Housing policy should rest on data.

STLCA · Data today ~ change tomorrow

STLCA examines Seattle’s publicly available landlord/tenant housing complaint, inspection, violation and enforcement records to ask a practical question: Can the public follow what happened after a complaint was filed?

We analyzed 233,543 records of Seattle Department of Construction and Inspections (SDCI) complaint and Notice of Violation data (2003 to 2025) and identified 56,258 landlord/tenant records.

These are the records available to the public to see how SDCI handles complaints, issues Notices of Violation and enforces tenant protections. Inspection activity cannot be measured from a field that does not publish dates, or from records marked Completed or Closed without a result.

The 56,258 records contain five types
40,540Complaints
9,961Notices of Violation
4,045Tenant relocation
1,394Citations
318Unfit buildings

Closed/Completed does not mean Corrected/Protected.

The investigation, from the beginning

We started with Seattle’s own complaint records.

We cleaned and organized the data, noticed patterns that did not make sense, and followed the questions from one dataset to the next. Here is what we learned: the public data can count enforcement activity, but it cannot consistently show what happened, why cases closed, or whether problems were corrected. Housing policy needs that complete record.

01Complaints

We began with the complaints.

Seattle publishes code enforcement records for construction, land use, housing, vacant buildings, noise and other subjects. We organized 233,543 records from 2003 to 2025 to see what people reported and how the City responded.

The first pattern was impossible to miss. From 2017 to 2025, complaints rose from 7,768 to 15,973. Notices of Violation went from 1,673 to 969. The lines moved in opposite directions.

When an inspector confirms a violation of the housing code, SDCI (Seattle Department of Construction and Inspections) issues a Notice of Violation: a written order with a deadline to fix the problem.

The second chart shows eight groups of record type labels from all 233,543 records. Land use (10,732) and weeds (7,270) account for 88.5% of all 20,340 citations. The Landlord/Tenant group (20,325 records) includes 500 Notices of Violation and 69 citations.

Citywide complaints and Notices of Violation, by year
ComplaintsNotices of Violation
04k8k12k16k2003200720112015201920232025RECORDS OPENED PER YEARYEAR RECORD OPENED20174621294,0661,8574,2991,8834,7172,4115,1412,0765,3201,5674,8901,4034,6481,2194,4391,1614,6581,2614,9461,2795,6511,4626,8731,6027,2061,4177,7681,67310,4571,30611,7961,42310,63797010,57096913,9531,00914,6731,11814,2881,24515,9739692017 → 2025Complaints +106%Notices −42%2018+35% in one yearNov 2003: data begins
Record type groups, all 233,543 records
ComplaintsNotices of ViolationCitationsTenant relocationUnfit building
020k40k60kENFORCEMENT RECORDSLand use60,996NOV 5,537 · Cit. 10,73297.4% Closed/CompletedWeeds45,866NOV 2,075 · Cit. 7,27097.9% Closed/CompletedConstruction36,801NOV 8,470 · Cit. 1091.7% Closed/CompletedNo category listed33,897NOV 8,514 · Cit. 1,28494.2% Closed/CompletedLandlord/Tenant20,325 · any labelNOV 500 · Cit. 6982.7% Closed/CompletedNoise13,822NOV 21 · Cit. 57398.0% Closed/CompletedVacant building11,946NOV 3,173 · Cit. 32096.8% Closed/CompletedOther and combinedlabels, 7,854NOV 2,172 · Cit. 4188.3% Closed/CompletedEmergency only2,036NOV 947 · Cit. 4196.7% Closed/CompletedRECORDS, 2003–2025
YearComplaintsNoticesNotices / 100 complaints
200346212927.9
20044,0661,85745.7
20054,2991,88343.8
20064,7172,41151.1
20075,1412,07640.4
20085,3201,56729.5
20094,8901,40328.7
20104,6481,21926.2
20114,4391,16126.2
20124,6581,26127.1
20134,9461,27925.9
20145,6511,46225.9
20156,8731,60223.3
20167,2061,41719.7
20177,7681,67321.5
201810,4571,30612.5
201911,7961,42312.1
202010,6379709.1
202110,5709699.2
202213,9531,0097.2
202314,6731,1187.6
202414,2881,2458.7
202515,9739696.1

Both lines count separately opened records, not case-matched enforcement results.

