County staff unveil new justice-system data dashboards; presentation highlights program outcomes, data limits

Marathon County officials presented a new justice-system data project that pulls 2024 program data from a vendor case-management system and displays it in Power BI dashboards and regular reports for staff and the public. The presentation described efforts to standardize hundreds of event codes, build repeatable reports, and produce weekly case-manager dashboards intended to improve transparency and management of diversion, treatment and supervision programs.

County Administrator Lance said the project is intended to help the county answer policy questions about program performance and to give the public clearer information about “how much do we do, how well do we do it, and is anyone better off.” Michael Schultz, Marathon County Data Officer, and Laura Yari, Justice Systems Coordinator, described the technical work: reducing inconsistent coding, building data-cleaning rules and automating Power BI reports that refresh from the vendor system (1Solution).

“Now that we're able to pull information through lots and lots of coding, that black box that you see there is just pages and pages of me manually programming what we need to pull out of these systems,” Michael Schultz said, explaining the manual ETL and reporting work he undertook to create sustainable dashboards.

The presentation covered multiple justice programs and example metrics drawn from 2024 entries in the vendor system and from the county’s pilot programs:

- Diversion programs (deferred prosecution agreements and deferred entry of judgment): one full-time case manager had 81 admits and served 133 people over 2024 (carryovers included); another caseload had 53 admits and served 82 people with a lower success rate attributed in part to staff turnover.

- Domestic-violence misdemeanor diversion: 71 admitted and 92 served in 2024; success rates and extensions were reported and discussed.

- OAR (operating after revocation) diversion and SSTOP: OAR had 28 admitted and 46 served (10 successful, 5 unsuccessful; 9 of 15 license reinstatements noted as a result); SSTOP (a program to get people assessed and into treatment to regain driving privileges) showed 44 discharges and 32 immediate legal reinstatements; the program reported reduced jail bed days where judges documented a saved sentence, though staff cautioned jail-day savings are undercounted if judges do not state alternative sentences on the record.

- CMTS (case management and treatment services): an intensive outpatient program shared by Marathon County and the Department of Corrections admitted 132 people and served 188 in 2024, with high dosage hours but lower-than-target success rates; staff noted this population is high-risk and housing instability complicates outcomes.

- Pretrial PSA pilot: Marathon County completed 1,504 Public Safety Assessment (PSA) pretrial assessments in 2024 as part of a pilot funded by the Wisconsin Department of Justice; the PSA results were used to inform release and supervision decisions and to assign participants to pretrial levels (levels 1–4). Staff reported the PSA appears predictive for reoffense among high-risk (level 4) individuals but that other level breakouts require further evaluation.

- Day Report drug-testing supervision: 2024 totals included 208 admits, 301 persons served, 7,498 face-to-face tests scheduled and 1,727 failures to report; positive tests and no-shows were routed to the referring agency or case manager for follow-up.

Laura Yari said the dataset now reports timestamps, the staff member who entered each record and program-specific event codes. She credited Michael’s technical work and cooperation from Attic case managers for enabling weekly automated reports. "These are the exact numbers that are coming from the data that's been entered for the year 2024," she said.

Speakers repeatedly cautioned that the data are “accurate but not precise.” Lance and Michael emphasized the project’s limitations: legacy system inconsistencies, human error in data entry, and necessary assumptions made when cleaning codes. Michael said the team reduced more than 600 event codes to about 250 standardized codes and created “cheat sheets” and training for case managers to sustain the improvement.

The dashboards are intended to prompt follow-up questions rather than to supply definitive causal conclusions. County staff recommended further analysis to compare treatment-court referrals with Department of Corrections referrals into CMTS and to identify whether referral timing or case-management practices explain differing success rates. Staff also said they intend to make 2025 dashboards filterable by month, program and case manager.

Ending

The committee received the presentation and asked follow-up questions about staffing, training, and how to use the dashboards to inform policy choices. County staff said the work will continue, with more program pages to be rolled out and additional validation with partner agencies.