For contact center and customer operations leaders
Score and coach every call, not one percent
Agents ask Tiersel out loud during a call and get the customer's history, the policy answer and a reply to read. Supervisors get every call scored against your rubric and a live read on the queues. Every refund, credit, coaching note and staffing move waits for a person to approve it.
Live demo on synthetic data · Built on Jev by TypeSafe AI
An animated example types requests as if they were spoken and shows the card Tiersel builds for each one.
For the VP of Customer Operations or Head of Contact Center at an enterprise or BPO
Your QA team hears a sliver of the calls your agents take.
An analyst can review a handful of calls per agent each month, so coaching rests on a sample too small to trust. During a call, agents switch between the CRM, the billing system and the knowledge base while the customer waits. Supervisors see a service level dip after it has already turned into abandons. Every agent who leaves takes their ramp time with them. The information to fix each of these is already in your systems. Nobody has the hours to pull it together while the call is still live.
4 agent workflows
Built for the work your team already does.
Each workflow starts from a sentence. The agent calls your systems, shows its evidence, and waits for a person before anything consequential happens. Every card below is the real, working interface.
Workflow 01 · Tier 1 agent
Live assist on a billing call
01 · Trigger
Customer calls about a charge
The agent asks for help out loud. The transcript streams as the customer speaks, with sentiment and intent updated on every turn.
02 · Tool call
Identify and verify
Matches the caller in the CRM and ticks off each verification and disclosure step as the agent completes it.
03 · Tool call
Check billing and policy
Finds both charges in the billing system and searches the knowledge base for the duplicate charge policy and the agent's refund limit.
04 · Reasoning
Draft the answer
Writes a suggested reply that covers the refund, when it posts and the disclosure still open, with each policy point cited.
05 · Human approval
Agent approves the refund
The refund is prepared at the right amount to the right card. It is issued only when the agent approves it.
Nothing happens until a person approves06 · Outcome
Resolved on the call
The customer hears the fix and the timeline before hanging up, and a case note is added with the approval.
What changes
The agent settles a billing dispute without leaving the call to search three systems, and every step is on the case record.
Workflow 02 · QA analyst or team lead
QA on every call
01 · Trigger
A call ends
Every completed call is queued for scoring, not a sample.
02 · Tool call
Transcribe and split speakers
Pulls the recording, transcribes it and separates the agent's speech from the customer's.
03 · Tool call
Score against your rubric
Scores greeting, verification, empathy, resolution, compliance and close with your weights, quoting the evidence for each.
04 · Reasoning
Find the coaching moment
Picks the weakest criterion, links the exact moment in the recording and drafts a short coaching note.
05 · Human approval
Lead sends the coaching
The team lead reads the evidence, edits the note if needed and approves sending it to the agent.
Nothing happens until a person approves06 · Outcome
Feedback tied to a real call
The agent gets specific feedback on a moment they remember, and the score joins their QA history.
What changes
QA coverage moves from a handful of calls per agent to every call, and analysts spend their time calibrating and coaching instead of listening.
Workflow 03 · Contact center supervisor or WFM analyst
Queue pulse and staffing moves
01 · Trigger
Supervisor checks the floor
A supervisor asks how the queues look, or a queue drops below its service level target.
02 · Tool call
Read live queue stats
Reads service level, calls waiting, longest wait, handle time and abandons for each queue from the contact center platform.
03 · Tool call
Compare with the forecast
Reads the intraday forecast and the skills matrix from workforce management to see which queues run over and who is cross-skilled.
04 · Reasoning
Model the move
Estimates how many cross-skilled agents bring the lagging queue back to target without pushing the donor queue below it.
05 · Human approval
Supervisor approves the move
The skill change is prepared with an automatic return time. Agents move only on approval.
Nothing happens until a person approves06 · Outcome
Act during the spike
Agents shift while the spike is still happening, and the change reverts on schedule.
What changes
Supervisors act on a service level dip in the interval it starts, and see the cost to the donor queue before they approve.
Workflow 04 · Agent, escalations specialist or retention team
Customer lookup with a next best action
01 · Trigger
A customer is on the line
The agent asks for the record by name, phone number or order number.
02 · Tool call
Pull the profile
Reads tier, lifetime value, plan and recent CSAT from the CRM and billing system.
03 · Tool call
Assemble every contact
Lists calls, chats and emails from the last 90 days with outcomes, plus open cases and their SLA status.
04 · Reasoning
Explain the churn risk
Weighs signals such as a downgrade question, repeated billing errors and a missed callback, and shows how much each one counts.
05 · Human approval
Choose and approve the action
Ranks save actions by expected save rate and cost. The agent picks one and approves it.
Nothing happens until a person approves06 · Outcome
One view, one decision
The agent starts the conversation knowing the history, and the action is logged to the customer timeline.
What changes
Agents stop asking customers to repeat their history, and every save offer carries its reason on the record.
How it plugs in
An agent on your systems, one component per workflow.
