Peninsula Permit Tracker

How long does your city take to approve housing? Ranked by friction score — median permit approval time, pipeline conversion rates, RHNA progress, and ADU statutory compliance — across San Mateo County and Peninsula jurisdictions.

Live: https://peninsulaforeveryone.github.io/peninsula-permit-tracker/

A project of Peninsula for Everyone.

---

What this shows

Metric	What it measures
Friction score (0–100)	Weighted composite: RHNA gap + conversion rate + timeline
RHNA progress	% of 6th-cycle (2023–2031) target reached via issued permits
App → permit conversion rate	% of applications that reach a building permit
Median days to permit	Application-deemed-complete → building permit issued
ADU compliance	% of ADU permits issued within 60-day statutory deadline (AB 68/881)

Cities that don't report required date fields to HCD are flagged — missing data is itself a compliance failure under Government Code 65400.

Data sources

Source	What	Years	URL
HCD Housing Element APR Table A	Applications deemed complete	2018–present	data.ca.gov
HCD Housing Element APR Table A2	Entitlements, permits, COs	2018–present	same
Redwood City Open Data	Building permits with dates	2018–present	redwoodcity.org
Menlo Park ArcGIS Hub	Permit layer	2018–present	data.menlopark.org

Quickstart

git clone https://github.com/PeninsulaForEveryone/peninsula-permit-tracker
cd peninsula-permit-tracker
make install
make run
# open docs/index.html in a browser

Downloads ~50 MB of raw APR data, filters to Peninsula jurisdictions, computes metrics, writes docs/data/viz_data.json.

Pipeline

pipeline/
  config.py       URLs, constants, RHNA targets, friction score weights
  fetch_apr.py    Download & cache HCD APR Table A and A2; normalize columns
  transform.py    Compute per-city metrics, ADU sub-metrics, friction score
  run_all.py      Orchestrator: --force to re-download, --step for single steps

Raw files cached in data/raw/ (gitignored). Final JSON (docs/data/viz_data.json) is committed and served by GitHub Pages.

Friction score methodology

score = RHNA gap (40%) + conversion gap (35%) + timeline score (25%)

RHNA gap        = 1 − (permits_6th_cycle / rhna_target)
Conversion gap  = 1 − (permits / applications)
Timeline score  = median_days_to_permit / 800-day ceiling

When a city doesn't report date fields, timeline weight is redistributed to the other two components. The full methodology string is embedded in viz_data.json under metadata.methodology.

ADU compliance

State law (AB 68/881) requires ministerial approval of ADU applications within 60 days. Any city exceeding this median is in documented violation. The tracker surfaces this as a separate metric with a "⚠ exceeds 60-day limit" flag.

Known data quality issues

1. Self-reported data: APR data is submitted by cities. HCD performs completeness checks but not field-level audits. Cities with incentives to underreport delays may do so.

2. Date field compliance: Many cities leave application/permit date fields blank, especially pre-2023. The tracker displays field-completion rate as its own accountability metric.

3. Join quality: Timeline medians are computed by joining Table A (applications) to Table A2 (permits) on jurisdiction + address + unit type. Match quality varies. Low match rates are surfaced in the drilldown.

4. Double-counting: The same project may appear in multiple APR years as it progresses through the pipeline. Unit count totals should be interpreted as cumulative activity, not net new units.

5. APR URL stability: The CSV resource URLs on data.ca.gov have changed once since 2022. fetch_apr.py includes a CKAN API fallback to auto-discover new URLs.

Deployment (GitHub Pages)

1. In repo Settings → Pages, set source to docs/ folder on main.
2. The refresh_data.yml workflow runs every Monday and commits an updated viz_data.json.
3. The frontend is a single static docs/index.html — no build step.

License

Data: California HCD APR (CC BY), city open data portals (public domain / CC BY). Code: MIT.

See also

• Peninsula enrollment vs. housing tracker — rebuts the "new housing overwhelms schools" argument