Last updated: October 13, 2026
Once you score every call, the hard part is acting on the results. Itero (iteroapp.ai), AI roleplay and call scoring software for customer-facing teams, scores 100% of the calls it receives against your own rubric. It flags a missed disclosure with the exact transcript moment, and a manager, or an automation you build on Itero's public API, routes the agent into roleplay practice on that skill.
Itero is not affiliated with iTero, the intraoral dental scanner made by Align Technology.
If your QA team reviews about 2% of calls, how do you score every call?
Replace sampling with automated scoring on every recorded conversation, and keep your analysts for calibration and coaching. With Itero, calls arrive automatically from your recording or dialer platform, and each one is scored against the scorecards you assign to that call type.
Calls reach Itero in four ways:
- Native integrations: Gong, Outreach and FrontSpin are documented on Integrations; Itero's scoring page also lists Microsoft Teams and Five9.
- Public API: push calls from any recorder or dialer, with metadata and tags, and receive scores back by webhook (Public API). Real chat transcripts sent this way are scored with the same scorecards as calls; there is no native helpdesk integration.
- No-code tools: n8n, Zapier or Make can move recordings into Itero without custom code.
- In-person conversations: reps record with the Itero mobile app, and the conversation is scored in about two minutes (Virtual Ride-Alongs).
For the cost and coverage case against sampling, see Manual Call Review vs Automated Scoring.
What software can score 100% of customer calls against a custom rubric?
Look for a tool that scores against your rubric rather than a generic one, shows its evidence, and lets you run more than one rubric per call. Here is how Itero handles each requirement.
| Requirement | Itero |
|---|---|
| Build from your existing rubric | Upload a rubric, playbook or QA checklist and Itero drafts the scorecard; drafts are reviewed before publishing |
| Compliance and quality in one pass | Stack several scorecards on one conversation, scored in the order you choose |
| Right rubric for each call type | Assign scorecards by call type, product line, user group, client or customer segment |
| Evidence for every score | Each criterion shows the question behind it and every supporting transcript citation |
| Re-scoring | Re-run one evaluation type without re-processing the rest |
| Channels | Voice calls, plus real chat transcripts sent in through the public API, with channel-aware scorecards |
| Public API | Standard on every account: push calls in, start evaluations, read results, and create or edit scorecards |
| Changing a scorecard | Edit in the app, or ask an AI agent with Itero's Agent Skills; new criteria can be added to a published scorecard without unpublishing it; changes apply to evaluations run after the change |
In one customer's comparison against their own QA rubric, Itero's scores matched 95% of the time (Itero, confirmed by the company, October 2026).
How other vendors handle real-call scoring
Based on each vendor's public pages as of October 13, 2026:
| Vendor | Real-conversation scoring, per its public pages |
|---|---|
| Hyperbound | Its Perform plan scores 100% of real conversations against custom methodology scorecards, with Gong, Salesloft and Chorus integrations, and assigns targeted roleplay when it finds a gap |
| Second Nature | Managers can pull a real conversation into the platform, have it analyzed, and assign practice; automatic scoring of every call is not described |
| Solidroad | Evaluates 100% of support interactions across phone, chat, video and email, including AI agents, and generates training simulations from QA findings |
| Itero | Scores every call it receives, and real chats sent through its API, against your scorecards, with yes/no compliance checks and transcript citations. Managers assign practice from the Coaching Inbox; automatic assignment is something teams build with the public API |
Hyperbound and Solidroad both describe a closed loop from scoring to practice that runs out of the box. Itero's loop is open instead: every step is in the public API, so a team or its AI agent decides the trigger and the response. Itero's other emphasis is regulated-industry compliance checks (insurance, lending, TCPA) and one rubric shared by practice and live calls across sales and service teams.
How does a missed compliance disclosure get flagged?
Write the disclosure as a yes/no criterion, such as "Did the agent read the call-recording disclosure?" or "Did the loan officer state the APR?". Itero checks it on every call that scorecard applies to.
When a call fails the check, the miss lands in the Coaching Inbox with the rep, the call, the requirement missed and the timestamp. The manager plays the audio snippet for that exact moment instead of scrubbing the full recording. Insurance teams can set requirements by state and product line; lenders can check TILA/RESPA, fair lending and TCPA items (AutoQA & Compliance).
For TCPA specifically, see Compliance Call Monitoring: The 2026 TCPA Playbook.
How do you keep the rubric current when a regulation or script changes?
Treat the scorecard like any other document your team maintains. With Itero's Agent Skills installed, an AI agent such as Claude Code or Codex can read the new disclosure language or policy memo and add or reword the matching criteria through Itero's public API. It shows each change and waits for your confirmation. New criteria can go into a scorecard that is already published, so scoring does not pause while you update it. Scorecard changes apply going forward: evaluations run after the change use the updated criteria, and calls already scored keep the scores they received, so trends stay comparable and nobody's past results shift under them.
The same agent can also work the results. The conversations skill searches calls, reads transcripts, starts evaluations and explains a score criterion by criterion, which helps a QA lead check whether a new criterion behaves as intended on real calls. See API & AI Agents.
How do you automate the loop between QA and agent coaching?
Connect three steps that usually live in separate tools: score, flag, practice. Itero runs all three on one scorecard.
Phase 1: Score every call
- Assign the compliance and quality scorecards to each call type.
- Let calls flow in from your integrations or the API.
Phase 2: Surface the misses
- Filter the Coaching Inbox by rep, skill area or severity.
- Open the flagged moment with its timestamp and transcript citations.
Phase 3: Route into practice
- Assign a roleplay scenario that drills the missed skill, such as a focus scenario that rehearses the disclosure moment. A manager can do this from the Coaching Inbox.
- To make it automatic, build it with the public API. A script or AI agent finds calls that failed a criterion, reads the criterion-level result and assigns the learning path that holds the drill scenario. The public endpoints are
POST /api/public/v1/call/get-calls(find scored calls and chats, filtered by scorecard and score range),GET /api/public/v1/evaluation/{evaluationId}(the criterion-by-criterion result) andPOST /api/public/v1/learning-path/{id}/assign(assign a learning path, whose stages are practice scenarios, with a due date). The same pieces support any other rule you want, such as assigning only after two misses in a week. - Use a certification with a pass threshold if the agent must re-qualify before taking that call type again.
Phase 4: Check the next real call
- Practice is scored with the same criteria as live calls.
- Track whether the violation rate for that rep and requirement drops on later calls.
If you want results in another system, Itero sends score events by webhook (CallQaScore and CallQualitativeScore), so a failed check can also open a ticket or post an alert.
