Whether you're a brand-new admin setting up your first call program or an experienced supervisor running daily QA on real recordings, the right walkthrough is here. Start with the Setup Guide, or jump straight to the workflow you need.
From signing up to your team's first scored call, in five focused steps. About 30 minutes start to finish.
Setup, Building Programs, Running Training, QA on Real Calls, Coaching, Reporting. Pick the one you need.
If you've never used Turn Around Talk before, this is the path. Each step takes a few minutes; the last one is a real conversation.
Name, industry, and a few details. Your account starts with free credits to explore.
Add employees one at a time or by CSV. Each gets a temporary password emailed automatically.
Auto Build asks 10 questions about the role and produces scenarios, a rubric, a script, and an agent scorecard.
Pick the scenario, an employee runs the call, the platform scores it instantly.
Open the session, read the per-criterion feedback, and use the coaching script in your next 1:1.
A Call Program bundles a script, scenarios, rubric, and scorecard under one name. Build them with Auto Build, or assemble them manually if you prefer.
Name the program, optionally upload your existing call script, answer 10 questions about the role. The platform produces a suite of training assets you can selectively add to your library.
Specify the persona, trigger event, constraints, voice, and call type. Used when you want full control or when Auto Build wasn't quite right.
Define five criteria, each with a question, an ideal answer, a category, and a weight that totals 100. The basis for every score the platform produces.
Compatibility Checker scores how well any rubric pairs with any scenario for a specific call type. Avoid mismatches before they hit your team.
Once a Call Program exists, an admin assigns a scenario to one or more employees. The employee logs in, runs the call, and the platform handles the rest.
Pick the scenario and the employee. Optionally set a due date. The employee sees the assignment on their dashboard the next time they log in.
Pre-flight audio check (speakers + mic), then the call connects. The agent and AI customer have a full voice conversation. Click End Call when finished.
Open the session. See the overall score, per-criterion breakdown, transcript, and the coaching script ready for your next 1:1.
Agents can flag any score with a written reason. Disputes appear on your dashboard in red — open, review the call, leave admin notes, mark resolved.
Audio QA grades real customer calls — single uploads or bulk batches — against any rubric in your library. The output flows into the same reports as role-play sessions.
Drop an MP3, WAV, or M4A. Pick the rubric and the employee. The platform transcribes, redacts PII, and scores in the background. You'll get an email when it's done.
Drop a ZIP of dozens of recordings. Each becomes its own scored entry under a single audio library. Filename can encode the agent name for auto-tagging.
Same scored-session detail page as role-play. Per-criterion feedback, full transcript, recording deleted from disk after scoring (only the transcript stays).
Each upload becomes a library you can name, browse, and delete. Libraries roll up into the company-wide reporting view.
Coaching Intelligence reads all your scored calls (role play + real audio) and surfaces what the team is weakest at. Coaching Log records every coaching conversation you have with an employee for HR-grade audit.
Pick a rubric and a date range, click Run Analysis. The engine identifies team weaknesses, individual flags, and recommends targeted scenarios you can build with one click.
Each insight comes with a one-click "+ Build Targeted Scenario" button. The platform generates and saves it to your library, attached to the insight.
After you coach an employee on a call, log the conversation: method (1:1, group, written), notes, optional link to the insight that drove it. Builds your audit trail.
Per-employee printable PDF showing every coaching session you've logged, scores over time, and a signature block. Use for performance reviews or compliance.
Six headline stats, score trend chart, distribution donut, top-five leaderboard, top-five scenarios, coaching alerts. Everything links to the underlying detail.
The Reports page has a Call Program filter that auto-selects the rubric. See exactly how a specific call type is performing across role play and real audio combined.
Any filtered report can export to CSV — every session, every score, every criterion. Build whatever offline analysis you want.
Every rubric has a Print Scorecard button. One-page agent-friendly summary they can keep at their desk: what to say, what to avoid, how they'll be graded.
One at a time or via CSV. Each new account gets a random temporary password emailed automatically and is required to change it on first login.
From the employees page, click Reset PW on any row. New temporary password is emailed and shown on screen — read it over the phone if email is delayed.
Use the inline role dropdown on the employees page. Supervisors get most admin powers minus billing. No delete-and-recreate needed.
Settings page has separate panels for Role Play sessions and Audio QA minutes — or use the Buy Both bundle to combine them in one Stripe transaction.
Most setup tasks take under 5 minutes. Sign in to start, or see what features you have at your disposal first.
See Features Sign In