mobility.rest Request a scan

Transit feed reliability scans

Know when your transit data stops working before riders do.

Managed 24-hour observation for GTFS Schedule and GTFS Realtime. Detect availability failures, stale information, coverage gaps, and silent regressions with evidence your agency and vendors can act on.

Request a 24-hour scan View a sample report
GTFS Schedule GTFS Realtime Public endpoints

Valid does not always mean usable.

A feed can return HTTP 200, parse cleanly, and pass validation while still serving stale, incomplete, or mismatched information to rider-facing systems.

The first product is a managed scan, backed by internal tooling, that separates endpoint failures, parser failures, scanner gaps, and rider-impact evidence.

Delivery flow

  1. Verify public endpointsBefore scan
  2. Observe for 24 hoursOutside-in
  3. Separate signal from noiseHuman review
  4. Deliver findings and next actionsPrivate report

Schedule comparisons are only called regressions when a defensible prior baseline exists.

Built Request intake

Form submissions create pending scope-review records.

Built Schedule audit

GTFS Schedule archives are parsed and stored by checksum.

Built Realtime observer

Protobuf payloads are fetched, parsed, and archived by hash.

Next Freshness metrics

The next metric layer turns observations into incident candidates.

A narrow workflow senior reviewers can audit.

The public offer is intentionally service-led. Each request moves through a controlled path before any observation starts, and every report separates what was observed from what is inferred.

Review

Scope review before observation

Confirm buyer context, public feed URLs, report audience, and whether the requested endpoints are reachable without credentials.

  • No private systems required
  • No scan starts from form submit alone
Archive

Archive the Schedule baseline

Store the Schedule archive by checksum, parse feed shape, and only call a regression when a defensible prior snapshot exists.

  • Service horizon and required files
  • Validator status and limitations
Observe

Observe Realtime for 24 hours

Poll public protobuf endpoints on cadence, preserve response metadata, and keep fetch, parser, and scanner gaps separate.

  • Trip Updates, Vehicle Positions, Service Alerts
  • Raw payload hashes and timestamps
Deliver

Human review before delivery

Turn raw evidence into prioritized findings with observed facts, potential rider impact, likely owner, and recommended next action.

  • Private report by default
  • No compliance certification claim

Active V1 scope is narrow by design.

The standard scan uses publicly accessible endpoints and requires no private integration, credentials, or rider data.

GTFS Schedule

Feed shape and baseline discipline

  • Public URL availability and zip structure
  • Required files, service horizon, and feed expiration
  • Route, stop, trip, calendar, and accessibility fields
  • Conditional comparison to a known prior snapshot
  • Change classification: defect, potential regression, expected change, advisory, or not comparable
Trip Updates

Freshness and matching

  • Endpoint uptime and response behavior
  • Protobuf parse validity
  • Timestamp presence rate before entity-age statistics
  • P50 and P95 age where entity timestamps exist
  • Static-to-Realtime matching with stated denominator and exclusions
Vehicle Positions

Route-level rider risk

  • Freshness distribution by route and time of day
  • Coverage gaps against scheduled service
  • Duplicated, canceled, added, and unscheduled trip treatment
  • Periods where the feed is globally fresh but specific routes are stale
  • Likely owner language kept cautious and evidence-based
Service Alerts

Alert lifecycle review

  • FeedHeader age and refresh behavior
  • Active-period plausibility
  • Expired or unchanged alert lifecycle
  • Required and expected field coverage
  • No invented per-alert P50 or P95 ages

Every problem is packaged as evidence, impact, owner, and action.

This is the wedge beyond validation: the report translates technical feed symptoms into a decision record that agency, vendor, consultant, and program teams can use together.

Finding anatomy

From one stale route window to a defensible next step.

Observed fact Route 14 Vehicle Positions exceeded the 90-second target in 63% of samples from 08:42 to 09:31.

Endpoint availability stayed high, so the finding is not reduced to a generic outage.

