Analyst PluginHeavySkillv0.12.3

Cohort Benchmarking

Benchmark a dealer group against a custom cohort — same geography, same brand mix, same size tier for apples-to-apples comparison

cohort benchmarkingcompare against similar groupssame-size dealer group comparisonbenchmark against comparable retailerspeer cohort analysisregion-matched dealer comparison
Heavy
Complexity
3–5
API Calls
$0.005 – $0.012
MC API Cost
$0.12 – $0.28
Total Cost

Purpose

Ranking AN against LAD is misleading when their scale, geography, and brand mix differ significantly. A more rigorous analysis compares a dealer group to its true peer cohort — similar scale, overlapping markets, similar OEM exposure. Cohort Benchmarking builds a custom peer set and runs the full comparison.

Cohort Benchmarking lets analysts define the comparison cohort by geography, brand exposure, or size tier — then runs the same multi-dimensional analysis as Group Benchmarking against only those comparable peers. Output includes a cohort-normalized quintile score and gap-to-best-in-cohort metrics.

Audience:Equity analysts running relative value screensM&A analysts doing target comparisonsActivist investors building peer frameworks

How It Works

Execution flow. MCP tool calls are shown inline on each step.

01
Cohort Definitionget_sold_summary

Loads profile or prompts user for cohort criteria: geography (states), brand/OEM exposure, or size tier (volume range). Identifies matching dealer groups in MarketCheck data.

02
Cohort Data Pullget_sold_summary

Calls get_sold_summary filtered to cohort geography and brand parameters for current and prior periods. Extracts all cohort groups' volume, ASP, DOM.

03
Cohort-Normalized Ranking

Ranks all groups within the defined cohort across Volume, ASP, DOM, and Efficiency Score. Assigns cohort quintile scores.

04
Gap Analysis

For the focus group, identifies gap-to-best and gap-to-median on each metric. Quantifies opportunity (e.g., 'closing DOM gap to cohort leader would add X efficiency points').

MCP Tool Calls

ToolCallsPurpose
get_sold_summary3–5Cohort group volume/ASP/DOM current and prior, plus optional brand filter calls

Example Output

COHORT BENCHMARKING — AutoNation (AN) | Southeast Region Cohort
══════════════════════════════════════════════════════════════════

COHORT DEFINITION: Southeast states (FL, GA, TX, NC, SC) | Multi-brand | >5,000 units/mo

COHORT MEMBERS: AutoNation, Penske, Group 1, Sonic, Asbury

  Group       Ticker  Vol (SE)  ASP       DOM      Eff Score  Cohort Rank
  ──────────  ──────  ────────  ────────  ───────  ─────────  ───────────
  Group 1     GPI     6,841     $40,200   24 days  285        1st
  Penske      PAG     5,214     $42,100   26 days  201        2nd
  AutoNation  AN      8,432     $41,800   31 days  272        3rd  ← Focus
  Asbury      ABG     3,107     $39,400   28 days  111        4th
  Sonic       SAH     2,441     $38,100   32 days  76         5th

AN vs Cohort Leader (GPI): DOM gap of +7 days — closing to cohort median (+2 days)
would add ~85 Efficiency Score points and align closer to Q1 position.

Cost Estimate

TierHeavy
API Calls3–5
MC API — best$0.005
MC API — worst$0.012
Claude — best$0.10
Claude — worst$0.25
Total range$0.12$0.28

10 analyses/month ≈ $1–3

Limitations

  • US market only.
  • Custom cohort quality depends on specificity of user criteria — overly narrow cohorts may have too few groups for meaningful quintile scoring.
  • Geographic cohort filtering relies on state-level data; multi-state groups are not automatically filtered to only their overlapping markets.
Analyst Plugin
View all skills & commands