2003 is a partial year: the data begins in November 2003.

A ratio of notice records to complaint records is not a matched complaint-to-enforcement rate.

Category labelRecordsComplaintsNOVCitationsClosed/ Completed
Land use60,99644,7275,53710,73297.4%
Weeds45,86636,5212,0757,27097.9%
Construction36,80128,3218,4701091.7%
No category listed33,89719,7368,5141,28494.2%
Landlord/Tenant (any label)20,32519,7565006982.7%
Noise13,82213,2282157398.0%
Vacant building11,9468,4533,17332096.8%
Other and combined labels7,8545,6412,1724188.3%
Emergency only2,0361,0489474196.7%
All records233,543177,43131,40920,34094.6%

Categories use the category label in the City record (RecordTypeDesc). Every record is counted once.

No category listed holds every tenant relocation (4,045) and unfit building (318) record, plus 19,736 complaints, 8,514 notices and 1,284 citations.

Landlord/Tenant (any label) counts every record whose label names Landlord/Tenant, alone or with another category: Landlord/Tenant alone (17,413), Emergency + Landlord/Tenant (2,776), Land use + Landlord/Tenant (83), Construction + Landlord/Tenant (26) and 8 smaller labels (27).

Emergency only is the single label Emergency. Blank labels are counted as No category listed.

Other and combined labels covers 58 other labels, led by Emergency + Vacant building (1,963), Tree (1,926), Shoreline (1,285) and Construction + Tree (544).

Landlord/Tenant (any label), 20,325 records, is not the 56,258-record landlord/tenant selection used in sections 02 to 07.

Closed/Completed is an administrative status. It does not show that a violation was corrected.

02Missing dates

Then we found the holes in the record.

Some of the information needed to follow a case was blank, especially the last inspection date. Across all code types, 92,564 of 233,543 records have no last inspection date.

When we narrowed it to landlord/tenant complaints, the change by year was striking: 34.5% of complaints opened in 2017 have no last inspection date, compared with 80.8% of complaints opened in 2018. The gap stays high every year after.

The chart groups complaints by the year they were opened, using the data as the City publishes it today. It is not a picture of what the field showed back in 2018.

Landlord/tenant complaints with no last inspection date, by year opened
Under 20%20–60%Over 60%
0%25%50%75%100%2003200720112015201920232025% WITH NO LAST INSPECTION DATEYEAR COMPLAINT OPENED34.5% → 80.8%2017 vs. 2018 opening years2018–2025 OPENINGS: 80.8%–92.6%4.8%low, 201392.6%
Year openedComplaints% no dateNotices / 100
20038135.8%38.3
200443726.8%89.2
200540526.2%83.5
200654421.0%75.6
200749013.9%92.2
200843912.1%85.2
20094659.5%77.2
201046811.3%75.2
20115077.1%68.2
20125555.0%71.5
20136464.8%71.2
20146237.9%75.4
20156138.5%72.3
201664914.3%72.9
20171,09534.5%49.9
20182,80880.8%14.3
20192,96084.0%15.7
20202,81289.6%22.3
20213,00491.5%31.4
20224,82892.6%14.9
20235,15991.2%6.5
20245,21089.8%6.6
20255,74291.4%5.0

Grouped by the year a complaint opened, in the later historical dataset. Not a snapshot of what the field displayed in 2018.

A blank last inspection date is not proof no inspection occurred.

The 2018 opening-year split and the July 2025 date-column correction are separate questions.

03Closed

And what does “Closed” actually mean?

A record can be marked Closed or Completed without telling the reader what was decided. Was the problem repaired? Did the City find no violation? Was the case closed for an office reason, or because no one could be reached? Those are different outcomes.

50,721 landlord/tenant records are marked Closed or Completed. 61.6% of them show no inspection result in the City’s public data. The status alone does not tell a renter what was resolved.

