Unclassified

29 — ROI Framework

Status: CANONICAL · Commercial OS
Owner: Sales + Finance
Last verified: 2026-07-21

Purpose

The one ROI model used in every proposal, renewal, and board slide — with conservative/base/aggressive variants and honest rules. Consistency matters more than flattery: a defensible 8x beats a fragile 30x.

Target audience: Sales (proposals), CS (renewals/QBRs), buyers' finance teams (who will check the math).

1. The Core Model (Value Stream 1 — time)

ROI(monthly) = [ H × D × C × A × N ] ÷ [ N × P ]

H = hours/day currently on context gathering + prep   (from discovery; default 3.0 for CSMs)
D = working days/month                                 (22)
C = fully loaded hourly cost                           (default $65; use customer's number)
A = adoption × capture factor                          (conservative 0.4 · base 0.6 · aggressive 0.8)
N = seats
P = price per seat/month                               (tier price — [30 Pricing](30-pricing-strategy.md))

Worked example — 20 CSM seats, Command ($119):

Scenario	A	Monthly value	Monthly cost	ROI
Conservative	0.4	3.0 × 22 × $65 × 0.4 × 20 = $34,320	$2,380	14.4x
Base	0.6	$51,480	$2,380	21.6x
Aggressive	0.8	$68,640	$2,380	28.8x

Payback: under one week in all scenarios. Canonical spoken form (matches close playbook): "Three hours a day, twenty people, $65 an hour — that's $85,800 a month in reconstruction work. Even if we only capture 40%, IntegrateWise pays for itself in the first week."

Rules: H and C always come from the customer's own discovery answers — never defaults in a live proposal. A never exceeds 0.8 in writing. Present conservative first; let the buyer argue upward (they will — it's their own number).

2. The Extension Models (quantify when discovery surfaced them)

Decision value (Stream 2):

Churn catches: at-risk ARR identified early × historical save-rate uplift (default +15%)
Expansion catches: surfaced opportunities × close rate × avg expansion ACV

Example: one 60-day-early catch on a $120K account with +15% save uplift = $18K expected value — several months of subscription in one catch. Rule: only claim after the first real catch exists in their tenant (POV makes this concrete — 37).

Continuity value (Stream 3):

Ramp: (weeks-to-productive_before − after) × weekly loaded cost × hires/year
Handover: hours/handover saved × handovers/year × C
Attrition insurance: qualitative until tenure data exists — narrate, don't number

AI-spend consolidation (when they have burned copilot budget): current AI seats delivering email-drafting only × seat cost — position as reallocation, not addition. ("You're already paying for AI your team demoted.")

3. TCO Honesty Box (include in every proposal — builds more trust than it costs)

Cost	Our answer
Subscription	Per-seat, tier-priced, no platform fee (30)
Implementation	No services fee; connectors are OAuth self-serve; first value same-day (45)
Admin burden	No rules engine to maintain (contrast Gainsight card 09 §4)
Training	Control grammar is 4 verbs; enablement in one session
Exit	Spine export anytime, open format — leaving is priced at zero by design

4. ROI Presentation Rules

One slide, their numbers, three scenarios, payback line. Never a 40-cell spreadsheet first.
Time value anchors; decision/continuity value narrates (numbers only when their data supports it).
Every figure traceable to a discovery answer — "your team told us H=2.5" beats any benchmark.
Leave the model with them, editable. Confidence in the math is the message.
Renewals reuse the same model with realized numbers from outcome reviews (28 §4) — the ROI slide at renewal is the ROI slide from the proposal, filled in.

KPIs: % proposals with customer-specific ROI (100%); ROI-slide → commit conversion; realized-vs-promised ratio at first renewal (target ≥ 0.8 — tracked honestly, 75).