Back to Details

sales-callminer

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

CallMiner Platform Help

CallMiner平台帮助

Help the user with CallMiner Eureka platform questions — from interaction capture and automated QA through agent coaching, real-time alerts, compliance monitoring, and API integration.
为用户解答CallMiner Eureka平台相关问题——从交互捕获、自动化QA到Agent辅导、实时告警、合规监控及API集成。

Step 1 — Gather context

步骤1 — 收集上下文

If 
references/learnings.md
exists, read it first for accumulated platform knowledge.
Ask the user:
  1. What area of CallMiner do you need help with?
    • A) Capture — recording, screen capture, redaction setup
    • B) Analyze — categories, scoring, sentiment, topic extraction
    • C) Coach — agent scorecards, coaching plans, performance tracking
    • D) RealTime — live alerts, next-best-action, real-time guidance
    • E) Compliance — PCI/HIPAA monitoring, PII redaction, risk flagging
    • F) API — Ingestion API (audio/text in), Data API (insights out)
    • G) Integrations — CRM, CCaaS, WFM, transcription engine
    • H) OmniAgent / Outreach / LiveTranslate — automation modules
    • I) Admin — licensing, user management, permissions
    • J) Something else
  2. What's your role?
    • A) QA Manager / QA Analyst
    • B) Contact center manager / supervisor
    • C) Compliance / risk officer
    • D) CX / VoC analyst
    • E) IT / admin / developer
    • F) Other
  3. What are you trying to accomplish? (describe your specific goal or issue)
If the user's request already provides most of this context, skip directly to the relevant step. Lead with your best-effort answer using reasonable assumptions (stated explicitly), then ask only the most critical 1-2 clarifying questions at the end.
若
references/learnings.md
文件存在，请先阅读以获取积累的平台知识。
询问用户：
  1. 你需要CallMiner哪个领域的帮助？
    • A) Capture — 录音、屏幕捕获、脱敏设置
    • B) Analyze — 分类、评分、情感分析、主题提取
    • C) Coach — Agent评分卡、辅导计划、绩效跟踪
    • D) RealTime — 实时告警、最佳下一步行动、实时指导
    • E) Compliance — PCI/HIPAA监控、PII脱敏、风险标记
    • F) API — Ingestion API（音频/文本导入）、Data API（洞察导出）
    • G) Integrations — CRM、CCaaS、WFM、转录引擎
    • H) OmniAgent / Outreach / LiveTranslate — 自动化模块
    • I) Admin — 许可、用户管理、权限设置
    • J) 其他
  2. 你的角色是什么？
    • A) QA经理/QA分析师
    • B) 联络中心经理/主管
    • C) 合规/风险专员
    • D) CX/VoC分析师
    • E) IT/管理员/开发人员
    • F) 其他
  3. 你想要达成什么目标？（描述你的具体目标或问题）
若用户的请求已提供大部分此类上下文，请直接跳至相关步骤。 先基于合理假设（需明确说明）给出最佳答案，最后仅询问最关键的1-2个澄清问题。

Step 2 — Route or answer directly

步骤2 — 转介或直接解答

Problem domainRoute to
CCaaS platform comparison/selection
/sales-ccaas-selection {question}
Agent coaching strategy (not CallMiner-specific)
/sales-coaching {question}
Customer feedback / NPS / CSAT strategy
/sales-customer-feedback {question}
Connecting CallMiner to other tools (architecture)
/sales-integration {question}
Otherwise, answer directly from platform knowledge using the reference below.
问题领域转至
CCaaS平台对比/选型
/sales-ccaas-selection {question}
Agent辅导策略（非CallMiner专属）
/sales-coaching {question}
客户反馈/NPS/CSAT策略
/sales-customer-feedback {question}
CallMiner与其他工具的连接（架构）
/sales-integration {question}
否则，直接利用下方参考资料中的平台知识解答。

Step 3 — CallMiner platform reference

步骤3 — CallMiner平台参考

Read 
references/platform-guide.md
 for the full platform reference — modules, pricing, integrations, data model, API overview, workflows.
Answer the user's question using only the relevant section. Don't dump the full reference.
阅读
references/platform-guide.md
 获取完整平台参考——模块、定价、集成、数据模型、API概述、工作流。
仅使用相关部分解答用户问题，不要输出完整参考内容。

Step 4 — Actionable guidance

步骤4 — 可操作指导

You no longer need the platform guide — focus on the user's specific situation.
  1. Step-by-step instructions for their goal in CallMiner
  2. Configuration recommendations — specific settings, category rules, scoring criteria
  3. Common pitfalls — what goes wrong and how to avoid it
  4. Verification — how to confirm the change worked
  5. For API questions — point to 
    references/callminer-api-reference.md
If you discover a gotcha, workaround, or tip not covered in 
references/learnings.md
, append it there.
无需再依赖平台指南——聚焦用户的具体场景。
  1. 分步说明：在CallMiner中实现目标的具体步骤
  2. 配置建议——特定设置、分类规则、评分标准
  3. 常见陷阱——易出现的问题及规避方法
  4. 验证方式——如何确认更改生效
  5. API相关问题——指向
    references/callminer-api-reference.md
若发现
references/learnings.md
未涵盖的陷阱、变通方案或技巧，请将其添加至该文件。

