Claim
Revenue becomes predictable only when prospecting (SDR), inbound qualification (MRR), and closing (AE) are split into three specialized roles. The "do-everything" rep model creates boom-bust pipelines, mismatches skills, and makes diagnosis impossible.
Mechanism
Three failure modes stack on top of each other when one rep does everything:
1. Time allocation collapses — closers stop prospecting while in deals, then face empty pipelines and restart from zero.
2. Skills are different — prospecting rewards persistence, volume, and pattern recognition; closing rewards consultative depth and stakeholder management. A rep can rarely be elite at both.
3. Metrics break — when one person owns the full funnel you cannot tell which stage is broken, so coaching and forecasting become guesswork.
The three-role split (SDR generates qualified meetings → MRR qualifies inbounds → AE closes) gives each function a dedicated metric (qualified opps / response time / revenue closed) and unlocks the pipeline formula: SDR count × qualified opps per SDR per month × deal size × win rate = revenue.
Conditions
Holds when:
- Company is past 5-10 reps and beginning to scale.
- Average deal size is large enough to fund SDR + AE specialization (typically ≥$10k ACV).
Fails when:
- Founder-led sales pre-PMF — premature specialization removes the founder's learning loop.
- Very small/transactional deals where one rep can run the full motion in minutes.
Evidence
"Mixing prospecting and closing in a single role creates unpredictable revenue... When one person does everything, you cannot identify which stage of the pipeline is broken."
"SDRs report to a Sales Development Manager, not AE managers. Prevents AEs from using SDRs as personal assistants."
— Aaron Ross, Predictable Revenue: 15-Minute Summary
Signals
- SDRs report into a Sales Development Manager, not into AE managers.
- SDRs are compensated on qualified opportunities, AEs on revenue — separate plans, not blended.
- Pipeline meetings can name which stage (prospecting / qualification / closing) is the bottleneck.
Counter-evidence
Pete Kazanjy and others argue early-stage startups should keep founders and first sellers running the full funnel until ~$1-2M ARR — the learning from doing every step compounds faster than the productivity gain from specialization. Bottoms-up SaaS (Figma, Notion) skips the SDR/AE split entirely.
Cross-references
- ins_seeds-nets-spears-lead-portfolio — same operator, complementary lens