What are case management tools?
Case management tools are software for recording the people a program serves and the work done with them, from intake and assessments to services, case notes, referrals, follow-up and the reports funders ask for. The twelve here range from dedicated case systems to Sopact, survey platforms and general AI assistants, grouped by role rather than ranked.
THE SHORT VERSION
- Decide first whether you need a system of record for casework, connected evidence around the person, or both.
- Ask who will change a form or field a year after launch: an admin or consultant, or the program team that uses it.
- Test each shortlisted tool on one client’s journey: intake, notes and follow-up on one ID, and an AI summary you can trace to its notes.
What should decide between case management tools?
Four questions separate these tools better than a feature list: is it the system of record or the evidence around the person, who governs it after launch, do surveys, notes and follow-up land on one ID, and what does AI see?
A system of record runs the daily casework: caseloads, service plans, appointments, referrals and the reports a contract requires. Evidence around the person is what you learn over time: intake answers, assessments, caseworker notes, the exit survey and the 90-day follow-up. Most tools are stronger at one of the two.
Governance is the question demos skip: when only the admin or consultant who built a platform understands it, nobody dares change a field after that person leaves; when the program team can add a form itself, the workflow keeps up with the program.
This guide follows one fictional program: Dana manages employment support with three caseworkers, and her funder asks, “Which clients were still working at 90 days, and what got in the way for the others?” Lesson 4 of the free course, Govern the work, covers who owns what once a tool is live.
The 12 options at a glance
Nine of the twelve are case or client systems meant to hold the casework record; three collect or read the evidence around the person. Ask each vendor the last column’s question on your own client journey.
Scroll horizontally to see all columns →
| Tool or category | Built mainly as | Ask in your demo |
|---|---|---|
| Bonterra Apricot / ETO | Participant, program and outcome records | Who changes a measure after launch? |
| Salesforce | Case workflows in a broader CRM | What is native or configured, and who administers it? |
| Casebook | Casework, goals and reporting | Does a late follow-up land on the same participant? |
| Penelope | Service delivery and client engagement | Do forms, documents and access fit one service episode? |
| CharityTracker | Assistance records, referrals and outcome tracking | Can your staff set up your goals? |
| CaseWorthy | Case and program management | How are your reporting requirements configured, and by whom? |
| ClientTrack | Multi-organization client records | How is one person matched and shared across organizations? |
| Bitfocus Clarity Human Services | Client-centered casework, including housing | Who maintains your required reporting after go-live? |
| Exponent Case Management | Casework app on Salesforce | Which Salesforce responsibilities stay with your team? |
| Sopact | Connected evidence around the person | Do notes and follow-up share one ID, and does each AI line open a record? |
| Survey platforms | Intake, assessments and follow-up | Does each response land on the right participant? |
| General AI assistants | Help with permitted analysis | Which records and fields does it read? |
Five systems built as the record of casework
Apricot and ETO, Salesforce, Casebook, Penelope and CharityTracker all hold client and service records; the differences show when you ask who maintains them and where follow-up lands.
1. Bonterra Apricot and ETO
Apricot publishes case-management, outcomes and reporting capabilities; ETO supports collection and analysis of program activity and outcomes. Treat them as distinct products. In the demo, ask your coordinator, not the vendor, to change a measure and rebuild a funder view.
Sources: Apricot product information and ETO overview.
2. Salesforce
Salesforce offers case functions through service and industry products, and its Nonprofit Cloud program tools include participant case and care-plan capabilities. If you already run it, ask which requirements are native, configured or another application, and who changes them a year from now.
Source: Salesforce participant case management.
3. Casebook
Casebook documents participant goal tracking and outcome reporting. Test a participant with two services, a changed goal and a late follow-up, and see whether the follow-up lands on the same person.
Source: Casebook outcome tracking.
4. Penelope
Penelope’s documentation covers client engagement, communications and configuration. Use one service episode to see how forms, documents and staff activity fit together, and confirm the current product scope.
Source: Penelope client-engagement documentation.
5. CharityTracker
CharityTracker describes outcome tracking through goals, follow-up dates, needs assessments, assistance records and changes-over-time reports. Test those with your own outcome definition and ask who would adjust a goal once it is live.
Source: CharityTracker outcomes tracking.
Four systems for coordinated services and specialized requirements
CaseWorthy, ClientTrack, Clarity and Exponent fit programs with several funding streams, several agencies or a platform already chosen, so test the configuration and who carries it after go-live.
6. CaseWorthy
CaseWorthy describes an integrated case and program platform for human services, including outcome reporting. Test the configured workflow for your funding streams and staff roles, and ask for evidence on specialized reporting.
Source: CaseWorthy platform.
7. ClientTrack
ClientTrack’s documentation includes client records, enrollments, assessments, goals, services, referrals, notes, reporting and organization-access controls. For multi-agency work, test what is shared, what stays restricted and how one person is matched across records, including a duplicate.
Source: ClientTrack user and administrator guides.
8. Bitfocus Clarity Human Services
Clarity supports client-centered casework, services and reporting, and its published scope covers housing and wider human-services settings. Confirm who configures and maintains your required reporting.
Source: Bitfocus case-management products.
9. Exponent Case Management
Exponent Case Management is a Salesforce application with intake, assessments and service-planning functions, relevant when a team wants casework on Salesforce. Clarify which responsibilities sit with the provider, your Salesforce administration and your own staff.
Source: Exponent Case Management.
Three tools for the evidence around the person
Sopact, survey platforms and general AI assistants collect or read what clients say and caseworkers write; they differ in whether that evidence stays on one ID and what AI may see.