Gotchas

注意事项

Best-effort from research — review these, especially items about plan-gated features and integration gotchas that may be outdated.
  • Expect a 3-6 month ramp-up. CallMiner is powerful but complex — building accurate categories, scoring models, and coaching workflows takes months. Don't expect instant ROI. Budget for dedicated analyst time during setup.
  • Auto-logout after 30 minutes is by design. The session timeout is short and affects all open tabs simultaneously. Save work frequently. This is a known pain point with no user-configurable override.
  • Category tuning is iterative, not one-shot. Initial category rules will have false positives/negatives. Plan for weekly refinement cycles during the first quarter. Test categories against known-good interactions before deploying to scoring.
  • Transcription engine choice affects everything downstream. CallMiner supports multiple ASR engines (Deepgram, Google, Azure, Nuance) via OVTS. Accuracy varies by accent, industry jargon, and audio quality. Test multiple engines with your actual call recordings before committing.
  • Pricing is opaque — negotiate hard. No public pricing. Average ~$102K/year but ranges widely. Seat-based vs. hours-analyzed licensing models have very different economics depending on your volume and agent count.
  • Self-improving: If you discover something not covered here, append it to 
    references/learnings.md
    with today's date.
基于研究的最佳实践——请仔细查看，尤其是关于计划受限功能和可能过时的集成陷阱的内容。
  • 预计需要3-6个月的上手周期：CallMiner功能强大但复杂度高——构建准确的分类、评分模型和辅导工作流需要数月时间。不要期望即时投资回报。在设置阶段需为专职分析师预留时间预算。
  • 30分钟后自动登出是设计使然：会话超时时间较短，且会同时影响所有打开的标签页。请频繁保存工作。这是已知痛点，暂无用户可配置的覆盖选项。
  • 分类调优是迭代过程，而非一次性操作：初始分类规则会存在误报/漏报情况。计划在第一季度每周进行优化。在部署到评分系统前，需针对已知有效交互测试分类。
  • 转录引擎的选择会影响后续所有环节：CallMiner通过OVTS支持多种ASR引擎（Deepgram、Google、Azure、Nuance）。识别准确率因口音、行业术语和音频质量而异。在确定前，请使用实际通话录音测试多个引擎。
  • 定价不透明——务必全力协商：无公开定价。平均约10.2万美元/年，但差异较大。基于席位和基于分析时长的许可模式，根据你的业务量和Agent数量，经济成本差异显著。
  • 自我优化：若发现此处未涵盖的内容，请将其添加至
    references/learnings.md
    并标注日期。

Related skills

相关技能

  • /sales-coaching
    — Sales coaching, QA, and agent training strategy (platform-agnostic)
  • /sales-ccaas-selection
    — Compare CCaaS platforms (Genesys, NICE, Talkdesk, Five9, etc.)
  • /sales-observe-ai
    — Observe.AI — contact center AI for QA and agent assist (CallMiner alternative)
  • /sales-cresta
    — Cresta — enterprise contact center AI (CallMiner alternative)
  • /sales-balto
    — Balto — real-time agent assist (CallMiner RealTime alternative)
  • /sales-convin
    — Convin — conversation intelligence with auto QA
  • /sales-enthu
    — Enthu.AI — contact center conversation intelligence
  • /sales-customer-feedback
    — Customer feedback, NPS, CSAT, VoC strategy
  • /sales-integration
    — Connect CallMiner to CRM, CCaaS, or other tools
  • /sales-do
    — Not sure which skill to use? The router matches any sales objective to the right skill. Install: 
    npx skills add sales-skills/sales --skill sales-do -a claude-code -y
  • /sales-coaching
    — 销售辅导、QA及Agent培训策略（平台无关）
  • /sales-ccaas-selection
    — 对比CCaaS平台（Genesys、NICE、Talkdesk、Five9等）
  • /sales-observe-ai
    — Observe.AI——用于QA和Agent辅助的联络中心AI（CallMiner竞品）
  • /sales-cresta
    — Cresta——企业级联络中心AI（CallMiner竞品）
  • /sales-balto
    — Balto——实时Agent辅助（CallMiner RealTime竞品）
  • /sales-convin
    — Convin——带自动QA的对话智能工具
  • /sales-enthu
    — Enthu.AI——联络中心对话智能工具
  • /sales-customer-feedback
    — 客户反馈、NPS、CSAT、VoC策略
  • /sales-integration
    — 将CallMiner连接至CRM、CCaaS或其他工具
  • /sales-do
    — 不确定使用哪个技能？该路由可将任何销售目标匹配至合适技能。安装命令：
    npx skills add sales-skills/sales --skill sales-do -a claude-code -y

