vs. named peers for each strength metric.

Section 3: Improvement Opportunities

For each KPI where the target group underperforms the industry median:

Write a focused paragraph covering:

1. The metric, its value, and the gap — how far below median or P80
2. What the data reveals — specific insight from MarketCheck data (e.g., which model lines are aging, which trims are slow)
3. Recommended action — specific, data-driven recommendation the dealer can execute using MarketCheck weekly data

Use a callout box format for the recommended action.

Section 4: Inventory Mix Analysis

From the lot-scanner results:

• New/used inventory ratio vs. new/used sales ratio — identify misalignment
• Used vehicle average mileage vs. industry median (52,289 mi from methodology) and named peers
• Body type distribution — is stocking aligned with local demand?
• New vehicle model mix — which nameplates contribute most to new vehicle DOM?

Section 5: How MarketCheck Data Can Help

Brief section (2-3 bullet points) on how ongoing weekly data access enables:

• Weekly competitive position monitoring (DOM, pricing, velocity vs. peers)
• Model-level new vehicle aging identification (target the 20% of inventory driving the average up)
• Used vehicle sourcing quality benchmarking (mileage, price point vs. local competitors)

Output Format

Present the full report with clear section headers. Use tables for benchmarks, narrative paragraphs for strengths/opportunities, and callout boxes for recommended actions.

The report should read as a standalone document that a dealer group executive or owner can review without needing additional context.

Important Notes

• US-only: Cohort benchmarking uses US get_sold_summary data. UK locations receive inventory composition only.
• Inventory type: Respect preferences.default_inventory_type from profile. If set to "used," used vehicle KPIs are primary. If "new," emphasize new vehicle KPIs.
• The target group may not appear in the cohort data if MarketCheck uses a different group name. If not found, note this and use the lot-scanner data + active inventory stats to compute what's possible.
• Strengths first, then opportunities. Always lead with what the dealer does well — this is a benchmarking report, not an audit.
• Named peer comparisons should use the actual KPI values from the cohort agent, not placeholder data.
• For franchise dealers, the new vehicle price/MSRP metric is particularly relevant. For independent/used-only dealers, skip it and note the exclusion.