10. Sopact
Sopact Sense gives each person a persistent ID at the first form, and assessments, case-note forms, exit surveys and follow-ups attach to that record as they arrive, managed by the program team without an IT ticket. It is not a full case system of record or a finance system, so test the connection if a case system stays.
An Intelligence Cell reads each note or open answer on arrival with your team’s prompt. Field selection keeps contact details out of AI, the AI Assistant stays locked until you pick its surveys, and every line of an answer links to a record.
11. Survey platforms
Surveys serve intake, assessments and follow-up, and some platforms support contact directories, identifiers and repeated collection. Test whether your setup puts each response on the right participant, using a changed email address, a second enrollment and a missing wave.
12. General AI assistants
An assistant can help with permitted analysis and documentation, but a consumer chat and an organization-managed workspace are different setups. It reads what you give it, so decide which records and fields it receives and check any summary against the notes.
What does one client’s journey show about two options?
Running one fictional client through two shortlisted options shows where each form lands, who can change it, and what the funder’s question costs to answer.
WORKED EXAMPLE · FICTIONAL
Each client completes intake, a needs assessment and an exit survey, and caseworkers write a note after each meeting. Ninety days after exit, a follow-up asks “Are you still working?” and “What has made it hard to keep the job?” Dana runs one client, Rosa, through a configured case system and through Sopact Sense.
In the case system Dana priced, a consultant sets up intake, assessments, notes and referrals with access rules per caseworker. The follow-up stays in the survey tool, so Rosa’s answer arrives under a new email and is matched by hand, and adding a childcare question means a request to the consultant. Some case systems send follow-ups themselves; test that on your shortlist.
In Sopact Sense, Rosa gets an ID at intake, and the assessment, notes, exit survey and follow-up attach to that record as they arrive. An Intelligence Cell reads her answer, “My shift changed and I lost my childcare,” with Dana’s prompt and returns Barrier: childcare, with the sentence attached. Dana adds that childcare question herself.
Sopact Sense does not run appointments, referral hand-offs or contract-mandated reporting. Dana has no mandated system and her pain is follow-up and notes, so she starts there, scoping the AI Assistant to those forms with contact fields left out.
Some limits no tool removes. Clients who did not answer stay unknown, not unemployed, and “still working” is self-reported. An AI summary is a proposal a caseworker checks against the notes, and showing the program caused a result needs a design such as those in attribution and contribution.
When is a dedicated case system the right call?
Choose a dedicated case system when casework itself is the daily job: caseloads, service plans, appointments and referrals, several agencies sharing one client record under access rules, or a contract that requires a specific system or reporting standard. If you already run Salesforce with an administrator who is staying, an option such as Exponent may fit better than a second platform.
Start with governed collection when the pain is evidence rather than casework: follow-ups matched by email, notes nobody reads, and the same report rebuilt every quarter. Many teams need both, connected.
How should tools connect to Salesforce, HubSpot or another system?
Evaluate the specific connection between your systems, not a promise that one interface solves every integration. APIs, exports, native apps and middleware can each be right; an AI-tool protocol such as MCP does not by itself resolve identity, permissions or field meaning.
Ask of any connection: which system owns identity and each outcome, which identifiers survive the transfer, who may see the connected records, how corrections are handled, who repairs a missed transfer, and whether you can export everything if you leave. Then have the vendor demonstrate your actual source and destination, since a connector logo proves nothing about your fields.
Test your shortlist on one client journey
Build a small, authorized test set around one fictional client, run it through every shortlisted tool, then run the next cycle too, because the second cycle shows who does the work.
- Write the funder’s question and the decision it serves, as Dana did, and agree the expected results before any demo.
- Create one fictional client with two enrollments, three case notes, a referral, a changed email and a missing follow-up.
- Have each vendor take that client from intake to follow-up on one ID, and note every manual match.
- Have your own staff add a question to the assessment, and write down who would make that change after launch.
- Ask the AI feature for the client’s email, then for the barriers in the notes: the first should be unavailable, and each line of the second should open its note.
- Reproduce one headline figure with its denominator and sources, and record the hours the second cycle takes.
After the first cycle you should have one client’s journey in each tool, a named owner for changes, a list of fields AI may read, and one figure you can trace. Related guides: case management CRM, outcome tracking software and funder outcome reporting.
Frequently asked questions
Which case management tool is best?
There is no universal best tool, only the best fit for your casework and your team. Start with requirements such as mandated reporting, multi-agency sharing and scheduling, then run the same fictional client through each shortlisted product. Choose the one your staff can maintain after launch and whose results trace back to records.
Do we need more than one tool?
Possibly, but do not assume it. One product may cover both casework and evidence, or a case system of record plus a connected evidence workflow may fit better when surveys, notes and follow-up keep ending up in separate files. If you run two, decide which one owns identity and test the connection before you commit.
Can historical case notes be migrated?
Often, depending on the source system, permissions, file formats and data quality. Test the migration with a real sample, preserve relationships and dates, and reconcile counts before and after. Similar names are a reason to review two records, not proof they are one person, and no connector guarantees a complete transfer.
Can AI read case notes safely?
AI can help code and summarize authorized notes, but safety depends on the setup. Ask which fields reach the model, whether names and contact details stay out, which records the assistant may use, and whether each line of a summary opens its note. A caseworker should still check a summary against the notes before it informs a decision.
Who should own a case management tool after launch?
The team that uses it should be able to make routine changes, such as adding a question or a follow-up form, without waiting on a consultant. Name one data steward for IDs, definitions and AI access, and keep a change log of who approved what. If only an outside admin can change the tool, plan for the day they leave.