Requests come in by voice, text or system event. The router picks the workflow, the agent calls your systems, and the answer lands as a live card where your team works. Nothing is written back until a person approves it.
In
Tiersel
Your systems
Out
Workflow blueprint · Tier 1 agent
Live assist on a billing call
- Starts when
- Customer calls about a charge. The agent asks for help out loud. The transcript streams as the customer speaks, with sentiment and intent updated on every turn.
- Reads
- Identify and verify. Matches the caller in the CRM and ticks off each verification and disclosure step as the agent completes it.
- Check billing and policy. Finds both charges in the billing system and searches the knowledge base for the duplicate charge policy and the agent's refund limit.
- Decides
- Draft the answer. Writes a suggested reply that covers the refund, when it posts and the disclosure still open, with each policy point cited.
- Waits for
- Agent approves the refund. The refund is prepared at the right amount to the right card. It is issued only when the agent approves it.
- Writes back
- Resolved on the call. The customer hears the fix and the timeline before hanging up, and a case note is added with the approval.
This workflow is one component
export const agentassist = defineSkill({id: "agentassist", category: "cx",// when this workflow, and not a neighbour, is meantcriterion: "A contact center agent is on a live call or chat with a customer and wants real-time help: what the customer needs, what to say, policy answers, and the resolution to approve. Not a customer's record on its own, and not reviewing a finished call.",examples: ["assist me on this call, the customer was double charged", "the caller wants to cancel, what do i say"],match: (text) => score, // intent score, runs on every wordparse: (text, ctx) => record, // entities + tool calls into your systemsreply: (record) => "…",Card: LiveAgentAssistCard, // the live UI, with its approval step});
The id, criterion, examples and reply are read live from the skill that renders the card above. Adding a workflow for your team means adding one of these: its intent, the calls it makes, what it says, and the interface your people approve in.
Fits your stack, answers to your people
Reads your systems. Acts only with approval.
What the agent connects to
Contact center platform
CRM and case management
Workforce management and QA
Knowledge base
Billing and orders
Product names are examples of systems in this category. Each deployment is connected to your systems through their APIs as part of the build.
A person approves every consequential action
Refunds, credits, retention offers, coaching notes and staffing moves are prepared by the agent and carried out only after a named person approves them.
Every step is logged
Each system call, the evidence it returned and the approval or rejection are recorded, so a supervisor or auditor can replay how a decision was made.
Synthetic data in this demo
Every customer, agent, call and number on this page is invented, so nothing you try here touches real customer data.
Least privilege, in your environment
When deployed, Tiersel reads and writes only through the permissions you grant, and customer PII stays in your environment.
The cost of doing it by hand
Put your own numbers in.
Move the sliders to match your operation. Nothing is sent anywhere unless you choose to email it.
Each year
12,000
staff hours back, about 5.8 full-time people
$360,000
of staff time, at your hourly cost
Email me a pilot plan for these numbersCounts only the minutes around the conversation: looking up the customer, searching the knowledge base, writing wrap-up notes and manual QA review. It assumes no shorter conversations and no fewer agents. The hourly default starts from the BLS median wage for customer service representatives, $21.53 in May 2025, plus about 40 percent for benefits and overhead.
Hours = calls a month × 12 × minutes each × share, ÷ 60. A full-time person is 2,080 hours a year. Default hourly cost from U.S. Bureau of Labor Statistics, Occupational Outlook Handbook, 2025.
Questions buyers ask
Straight answers.
No. Tiersel sits on top of the systems you already run. When deployed, it reads from them and writes back through their APIs with the permissions you grant. The demo on this page uses synthetic data and connects to nothing.
When deployed, it stays in your environment, and each tool call is limited to the fields the workflow needs. In this demo there is no customer data at all: every record is synthetic.
Every score shows the quote and timestamp it rests on, so an analyst can check it in seconds. In a pilot you calibrate it against your own analysts on the same calls before anyone relies on it, and a lead still approves every coaching note.
No. It prepares the refund, credit, offer or staffing move and shows why. Nothing happens until a person approves it, and the approval is logged with the evidence.
They stay on every conversation. Tiersel does the lookup, the note and the first pass at QA so agents can listen to the customer and leads can spend their time coaching.
Try the live demo on this page now. A pilot starts with one queue and one workflow, read-only at first, so your team can compare its scores and suggestions with your own before anything writes back.
See it on your own workflow.
Watch the contact centers workflows run end to end, or speak to the agent yourself. Then tell us which one to build on your data.
Scope a pilot with hello@sasid.ai- 01
Pick one workflow
A short call to choose the workflow that costs you the most time, and who approves its actions.
- 02
See it on your data
We build that workflow against your sample or de-identified data so your team judges the real thing.
- 03
Run it with people in charge
Your staff use it on live work. The agent drafts and prepares; a person approves every consequential step.
- 04
Decide from results
You keep what measurably saves time and drop what does not.
Other operations
