Glossary

A few words.
A clearer picture.

From customer satisfaction to AI handoffs, get to know the terms behind better support.

Tickets & volume

The work in your queue.

The questions customers ask and how support teams handle them.

WISMO Where Is My Order

“Where is my order?” A customer asking about the status or delivery of a purchase. A useful reply starts with the current order details.

Related: Stop WISMO tickets

WISMR Where Is My Return / Refund

“Where is my return or refund?” A customer asking what happened after sending an item back or requesting money back. The answer depends on the return and refund records.

Related: Automate return & refund status

Ticket deflection

Answering a customer’s question through self-service or AI before it needs a support representative. A useful measure counts solved questions, not abandoned attempts.

Related: Deflect without frustrating customers

Deflection rate

The share of contacts resolved without a person’s help. Read it alongside reopened tickets and satisfaction to check whether customers got the answer they needed.

Related: Measure deflection rate honestly

AI agents

AI and your team.

The different ways AI can help, from drafting a reply to carrying out an approved action.

AI agent

Software that uses information to plan and carry out tasks. In BearScope, an AI agent has a defined role and permissions that govern what it can do.

Related: AI agents for support, explained

Agent assist copilot

AI that helps a person handle a conversation. It can bring up relevant information and draft a reply for the person to review and send.

Related: A copilot that drafts replies

Autonomous resolution

An AI agent resolving an issue without human review of each step. It still needs reliable sources, defined permissions, and a record of its actions.

Related: What autonomous resolution requires

AI-to-human handoff

Passing a conversation from AI to a person, with the history, available records, and unresolved request. The next person can continue from the work already done.

Related: Hand off with context

Quality & QA

A shared standard.

The terms behind conversation reviews and useful coaching.

Conversation QA

Reviewing customer conversations against a rubric. Scores show how the reply met each criterion, and coaching identifies a specific way to improve.

Related: How BearScope scores conversations

100% scoring vs. sampling

Full scoring reviews every conversation in scope. Sampling reviews a selected subset. The two approaches offer different coverage and review workloads.

Related: 100% scoring vs. sampling

QA calibration

Aligning reviewers on how to apply the rubric. The team scores the same conversations, discusses differences, and refines the criteria.

Related: Running QA calibration sessions

Reopen rate

The share of resolved tickets that reopen. It can help you understand whether a resolution lasted, especially alongside first-contact resolution and satisfaction.

Related: Reopen rate as a quality metric

Metrics

Get to know your numbers.

Different measures answer different questions about the customer experience.

CSAT Customer Satisfaction

A survey of how satisfied a customer was with an interaction. Read it alongside conversation reviews, and remember that it reflects the customers who respond.

Related: CSAT vs. NPS vs. CES

NPS Net Promoter Score

A survey asking how likely customers are to recommend a brand, from 0 to 10. Subtract the percentage of detractors (0–6) from promoters (9–10) to get a score from −100 to 100.

Related: CSAT vs. NPS vs. CES

CES Customer Effort Score

A measure of how easy or difficult a customer found an interaction. It helps identify unnecessary steps, repeated questions, or other sources of effort.

Related: Reduce customer effort

FCR First Contact Resolution

The share of issues resolved during the first contact. Check how resolution is defined and read it alongside reopened tickets.

Related: First contact resolution, explained

AHT Average Handle Time

The average time spent handling a contact, including the conversation and follow-up work. Use it with quality measures so speed doesn’t become the only goal.

Related: The tradeoffs of average handle time

SLA Service Level Agreement

An agreed response or resolution target, such as replying within one hour. Targets can vary by channel, service, or plan.

Related: Setting achievable SLAs

Cost to serve

The cost of handling customer work, including people, tools, and overhead. Divide by the conversations handled to calculate an average cost per conversation.

Related: Cost to serve per conversation

Trust & safety

Control and confidence.

How AI uses information, follows permissions, and records its work.

Receipt the audit record of an action

A record of an action, including what was attempted, who authorized it, and the result. It lets your team review the work afterward.

Related: Reading an AI receipt

Guardrails

The rules that define what an AI agent can do, what needs approval, and when it must hand over. These rules are checked before actions run.

Related: Guardrails checklist

Grounding

Using source information to support an AI answer. A grounded response can point to the order, conversation, or policy behind an important detail.

Related: Grounding AI support in real data

Fail-closed

A control that blocks an action when the required checks cannot pass. The action stays stopped until its permissions or other requirements are satisfied.

Related: Why AI support should fail closed

CX operations

The work of running customer support: conversations, people, AI, quality reviews, and service improvement. BearScope brings these into one workspace.

Related: What is CX operations?

Book a demo

Get to know BearScope.

Bring your questions to a 30-minute demo. We’ll explore the product and what fits your team.