菜单
Use this skill when configuring CRM workflows, managing sales pipelines, building forecasting models, or optimizing CRM data hygiene. Triggers on Salesforce, HubSpot, CRM workflows, pipeline management, deal stages, forecasting, CRM automation, and any task requiring CRM architecture or process optimization.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/mkurman/zorai --skill crm-management命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| name | crm-management |
| version | 0.1.0 |
| description | Use this skill when configuring CRM workflows, managing sales pipelines, building forecasting models, or optimizing CRM data hygiene. Triggers on Salesforce, HubSpot, CRM workflows, pipeline management, deal stages, forecasting, CRM automation, and any task requiring CRM architecture or process optimization. |
| tags | ["crm","salesforce","hubspot","pipeline","forecasting","automation","workflow","sales","performance","finance"] |
| category | sales |
| recommended_skills | ["sales-playbook","lead-scoring","account-management","sales-enablement"] |
| platforms | ["claude-code","gemini-cli","openai-codex"] |
| license | MIT |
| maintainers | [{"github":"maddhruv"}] |
Data hygiene is non-negotiable - A CRM full of stale, duplicated, or manually-entered guesswork is worse than no CRM. Garbage in, garbage out applies to forecasts, reports, and automation. Treat data quality as a first-class engineering concern: define ownership, set decay rules, and automate enrichment from day one.
Automate the boring stuff - Reps should spend time selling, not updating fields. Any task that follows a predictable rule (create follow-up task when stage advances, notify manager when deal exceeds threshold, enrich lead on creation) should be automated. Human judgment is reserved for exceptions.
Pipeline reflects reality - Every stage must represent a verifiable buyer action, not a rep's optimism. Stages without exit criteria are opinions. Exit criteria must be objective and observable: "Demo completed" not "Rep thinks they're interested." Review pipeline stages whenever win rates diverge from forecast accuracy.
Forecast with methodology - Never let reps enter a single probability number. Pick one forecasting method (weighted, categorical, or AI) and apply it consistently. Mix methods only at the rollup layer. A forecast is only as good as the pipeline data behind it - fix pipeline hygiene before blaming the model.
Less fields, more adoption - Every field added to a record is friction. Every required field that reps don't understand is a source of garbage data. Audit fields quarterly: if a field hasn't been used in reporting in 90 days, archive it. Default to fewer, well-defined fields with validation rules over many optional ones nobody fills in.
CRM platforms organize data around a standard object hierarchy. Understanding the relationships prevents misdesign.
| Object | Represents | Key relationships |
|---|---|---|
| Lead | An unqualified inbound contact, not yet associated to an account | Converts to Contact + Account + Opportunity |
| Contact | A known individual at a company | Belongs to Account; linked to Opportunities |
| Account | A company or organization | Parent of Contacts and Opportunities |
| Opportunity | A specific deal or revenue event in progress | Belongs to Account; has a Stage, Amount, and Close Date |
Lead vs Contact: Leads are pre-qualification. Once a lead meets your ICP criteria (or a sales rep accepts it), convert it. Do not store active selling conversations on Lead records - move to Opportunity.
Account hierarchy: Enterprise deals often span subsidiaries. Model parent-child account relationships to roll up ARR accurately.
A pipeline stage is a milestone in the buyer's journey, not the seller's activity. Each stage must have:
Standard properties every opportunity should carry:
| Property | Type | Purpose |
|---|---|---|
amount | Currency | ACV or total contract value |
close_date | Date | Expected close, used in forecasting |
stage | Enum | Current pipeline stage |
forecast_category | Enum | Committed / Best Case / Pipeline / Omitted |
deal_source | Enum | Inbound / Outbound / Channel / Expansion |
next_step | Text | Single next action with owner and date |
competitor | Multi-select | Competitors actively in the deal |
loss_reason | Enum | Required on Closed Lost; drives win/loss analysis |
CRM workflows are event-driven. Standard trigger types:
Define stages bottom-up: start from Closed Won and work backward to the first meaningful buyer commitment. A typical B2B SaaS pipeline:
| Stage | Definition | Exit criteria | Default probability |
|---|---|---|---|
| Prospecting | Identified as target, no contact yet | Meeting booked | 5% |
| Discovery | First meeting held; pain and budget being explored | Discovery call completed, MEDDIC/BANT fields populated | 15% |
| Demo / Evaluation | Product demonstrated; evaluating fit | Demo completed; champion identified | 30% |
| Proposal | Pricing and scope sent | Verbal interest in proposal | 50% |
| Negotiation | Legal or commercial back-and-forth | Legal review initiated | 70% |
| Closed Won | Contract signed | Signed document received | 100% |
| Closed Lost | Deal dead | Loss reason entered | 0% |
More than 7 active stages is almost always too many. Stages that reps skip consistently signal the stage does not reflect a real buyer milestone.
For SaaS, enterprise, and PLG templates, see references/pipeline-templates.md.
Lead scoring combines demographic fit (ICP match) and behavioral engagement. Use two dimensions to avoid conflating them:
Profile score (ICP fit):
Engagement score (interest signals):
Routing rule: Route to sales when profile score >= 40 AND engagement score >= 30. Never route on engagement alone - a curious student visiting your pricing page is not an MQL.
