Social work case management software compared — Penelope, Bonterra Apricot, Casebook, and the AI-native alternative. Documentation in 30 minutes, not 3 hours.
The case management software in front of you was designed when the bottleneck was collection — get the intake form on the worker’s caseload screen, land the right fields, write the case note in a Word doc, file the year-end outcome report. That bottleneck moved.
Collection is solved. The new bottleneck is the workflow that reads every case note, every assessment, and every follow-up survey on arrival — and surfaces the risk before the next supervision meeting. The year-end outcome report is no longer the answer. The client is the unit of work, and the client record has to be intelligent.
That is the difference between social work case management — the collection-era stack of intake form, case notes in Word, services log, and year-end outcome export — and social work case intelligence: one persistent client record, read on every arrival, with caseload, supervision, and funder reports generated as queries.
AI without a workflow is a clever intern with no desk. The social work teams winning with AI are the ones whose case data has a place to land — one record, one ID, one story.
Social work case management software is a platform that holds one persistent record per client across every interaction — referral, intake assessment, service or treatment plan, case notes, supervision, services rendered, and outcome follow-up — and reads that record so social workers and supervisors get answers without a spreadsheet merge. It replaces the typical stack of intake form, case-note Word documents, services log, supervision spreadsheet, and year-end outcome report with one connected record.
Used by:
Case management software for social workers, social services case management, social services software, social work software, caseload management software, case notes software, social work client database — different terms point to the same software category. What varies is which stage of the lifecycle the buyer is thinking about when they search.
A nonprofit CRM (Salesforce NPSP, Blackbaud, HubSpot) tracks donor relationships. EHR (Epic, Cerner) tracks clinical visits in a hospital system. Social work case management software tracks clients through a service delivery lifecycle with case-note narrative at its center.
The case management software your agency uses was designed when the bottleneck was collection — get the intake into the worker’s caseload, write the notes by hand, export the outcome report at year-end. The dashboard is no longer the answer.
Collection is solved. The new bottleneck is the workflow that reads every case note on arrival, surfaces the risk before the next supervision meeting, and keeps one persistent client record across child welfare, behavioral health, housing, and family services inside the same agency. The client is the unit of work, and the client record has to be intelligent.
AI without a workflow is a clever intern with no desk. The social work teams winning with AI are the ones whose case data has a place to land — one record, one ID, one story.
For years, Open Play’s program and facility data sat in paper logs and spreadsheets — including a water log Marco Botha had kept by hand for nearly three years. The foundation needed comparable, real-time evidence to put in front of investors and partners, not a quarter-end export — and nothing in the old setup could flag an operational problem while it was still fixable.
After a stretch of heavy rain, Open Play’s water purification system reported it had run out of rainwater — which Marco knew was impossible. Because the reserve, groundwater-extraction, and harvesting figures were now live on the record, he cross-checked them in minutes. The numbers immediately surfaced the real cause: a probable reservoir leak or a failed float switch — caught while still fixable. Same logic for a social worker: when intake assessments, services delivered, supervision notes, and follow-up surveys all live on one record, the impossible reading shows up on Tuesday, not at year-end.
“Those statistics that we’re now running on Sopact immediately showed me there’s something significantly wrong … things like that, we would never have been able to do in the past.”
Every effective social work case workflow moves through the same five stages — whether the program is child welfare, behavioral health, adult protective services, housing, family services, or a multi-program agency. Memorize them; the rest of this page is detail under each.
Referral, screening, eligibility, demographics, validated assessments.
Theory of Change, Logic Model, service plan, treatment plan — the agency’s theory.
Field definitions, validated-screen scoring, indicators, ownership.
Case notes scored, risks flagged, services joined to outcomes.
Caseload, supervisor, funder, court, HMIS — one query each.
Pick the lifecycle the client record carries (next section), assign a persistent client ID at first contact, then let every later stage write back to the same row.
The program changes, the documentation requirements change, the funder report at year-end changes. The record underneath — one client carried across every stage — does not.
In legacy social work case management, the record resets at case closure. In case intelligence, the record continues — referral through year-three follow-up — on the same persistent client ID.
One persistent client ID carries every case note, service delivered, supervision thread, and follow-up survey from referral through year three. Re-engagement arrives with the full case history attached.
Most social work agencies evaluating case management software are choosing between five recognizable names. Each was built for a different era; each made different trade-offs.
All five vendors will run a basic referral → service → closure workflow. The differences show up at the third question your funder, board, or county auditor asks: how do you know the outcome happened, where is the supporting case note, and what does the agency-level report look like for the same family across three programs?
No per-seat tax. No per-caseload meter. The line items are the things that actually drive work — how many programs share the client record, how custom the data dictionary needs to be, which built-in skills get activated, and how branded the funder-facing reports have to look.
Your case-note fields, your validated screens, your eligibility codes, your demographic categories — drafted from your existing intake form in one working session.
Case-Note Reader skill, Treatment Plan skill, Service Plan skill, Outcome Rubric skill, Cohort Roll-up skill — turned on by default, ready to read on arrival. Add domain skills like Court Report or Coordinated Entry when needed.
Logo, color palette, intake banner styled to your agency. White-label form, banner, and report design rolling out — funder and court-facing pages match your identity.