Examples

示例

Example 1: Automated QA scoring setup

示例1：自动化QA评分设置

User says: "We're only scoring 2% of calls manually — how do I set up automated QA in CallMiner?" Skill does:
  1. Reads platform guide for Analyze and Coach modules
  2. Explains category-based scoring — define quality criteria as categories, assign weights, auto-score 100% of interactions
  3. Walks through creating a QA scorecard with weighted categories (compliance, empathy, resolution, script adherence)
  4. Recommends starting with a calibration period comparing auto-scores to manual scores Result: User has a plan to move from 2% manual sampling to 100% automated QA scoring
用户提问：“我们仅手动评分2%的通话——如何在CallMiner中设置自动化QA？” 技能操作：
  1. 查阅Analyze和Coach模块的平台指南
  2. 解释基于分类的评分——将质量标准定义为分类，分配权重，自动评分100%的交互
  3. 逐步演示创建带权重分类（合规、同理心、问题解决、脚本遵循）的QA评分卡
  4. 建议先设置校准周期，对比自动评分与手动评分 结果：用户获得从2%手动抽样转向100%自动化QA评分的方案

Example 2: Compliance monitoring

示例2：合规监控

User says: "We need to monitor all calls for PCI compliance — agents sometimes read back full card numbers" Skill does:
  1. Reads platform guide for compliance and Redact modules
  2. Explains how to create categories detecting PCI-sensitive language patterns
  3. Walks through Redact configuration for automatic PII masking in transcripts and audio
  4. Recommends RealTime alerts for live intervention when PCI violations detected Result: User has PCI compliance monitoring with auto-redaction and real-time agent alerts
用户提问：“我们需要监控所有通话是否符合PCI合规——Agent有时会完整读出卡号” 技能操作：
  1. 查阅合规和Redact模块的平台指南
  2. 解释如何创建检测PCI敏感语言模式的分类
  3. 逐步演示Redact配置，实现转录文本和音频中的PII自动脱敏
  4. 建议使用RealTime告警，在PCI违规时实时干预 结果：用户获得带自动脱敏和实时Agent告警的PCI合规监控方案

Troubleshooting

故障排除

Steep learning curve — don't know where to start

学习曲线陡峭——不知从何入手

Symptom: Platform feels overwhelming, too many modules and configuration options Cause: CallMiner is enterprise-grade — it's designed for dedicated analysts, not casual users Solution: Start with ONE use case (e.g., automated QA on a single queue). Build 5-10 categories, validate against known interactions, then expand. CallMiner offers a sandbox environment for testing. Budget 3-6 months for the first production deployment. Request CallMiner's onboarding support — initial hand-holding is available.
症状：平台感觉过于复杂，模块和配置选项过多 原因：CallMiner是企业级工具——专为专职分析师设计，而非普通用户 解决方案：从单一用例开始（例如，对单个队列进行自动化QA）。创建5-10个分类，针对已知交互验证后再扩展。CallMiner提供沙箱环境用于测试。为首次生产部署预留3-6个月时间。请求CallMiner的入职支持——初始指导服务可用。

Categories not matching expected interactions

分类未匹配预期交互

Symptom: Auto-scoring misses interactions that should match, or flags ones that shouldn't Cause: Category rules are too narrow (missing synonyms/variations) or too broad (matching unrelated phrases) Solution: Review the category's keyword/phrase rules. Use CallMiner's "Test" feature to run categories against a sample set. Add phrase variations and exclusions iteratively. Check transcription accuracy first — poor ASR = poor category matching regardless of rule quality.
症状：自动评分遗漏应匹配的交互，或标记不应匹配的交互 原因：分类规则过于狭窄（缺少同义词/变体）或过于宽泛（匹配无关短语） 解决方案：查看分类的关键词/短语规则。使用CallMiner的“测试”功能在样本集上运行分类。迭代添加短语变体和排除项。首先检查转录准确率——ASR识别差会导致无论规则质量如何，分类匹配效果都差。

Pages timing out / session drops

页面超时/会话中断

Symptom: UI tabs time out and log you out across all open pages simultaneously Cause: 30-minute session timeout is system-wide; one tab expiring kills all sessions Solution: Save work frequently. Avoid having many tabs open simultaneously. If running long analysis queries, use the API (Data API) for batch exports instead of the UI. This is a known limitation — no workaround for extending the timeout.
症状：UI标签页超时，并同时登出所有打开的页面 原因：30分钟会话超时为系统全局设置；一个标签页过期会终止所有会话 解决方案：频繁保存工作。避免同时打开多个标签页。若运行长时间分析查询，请使用API（Data API）进行批量导出，而非UI。这是已知限制——暂无延长超时时间的变通方案。