Potential rider impact Arrival displays may have shown stale vehicle locations during a high-demand service period.

The report avoids claiming customer harm that was not directly observed.

Likely owner Evidence suggests realtime vehicle export or CAD/AVL handoff.

Outside-in observation cannot prove a private-system root cause.

Next action Compare vehicle export logs for the incident window against the public Vehicle Positions feed.

The scan ends with an operational handoff, not a vague severity label.

Route coverage

Coverage is reported with a denominator.

Route 14 42% usable
Route 21 68% usable
Route 7 91% usable
Route 3 96% usable

Fictional denominator: scheduled trips with expected realtime coverage during the observed service span. Exclusions are stated in the report.

Incident timeline

Correlation stays separate from causation.

  1. 03:10Schedule archive published and stored by checksum.
  2. 08:42Vehicle position age begins exceeding the target on Route 14.
  3. 09:31Freshness recovers, but coverage denominator remains disclosed.
  4. ReviewReport states what was observed and what evidence suggests.

The paid artifact is a private evidence report.

Every named agency, URL, incident, and metric in this sample is fictional demonstration data. Real customer reports remain private unless approval is given.

Open the full sample report
24-Hour Rider Information Reliability Scan River County Transit, fictional demo
Action needed
99.71%Endpoint availability
242sVehicle P95 age
86%Route coverage denominator disclosed
Report structure
  1. Scope and methodology
  2. Executive summary
  3. Endpoint reliability
  4. GTFS Schedule feed
  5. GTFS Realtime freshness
  6. Realtime coverage
  7. Static-to-Realtime matching
  8. Incident timeline
  9. Remediation plan
  10. Technical appendix
High confidence

Vehicle position freshness degraded after 08:42.

Observed fact
Route 14 vehicle entities exceeded the 90-second freshness target for 63% of observations between 08:42 and 09:31.
Potential rider impact
Arrival displays may have shown stale vehicle positions during the morning peak.
Likely owner
Evidence suggests realtime vehicle export or CAD/AVL handoff. Outside-in observation cannot prove a private-system root cause.
Recommended next action
Compare vehicle export logs for route 14 between 08:30 and 09:45 against the public Vehicle Positions feed.
Endpoint HTTP Feed age Parse
/gtfs-rt/vehicle-positions 200 6s header, 242s P95 entities valid protobuf
/gtfs-rt/trip-updates 200 14s median valid protobuf
/gtfs/schedule.zip 200 43d horizon parsed archive
Baseline: prior customer-supplied Schedule snapshot from 2026-05-18. Sample URL: https://feeds.rivercounty.example/gtfs.zip
Executive

Decision summary

Overall status, top findings, likely business risk, and the smallest next action that moves the issue forward.

Operations and planning

Service context

Route, time-of-day, service-horizon, and rider-information implications with baseline discipline where comparison is possible.

Technical and vendor

Reproducible appendix

Endpoint attempts, parser status, payload hashes, denominator rules, raw evidence rows, and limitations for follow-up work.

Start with a scan. Monitor only when the workflow proves useful.

The one-time 24-hour scan is the dominant launch offer. Ongoing monitoring comes after an initial audit establishes fit and operational value.

One time

Schedule Audit

$750

Current GTFS Schedule feed review with endpoint checks, feed shape, validator evidence, and baseline comparison when available.

  • One public Schedule feed
  • Service horizon and required-file review
  • Prioritized remediation plan
Primary launch offer

24-Hour Rider Information Reliability Scan

$1,500

One Schedule feed plus up to three GTFS Realtime endpoints observed for 24 hours after scope verification.

  • Trip Updates, Vehicle Positions, and Service Alerts
  • Freshness, coverage, matching, and incident timeline
  • Executive summary, remediation plan, and technical appendix
After initial audit

Ongoing Monitoring

from $399/mo

Recurring feed reliability monitoring for teams that need continued outside-in observation after an initial scan.

