The screen between your systems and your agents.
Five surfaces, one shared record, every action written to the ledger.
Claim it, check it, call it.
One prioritized queue, one contact at a time, with the preflight, the phone handoff, and the disposition on the same screen.
Presence and status policy
Available, On call, Wrap-up, or Break; the time in each status is visible to managers and governed by policy.
Prioritized queue views
Power Hour, Callbacks, Inbound, Coverage, and Overnight are views over one canonical work item, never copies.
Claim and release with leases
One winner per work item; a lease is released or expires, so two agents never hold the same contact.
Contactability preflight
Consent, local time, attempt caps, and channel blocks are checked before every touch, per channel.
Click-to-call and transfer handoff
The number opens in the agent's own phone app and transfers hand off to the provider; DialSplice never carries the call.
Structured dispositions and versioned guidance
Every outcome is a disposition from your list; the script beside it is versioned and approved before agents see it.
Five-minute inactivity protection locks the workspace; unacknowledged alerts stay in an in-app recovery inbox.
Coverage you can see, not guess.
Queue health, presence, holds, and the call log on one desk, with the approvals that keep callbacks inside policy.
Queue health per view
Counts, SLA risk, and the oldest item for Power Hour, Callbacks, Inbound, Coverage, and Overnight.
Overnight and coverage assignment with batch review
Automatic assignments arrive as a location-scoped batch with fairness evidence; a manager releases them.
SLA-risk deferral approvals
A callback that would slip past its window needs a manager's approval to move, and the approval is logged.
Attempt-cap and reconciliation holds
Work that hits its cap or returns an ambiguous provider outcome is held for a person, not retried.
Call log with content-limited CSV export
Time, direction, location, agent, outcome, and sync state; no message bodies, no recordings.
Non-PHI feedback triage
Agent feedback moves through new · triaged · planned · closed without carrying contact identity.
One contact. One history.
Ownership, language, location, inquiries, appointments, payments, and every call, SMS, and email — read from the systems that hold them.
Permissioned search
Results are filtered by role and location scope before they leave the server.
Existing patient record lookup
Before booking, the agent sees whether the contact already has a patient record in your EHR, and at which locations.
Repeat-lead rule
A returning contact updates an open opportunity; a new one is created only after close or abandonment.
Cross-location history where permitted
A contact seen at two locations shows both histories only to roles scoped to both; an ambiguous match is held, never guessed.
Payments and card on file
Take a payment over the phone or update the card on file without leaving the workspace; the action is logged like every other.
Booking into the system of record
Bookings write to your EHR with identity resolved first; the CRM is updated from the same event.
Versioned, approved, and on the ledger.
Configuration changes are revisions with a content hash; every activation, rollback, and read lands in the same append-only ledger.
Versioned location configuration
Each revision carries a content hash; approval, activation, and rollback are separate, logged steps.
Roles and scopes
Capability and location scope are assigned per role and validated at request time, not at login.
Published guidance assets
Scripts and playbooks are versioned and approved before an agent sees them beside a contact.
Hash-chained audit ledger
Every event links to the previous by SHA-256; the ledger stores events and hashes, not content.
Operational telemetry with attribute allowlisting
Only allowlisted attributes leave the request path; identity and content never enter telemetry.
Measure the work. Show the impact.
Every call, callback, and booking is a ledger row, so the numbers come from what happened, not what was reported.
Speed to lead
Time from a new lead to the first attempt, by agent, location, and queue view.
Connection and schedule rate
Attempts that connected, and connections that booked, counted from the ledger rather than self-report.
Call duration
Connected-call time per agent and per outcome, so coaching starts from evidence.
Show rate and downstream impact
Appointments kept, read back from your system of record and tied to the calls that produced them.
Accountability views
Per-agent and per-location figures that managers and agents both see; the leaderboard uses the same numbers.
Export to your data platform
Content-free facts flow to Fabric, AWS, or Google Cloud for your own reporting.
Not in the box.
DialSplice is bring-your-own-telephony. Several of these already live in your phone system or your other tools; DialSplice does not duplicate them, and it does not replace them.
Predictive or power dialing
Your telephony provider's, where it offers one
Recording, transcription, or automated quality scoring
Stays in your telephony provider or QA tool
Workforce forecasting
Your workforce-management tool's
A patient or customer self-service portal
Your EHR's or CRM's
If you need those, we'll point you to the tools we've seen work beside DialSplice.