Closed-ended validated screens and open-text case notes scored together. Indicators populate automatically, with citation trails and attribution rules — which service drove which outcome, across the caseload.
Caseload · supervisor · court · HMIS / CoC · county BH commission · funder — each generated from one place with citation trails. The data dictionary keeps every metric reproducible.
One program is the simplest case. Three programs (child welfare + behavioral health + housing) sharing one client ID adds dictionary reconciliation and cross-program reporting.
One office is straightforward. Multiple sites under one agency add HMIS-aligned fields, federation-level roll-up, and BI integration to Tableau, Power BI, Looker, or Snowflake.
Closure summary is light. Year-1, year-3, and year-5 outcome cohorts attributed back to baseline assessment add depth to the dictionary.
Built-in skills cover common patterns. Your agency’s domain rubrics — encoded in plain English, versioned, composed with the built-ins — let the skill library grow with the practice.
One brand on one form is simplest. Multi-brand white-label — separate funder portals, separate client-facing portals — useful for fiscal sponsors and coalitions.
Reading reports in Sopact is included. Piping data via API to HMIS, billing systems (Medicaid, insurance), court reports, or your BI stack adds integration setup.
A 12-person family services agency running one program pays less than a 50-person multi-site community-action agency running six. Both pay for the complexity they actually use — the data dictionary, the skills, the white-label depth, and the reporting reach. Talk to us with your program list; we will quote against it directly.
Social work casework touches confidential information — child welfare records, behavioral health notes, family functioning, eligibility status. Here is what Sopact actually provides, named without overclaim.
AES-256 at rest, TLS 1.3 in transit, encrypted backups, scheduled key rotation. The reading model returns structured output and does not retain the underlying text.
Role-based access at the field level (worker, supervisor, director, auditor), single sign-on, multi-factor authentication, complete audit trail on every read and write.
All AI calls run under enterprise-grade service-level agreements with no training-data retention. The model reads, returns structured output, and forgets.
Sopact is not currently HIPAA-certified or covered by a Business Associate Agreement (BAA). The controls above support a security-conscious deployment, but if your agency is subject to HIPAA, FERPA, 42 CFR Part 2, state child-welfare confidentiality rules, or county behavioral-health requirements, you should evaluate the specific controls against your compliance program and consult your counsel and county/state contracts before storing protected information. We are happy to walk through the technical detail with your security and compliance team.
Reports are not formats — they are questions. Case intelligence produces four distinct shapes from the same accumulating client record. Naming the shape tells you which one a supervisor, funder, or court is actually asking for.
The case note overdue this week, the 90-day follow-up nobody returned, the supervision sign-off missed, the court report due tomorrow. Run continuously — surface gaps before the supervisor’s Monday meeting.
A case where session frequency dropped sharply, a screening-score swing, a length-of-stay that doubled, a worker’s caseload drifting from agency norms. Anomaly against history or against the group.
Everything known about one client — first referral to today. Every assessment, every case note, every service rendered, every supervision thread, every outcome survey. Court-ready and supervisor-ready in two clicks.
Caseload distribution, supervisor reports, HMIS / CoC, county BH commission, court compliance, funder outcomes — all as queries against the client records. The annual report drawn from one place.
Strip away the marketing and four properties separate social work case intelligence from everything before it. If a system you’re evaluating lacks any of them, it belongs to the era that ended.
Generic generative-AI tools are easy to adopt and hard to trust: ask the same question twice and you can get two answers. For social work casework that supervisors, courts, auditors, or boards will scrutinize, that variance is disqualifying. Sopact runs the model to read, then locks the answer, so the result is reproducible and auditable every time.
A validated screen (PHQ-9, GAD-7, VI-SPDAT, AAPI-2, custom indicator) tells you what changed. The case-note narrative tells you why. Sopact joins both on the same persistent client ID — so the score and the case worker’s clinical reasoning live on the same row in the supervisor or funder report.
Encryption at rest and in transit, role-based access at the field level (worker / supervisor / director / auditor), no training-data retention on AI calls, complete audit trail. Useful for any program; required for any agency whose contract or county auditor asks how the data is handled.
New intake question, new validated-screen scoring rule, new outcome cohort, new white-label brand — described in language, live immediately. The system bends to the agency’s practice instead of the agency’s practice bending to the system.
The architecture is the same; the configuration scales with the work. Here is what each size of agency typically runs through Sopact.
A single supervisor and 5–10 case workers. Intake form, case-note template, validated screen, and one outcome survey. Live in a week. No integrator, no IT staff required.
Multiple program directors, mixed funder portfolio. One family appears across child welfare, behavioral health, and housing — joined on a persistent client ID. Supervisor dashboards, court reports, and funder reports draw from one place.
Multiple physical sites, HMIS reporting, CoC compliance, county behavioral-health contracts, court oversight. Sopact reads HMIS fields, runs the agency roll-up across sites, exposes data via API.
Ten questions that come up in nearly every social work case management evaluation — answered here so the comparison work happens before the sales call, not during it.
Social work case management is one shape of the stakeholder record. The same approach holds applicants, students, trainees, clients, and grantees across every program an agency, foundation, or university runs.
One intake form. One validated screen. One year’s worth of case notes you already have. We’ll walk through how it would live as one record per client, what the AI would pull out of the long-form notes, and what the supervisor caseload and funder outcome reports would look like coming out of one place.
