NY Tolling Program · TDAP · Findings & recommendations topology

Insights

Last updated May 14, 2026 at 19:00 hrs EST

Seven views over the same evidence base: 33 findings across seven dimensions and 0 recommendations grouped by horizon and change-type. Each panel re-cuts the corpus along a different axis so patterns invisible on a table become legible at a glance.

Dimensions 1-6 per TDAP SOW Schedule 2; Dimension 7 (Agency Relationship) added by Primus and presented to the agencies on Apr 30 — flagged with P where dimensions are listed individually.

◇ Dimension pressure

Two overlapping polygons across the seven dimensions. The filled polygon is total-findings density per dim; the inner stroke is triangulated-high count alone. A wide gap between the two means the dim has many findings but few that meet the strongest evidence bar.

  • Total findings (filled)
  • Triangulated-high only (inner stroke)

⌬ CRTA capability heat-map

Where CRTA's in-house capability is strong, partial, or has a real gap — relative to what each dimension's findings show the program needs. This is the assessment team's honest external read; treat it as the starting point for a build/buy/partner conversation. Click any card to see the named internal assets and the specific gap.

No capability assessment yet.

One Sonnet 4.6 pass over the findings + recommendations. Output is 7 per-dim assessments (~$1, ~30s).

▦ Severity heatmap

Each cell is a count: how many findings in that dimension hit that evidence tier. Cell shade scales to the matrix max (4). Read across a row to see a dim's strength profile; read down a column to see where the program clusters at each tier.

Trig·High Trig·Med Single·High Other Total
D1 QA 4 · · · 4
D2 Resources 4 1 · · 5
D3 Release 4 · · · 4
D4 Environment 4 2 · · 6
D5 Transition 4 · · · 4
D6 O&M Model 3 1 · · 4
D7 Agency 4 2 · · 6
Σ 27 6 0 0 33

✦ Findings constellation

Every finding plotted as a bubble. X = evidence count (how many citations back it); Y = tier rank (triangulated-high at top); color = dimension; bubble area = evidence count. The cluster top-right is the program's "load-bearing" set — strong-tier findings with deep evidence underneath.

Triangulated-high Triangulated-medium Single-high Other 0 2 4 6 8 10 evidence count per finding →
  • D1 QA
  • D2 Resources
  • D3 Release
  • D4 Environment
  • D5 Transition
  • D6 O&M Model
  • D7 Agency

▣ Recommendation matrix

Each recommendation is a dot, colored by its dimension, placed in the 3 × 6 cell of (horizon, change-type) it occupies. Dense cells flag the "shape" of the recommendation set — quick-wins clustered short-term, structural moves clustered long-term, etc.

short-term medium-term long-term Σ
quick-win · · · 0
procedural · · · 0
architectural · · · 0
organizational · · · 0
cultural · · · 0
structural · · · 0
Σ 0 0 0 0

◐ Evidence provenance

Polar wedges, one per dimension. Inner band = citations sourced from Sessions (CRTA SWAT interviews); outer band = citations sourced from the VOC (NY tolling agency leadership). Wedge size is the total. Tells you per dim how much weight each side of the evidence base carries.

288 total citations
  • Inner band — Sessions citations
  • Outer band — VOC citations

↦ Findings → Recommendations

Flow widths show how many recommendations attribute to each dimension and which horizon they target. A thick band into "long-term" from a single dim signals a structural-debt cluster; broad fan-out into "short-term" signals quick-wins across many dims.

CC 0 D1 0 D2 0 D3 0 D4 0 D5 0 D6 0 D7 0 Short-term 0 recs Medium-term 0 recs Long-term 0 recs

⋹ Recommendation beeswarm

Each recommendation as a single dot, colored by change-type, positioned within its horizon band. Within a band, dots are jittered vertically (deterministic) to make density visible — a tall pile means many recs of the same horizon, a sparse band means fewer.

short-term 0 recs medium-term 0 recs long-term 0 recs
  • quick-win
  • procedural
  • architectural
  • organizational
  • cultural
  • structural