One public status, several possible explanations
50,721Closed or Completedlandlord/tenant records90.2%of all five record types marked Closed/Completed56,258 records · 2003–2025WHAT “CLOSED” COULD MEANRepaired?No violation found?Closed administratively?Contact could not be made?Another reason?The status alone does not say which.Blue ring: other statuses, 9.8%
Status, all five record typesRecordsShare
Completed39,80970.8%
Closed10,91219.4%
Closed or Completed50,72190.2%
Under Investigation3,7496.7%
14 other statuses1,7883.2%
All selected records56,258100%
Record typeClosed or CompletedShare of type
Complaint36,559 of 40,54090.2%
Notice of Violation8,921 of 9,96189.6%
Tenant Relocation3,706 of 4,04591.6%
Citation1,258 of 1,39490.2%
Unfit Building277 of 31887.1%

Closed/Completed is an administrative status. It does not by itself establish verified repair.

How often each closure reason occurs is not published in the record.

04July 2025

Then the last inspection date column changed.

Before July 2025, the City’s public last inspection date column showed geographic coordinates (point data) instead of dates. This was not a handful of blank records. The column was not publishing dates at all.

In July 2025, dates appeared for past records. Only then could we see which records had a date and which did not, by the year they were opened. The July 2025 change and the 2018 jump are two separate questions.

We asked the City why the column held point data. While we waited for its records, we kept following how complaints moved through the system. See the report →

Two different data problems
A · THE PUBLISHED LAST INSPECTION DATE COLUMNMar 2025 · STLCA observes coordinatesJul 2025 · dates appear (historical backfill)Coordinates, not dates, until July 2025start not documentedB · COMPLAINTS WITH NO DATE, BY OPENING YEAR100%50%0%2018 cohort break200320082013201820232026YEAR (TRACK B: YEAR COMPLAINT OPENED)Track B is read from the backfilled column.Problem A is about the field. Problem B is about the cohorts.
Before Jul 2025The last inspection date displayed geographic coordinates (point data) rather than dates.
Mar 2025STLCA observes the coordinate values in the column.
Jul 2025Dates appear through a historical backfill.
2018 cohortSeparate question: complaints with no date rise from 34.5% (2017 openings) to 80.8% (2018 openings).

The 2018 opening-year split and the July 2025 date-column correction are separate questions.

05Landlord/tenant records

So we focused on landlord/tenant records.

A noise complaint or a land use case is not the same as a renter reporting a broken heater, mold, an eviction notice or an unsafe building. So we separated out the 56,258 landlord/tenant records: 40,540 complaints, 9,961 Notices of Violation, and the tenant relocation, citation and unfit building records.

Of those 56,258 records, 36,106 have no last inspection date, including 30,374 of the complaints.

What makes up the landlord/tenant records
ALL 56,258 RECORDS, BY RECORD TYPEComplaints 40,540NOV 9,961Tenant relocation 4,045Citation 1,394Unfit building 318COMPLAINT RECORDS ONLY · LAST INSPECTION DATE FIELDNo date · 30,374Date shown · 10,16630,374 / 40,540 complaints = 74.9% with no public dateNotice, relocation, citation and unfit-building records are excluded from this rate.010k20k30k40k50k56,258NUMBER OF RECORDSComplaints: 72.1% of the selection. Other types: 15,718.
Record typeRecordsShare of selection
Complaint40,54072.1%
Notice of Violation9,96117.7%
Tenant Relocation4,0457.2%
Citation1,3942.5%
Unfit Building3180.6%
All selected records56,258100%
Complaints onlyRecordsShare of complaints
No public last inspection date30,37474.9%
Date shown10,16625.1%

The selected 56,258 records are not 56,258 complaints. They are five record types.

The last inspection date rate counts complaints only. Other record types have different documentation workflows and are not part of it.

All-code records and landlord/tenant records are different populations.

06Living conditions

Were these housing conditions or paperwork?

Not every renter issue needs the same response. We sorted what renters wrote into problems with conditions in the home, paperwork and tenant protections, both, and other topics.

Among closed complaints about conditions in the home, 848 of 7,035 opened before 2018 have no last inspection date. For complaints opened from 2018 to 2025, it is 8,226 of 10,954. That makes the gap hard to explain as paperwork alone.

See where these complaints are on the citywide maps →

The maps count records, not confirmed problems or unique buildings.