  • Monitoring cadence scoped after audit
  • Private recurring reports
  • Incident follow-up support
Scoped by feed count

Portfolio Monitoring

from $1,250/mo

Monitoring for vendors, consultants, and regional programs managing multiple agency feeds.

  • Multiple agencies and endpoints
  • Portfolio summary views
  • White-label reporting available

Clear limits make the report more credible.

The scan is an outside-in review of public feeds. It is built to identify evidence and likely ownership, not to certify compliance or make proprietary trip planners ingest data.

Data handling

  • Public feed URLs only for the standard scan.
  • No rider PII required or intentionally collected.
  • Customer reports remain private.
  • Named negative findings are not published without approval.
  • Raw snapshots are retained for 30 days unless another period is agreed in writing.

Limitations

  • The service does not certify GTFS or GTFS Realtime compliance.
  • The service does not guarantee ingestion by Google, Apple, Transit, or any proprietary trip planner.
  • Outside-in observation cannot prove a private-system root cause.
  • A 24-hour window may not represent weekend, seasonal, school, event, or low-frequency service.
  • Schedule regression claims require a defensible prior baseline.

Print-ready service sheet

24-Hour Rider Information Reliability Scan

Outside-in observation for one GTFS Schedule feed and up to three GTFS Realtime endpoints.

$1,500 one time

Included scope

  • One current GTFS Schedule feed
  • Up to three GTFS Realtime endpoints
  • Endpoint verification before observation
  • 24-hour outside-in observation
  • Public-endpoint and privacy limitations

Checks performed

  • Endpoint reliability
  • Timestamp and freshness analysis
  • Route-level coverage
  • Static-to-Realtime matching
  • Conditional Schedule comparison when a baseline exists

Deliverables

  • Top three prioritized findings
  • Executive summary
  • Remediation plan
  • Technical appendix
  • Private report for internal review

Smaller alternative

Schedule-only audit - $750

Use when the immediate need is GTFS Schedule feed shape, expiration, validator evidence, and baseline comparison.

mobility.rest Request a scan

Request a 24-hour scan.

Share public feed URLs and buyer context. Requests are reviewed for endpoint accessibility and scope before the observation window starts.

Review Scope review

We confirm the buyer context, public URLs, report audience, and whether the request fits the fixed scan.

Verify Endpoint verification

We check reachability before observation. A submitted form does not start a scan automatically.

Report Private report

If scope is accepted, the observation window starts and the final report remains private by default.

Before you submit

  • Use public HTTPS feed URLs only.
  • Do not include passwords, private tokens, or rider data.
  • Use the notes field for the operational question the report should answer.

Scope request

Send the minimum facts for review.

Required fields are marked in the labels. Schedule URL is required; Realtime URLs are optional for the initial request.

01 Buyer context 02 Public feed URLs 03 Review question
Requester and report context

Agency, vendor, consultant, MPO, state program, or operating entity.

Used only to coordinate scope review and report delivery.

Public feed URLs

Include direct public URLs. Use `.example` values only for demos; real requests should use reachable HTTPS endpoints.

Usually a direct `.zip` download for the current GTFS Schedule feed.

Report question

Review the offer before requesting scope.

These public resources describe what the scan checks, what the report includes, and where the service deliberately stops.

Sample artifact

Fictional 24-hour report

Review the sample report structure, evidence rows, limitations, and remediation language.

Open sample report
Field guide

GTFS Realtime freshness checks

Read the source-backed guide on detecting stale rider-facing data after validation passes.

Read guide
Monitoring

GTFS Realtime monitoring

See what to watch after a realtime feed is reachable, parseable, and still capable of going stale.

Read monitoring page
Audit

GTFS feed audit

Review the fixed-scope audit path for Schedule and Realtime feeds before a deadline or handoff.

Read audit page
Comparison

Validator vs observation

Understand what validators answer and when continuous observation becomes the missing layer.

Compare workflows
Next step

Scope review request

Send public feed URLs and the operational question the report should answer.

Request a scan