Choose one primary methodology. Do not mix until you understand the trade-offs.
Weighted pipeline (default):
Categorical (commit-based):
AI / predictive:
Rollup structure: Rep -> Manager -> VP -> CRO. Each level reviews the layer below before submitting up. Lock forecasts weekly on Monday; review actuals Friday.
Automate repetitive mechanics, not judgment calls. Standard automation patterns:
| Trigger | Action | Purpose |
|---|---|---|
| Opportunity stage = Demo | Create task: "Send follow-up email within 24h" assigned to owner | Enforces follow-through |
| Opportunity stage = Proposal | Notify manager via Slack | Deal visibility |
| Opportunity amount > $50k | Flag as "Strategic Deal", notify VP | Escalation routing |
| Close date passes with stage not Closed | Send stale deal alert to rep and manager | Pipeline hygiene |
| Lead created from website form | Enrich via Clearbit/Apollo, route by territory | Speed to lead |
| Deal moves to Closed Lost | Require loss_reason before save | Win/loss data integrity |
Automation should enforce process, not replace it. If an automation creates a task that reps always dismiss, the process is wrong, not the automation.
Data hygiene has four levers: deduplication, enrichment, decay management, and field governance.
Deduplication:
Enrichment:
Decay management:
Field governance:
Every sales dashboard should answer one of three questions: Where are we? Where are we going? Why did deals win or lose?
| Dashboard | Key metrics |
|---|---|
| Pipeline health | Open pipeline by stage, pipeline coverage ratio (pipeline / quota), average deal age per stage |
| Forecast | Committed vs Best Case vs quota, forecast vs prior week delta, at-risk deals (close date < 14 days, no activity in 7 days) |
| Activity | Calls, emails, meetings per rep per week; stage conversion rates |
| Win/loss analysis | Win rate by deal source, competitor, deal size, industry; average sales cycle by segment |
| Rep performance | Quota attainment, pipeline created, average deal size, stage conversion funnel |
Report cadences: Daily - pipeline alerts. Weekly - forecast review. Monthly - win/loss and funnel analysis. Quarterly - field governance and process audit.
CRM-MAP integration is a bidirectional sync. Design the data contract carefully:
CRM to MAP:
MAP to CRM:
Sync rules:
| Anti-pattern | Why it's wrong | What to do instead |
|---|---|---|
| Stages based on rep activity ("Proposal Sent") | Tracks what the seller did, not what the buyer decided | Redefine stages around verifiable buyer actions and decisions |
| Single probability field reps fill manually | Reps game it to match their gut; forecasts become meaningless | Derive probability from stage; use forecast category for rep judgment |
| Required fields without picklists | Reps type anything to get past validation; data is unqueryable | Replace free-text required fields with controlled picklists |
| CRM fields duplicated in spreadsheets | Shadow systems diverge; actual data is always "in the spreadsheet" | Mandate CRM as system of record; kill the spreadsheets |
| Automating before stages are stable | Automation bakes in bad process; expensive to unwind | Freeze stage definitions for one full quarter before automating |
| Enrichment overwriting sales data | Reps lose trust in CRM when their updates get overwritten | Set enrichment to fill empty fields only; never overwrite |
Automating before stage definitions are stable - Building workflow automations on top of pipeline stages that are still being debated bakes bad process into code. When stages change, you have to unwind automations, field mappings, and reports simultaneously. Freeze stage definitions for one full quarter before automating them.
Enrichment overwriting sales rep data - When a data enrichment provider (Clearbit, ZoomInfo) updates a field like company size or industry, it can silently overwrite a value a rep manually entered from a real sales conversation. Reps notice, stop trusting the CRM, and revert to spreadsheets. Configure enrichment to fill empty fields only, never overwrite populated ones.
Lead routing on engagement score alone - A high engagement score means someone is interested - not that they are a qualified buyer. Routing a university student who visits your pricing page 10 times to sales wastes rep time and trains reps to distrust MQL routing. Always require a minimum profile (ICP fit) score alongside engagement before routing.
Forecast categories without commit culture - A categorical forecast ("Committed / Best Case / Pipeline") only works if reps treat "Committed" as a hard promise. Without explicit commit culture training and consequences for consistent miss-commits, reps either over-commit to look good or under-commit to sandbag. The methodology is useless without the discipline.
Required free-text fields - Making a free-text field required (like "Next Steps" as a text box) guarantees garbage data. Reps type anything to save the record: "TBD", "follow up", or nothing meaningful. Replace free-text required fields with picklists that have clear, actionable options.
For detailed templates and implementation guidance, read the relevant file from
the references/ folder:
references/pipeline-templates.md - Pipeline stage templates for SaaS, enterprise, and PLG motionsOnly load a references file if the current task requires it - they are detailed and will consume context.
On first activation of this skill in a conversation: check which companion skills are installed by running
ls ~/.claude/skills/ ~/.agent/skills/ ~/.agents/skills/ .claude/skills/ .agent/skills/ .agents/skills/ 2>/dev/null. Compare the results against therecommended_skillsfield in this file's frontmatter. For any that are missing, mention them once and offer to install:npx skills add AbsolutelySkilled/AbsolutelySkilled --skill <name>Skip entirely if
recommended_skillsis empty or all companions are already installed.