Closed living-condition complaints with no last inspection date
0%25%50%75%100%12.1%848 of 7,03575.1%8,226 of 10,954Opened before 2018Opened 2018–2025% WITH NO LAST INSPECTION DATECOMPLAINT OPENING PERIOD+63 percentage points
What the complaint descriptions match
All complaints in categoryNo public last inspection date
05k10k15kHousing condition8,458 of 16,719 have no date · 50.6%Paperwork / tenant protection11,205 of 11,330 have no date · 98.9%Both1,606 of 2,329 have no date · 69.0%Other / unclassified9,105 of 10,162 have no date · 89.6%COMPLAINT RECORDS
CategoryComplaintsNo dateShare
Housing condition16,7198,45850.6%
Paperwork / tenant protection11,33011,20598.9%
Both2,3291,60669.0%
Other / unclassified10,1629,10589.6%
Habitability subset, opened before 20187,03584812.1%
Habitability subset, opened 2018–202510,9548,22675.1%

Topic matches do not establish legally verified violations.

The habitability comparison uses a separate strict text-match subset; 6,925 of its 9,074 missing-date complaints have a review-only housing-relevance flag.

07Complaints and Notices of Violation

We tried to connect complaints to Notices of Violation.

Counting complaints and notices is only a start. We tested which Notices of Violation could be tied to a specific complaint, using addresses and descriptions.

6,240 of 9,961 notices matched a complaint, and 3,721 could not be matched. At least 5,323 complaints connect to a notice.

That does not prove the rest had no action. It shows the public trail is incomplete: people have to rebuild the link between a complaint and a notice themselves. Read the full matching analysis →

Notices of Violation matched to complaints
9,961 NOTICE OF VIOLATION RECORDS6,240matched · 62.6%3,721set aside, unmatched≥5,323complaintrecordsSCALE: 1,000 RECORDSOne notice can link to a complaint;several notices can share one.Not established by this method.Not proof of no action.
Notice of Violation records reviewed9,961
Associated with a complaint6,240 · 62.6%
Set aside as unmatched3,721 · 37.4%
Complaint records with an observed matchat least 5,323

Method: address and description matches. Not exhaustive; no exact conversion rate.

Unmatched means not established by this method, not necessarily unrelated to rental housing.

08RRIO

Then RRIO raised another question.

Seattle’s Rental Registration and Inspection Ordinance (RRIO) checks basic rental conditions through registration and scheduled inspections. But a rental certificate and a later tenant complaint are two different records, kept in two different systems.

A program can report inspections and certificates while the public still cannot see what happened when a renter reported a problem afterward.

One property, two separate records
RRIO · REGISTRATION & INSPECTION RECORDSDCI · TENANT COMPLAINT RECORDRegisteredChecklist inspectionCertificateComplaint filedFindings?Resolution??TIME →Schematic of record structure, not dataA certificate records a moment. A complaint starts a new question.

RRIO certification is not a response to a subsequent complaint.

The schematic shows how the two records are structured. It does not count properties.

09What the record needs

We can count complaints. Now the public needs to see what happened next.

A useful enforcement record connects the complaint, first response, inspections, findings, Notices of Violation, deadlines, repair, verification and the reason for closing. That is the evidence policymakers need, not just a count marked Closed.

The questions the record should answer:

  • What happened after each complaint was filed?
  • Why was each case closed, and what shows it?
  • Why did the last inspection date column publish point data, and what changed it?
  • Where is the link between a complaint and its Notice of Violation?

STLCA brings public records together for renters, journalists, attorneys, researchers and policymakers. We are asking City Council for complete, linked enforcement data through a proposed Statement of Legislative Intent. See what we asked →

Housing policy should rest on evidence the public can follow.

The linked enforcement record
1Complaint2First response3Inspections4Findings5Notice ofViolation6Deadlines7Correction8Verification9Reason for closurePublished, as separate recordsPartly visibleNeeded, linkedOne case history, from the first report to the reason it closed.
1. ComplaintPublished, as separately opened records
2. First responseNeeded in a linked case history
3. InspectionsPartly visible in published fields
4. FindingsPartly visible in published fields
5. Notice of ViolationPublished, as separately opened records
6. DeadlinesNeeded in a linked case history
7. CorrectionNeeded in a linked case history
8. VerificationNeeded in a linked case history
9. Reason for closureNeeded in a linked case history

See the districts

Auto-reading01 / 14