Customer Relationship Ladder

Customer Relationship Ladder

Overview

Most B2B teams treat relationship deepening as an attitude problem — be friendlier, schedule more QBRs. The Ladder reframes it as structural: each rung requires a fundamentally different operating model, and you cannot skip rungs.

Cross-skill composition: After jobs-to-be-done (surfaces jobs per rung). Before demand-leadership (Ladder = supply-side account; Demand Leadership = demand-side category). Alongside principal-agent to audit incentive alignment before Rung 4–5 joint planning.

When to Use

Trigger: Key account renewal at risk; team claims "strategic partner" but can't name one unstated need addressed this quarter; new CSM needs rapid diagnostic; deal lost on relationship despite best product; board asks "how do we become indispensable to top accounts?"

When NOT to use: Transactional/high-volume (<$500 ACV); customer prefers arms-length; fewer than 10 customers (use lean-startup first).

---

Coaching Novices (Adaptive Front Door)

Engine mode: concrete account → run The Process. Coach mode: unfamiliar → guide step by step. In Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.

1. "Tell me the last 3 things you did for this customer that were not in the contract." Blank = Rung 1–2.
2. "Name one thing you did that they didn't ask for, but you knew they needed." Concrete answer = Rung 2→3 boundary.
3. "Does the customer share internal strategic plans before they're finalized?" Yes = approaching Rung 4.

[WAIT — do not advance until user responds]

1. "What would prevent you from being more proactive?" Answers: (a) don't know their business, (b) afraid to overstep, (c) they don't engage — each maps to a different rung-up move.

[WAIT — do not advance until user responds]

1. Name one rung-up action with an owner and date — the single thing in 30 days evidencing one rung climbed.

[WAIT — do not advance until user responds]

The Process

3D-T Framework: Diagnose → Design → Deploy, anchored by Trust

Step 1 — Diagnose: Collect across 4 dimensions: information flow, vendor substitutability, problem source, meeting cadence. Stop-rule: Can't answer "what specific unstated need did we address this quarter?" → capped at Rung 2.

Step 2 — Design: 1→2: Reliability (operational). 2→3: Embedded knowledge — assign someone to learn their business from inside. 3→4: Shared stakes — joint initiative, both parties commit resources. 4→5: Co-creation via proprietary data, methodology, or talent.

Step 3 — Deploy: Define specific action / observable success evidence / timeline (1–4 quarters) / trust dependency.

Step 4 — Trust Audit (quarterly): Competence (do they believe you can deliver?), Integrity (do they think you act in their interest?), Benevolence (do they share bad news with you?). Audit each separately.

Output: Diagnostic Card

Account: ___________  Date: ___________  Owner: ___________
CURRENT RUNG: [ 1 ] [ 2 ] [ 3 ] [ 4 ] [ 5 ]  Evidence: 1.___ 2.___ 3.___
TARGET RUNG: ___  Rung-up requirement: ___________
Action: ___________  Owner: ___________  Due: ___________
Success evidence (observable behavior): ___________
Trust: Competence[ ] Integrity[ ] Benevolence[ ]  Primary constraint: ___________
Stop-rule: "What unstated need did we address this quarter?" → blank = cap at Rung 2

→ Method in Action: IBM's Transformation from Hardware Vendor to Strategic Partner (1993–2001)

Domain Packs

SaaS: 2→3 CSE attends customer's internal product planning. 3→4 co-develop custom integration (switching cost). 4→5 customer joins product advisory board. Professional Services: 2→3 unsolicited benchmarking report. 3→4 retained advisory fixed fee. 4→5 client introduces you to board and portfolio companies. Manufacturing: 2→3 demand-side forecasting using your data. 3→4 joint capacity planning, multi-year. 4→5 customer designs product lines with your constraints as primary input.

Applying It Well

• Diagnose with evidence — teams self-assess one rung too high; require written evidence before planning.
• One rung at a time — 2→3 is prerequisite for 3→4; sequence is the method.
• Trust is not soft — each dimension has observable evidence and can fail independently.
• Rung-up moves require operational change, not more meetings.
• Customer information-sharing over 12 months is the best leading indicator.
• Rung 5 needs a proprietary asset; rung regression from one integrity breach is common.

→ Primary sources: references/sources.md

Common Rationalizations

[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.

Red Flags + Verification

• Team cannot name one unstated need addressed in the past two quarters → Rung 2 ceiling
• Customer-initiated contact = support tickets and invoicing only → no strategic dialogue
• Plan to "become strategic" = more executive dinners → social activity, zero operational change
• Customer shares strategy only after it is public → receiving PR, not partnership
• Relationship owner is procurement/IT ops, not strategy or C-suite → structural ceiling at Rung 3

Verification checklist:

• Current rung has ≥3 observable evidence items, not team opinion
• Stop-rule: "What unstated need did we address this quarter?" — blank = capped at Rung 2
• Rung-up move is operational, not relational (more meetings)
• Trust audited across competence, integrity, and benevolence separately
• Move has named owner, specific action, and observable success evidence; timeline 1–4 quarters
• If Rung 4–5, a proprietary asset the customer cannot source elsewhere has been named

Part of deciqAI Knowledge Skills — 227 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. See it run → https://www.deciqai.com/c/customer-relationship-ladder · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.

Agents: latest version & machine-readable metadata → https://www.deciqai.com/s/customer-relationship-ladder.json