The Playbook

A Prescription, Not a Forecast

What follows is the report’s editorial recommendation for a company that accepts the diagnosis and intends to act. The phase durations, target percentages, and discount levels are design choices, reasonable starting points drawn from the levers the data validates, not predictions of what any specific company will achieve. The findings underneath them are certified, multi-year contracts are associated with roughly 79% lower churn, contract adoption fell from 48% to 26% in a single year, gross churn has held near 14%, downsell is about a third of all revenue loss, and the coverage ratio sat at 1.07× in 2024, but the program built on those findings is a prescription to be adapted, measured, and revised, not a plan whose outputs are guaranteed. Inference

Readiness: Who Can Make the Change

Not every organization can make this change at the same moment, and attempting the full redesign without the prerequisites risks the worst outcome, an initiative that launches, meets resistance, produces conflict, and is abandoned, leaving a leadership team more cynical and less likely to try again. Six indicators identify the organizations with the conditions for it to succeed. Inference

The Prerequisite Diagnostic

No redesign should begin without an honest read of the current state. The diagnostic has two parts: a quantitative read that establishes what the system is doing, and a qualitative read that maps its leaders against the readiness indicators above. Inference

Run the Revenue System Diagnostic to produce this read for your own company. It takes 12 to 15 minutes with your numbers at hand, and a number you cannot produce is itself the first finding.

Quantitative: five data points

Produce all five. Each is an input the phases that follow depend on:

1. Decomposed ARR movement. Beginning ARR, expansion ARR, gross churn ARR, and downsell ARR for the most recent four quarters, reported separately, not combined into NDR.
2. Contract-length mix. The share of ARR on sub-annual, annual, two-year, and three-year-plus terms.
3. Decomposed revenue responsibility. Which function, Sales or Customer Success, is accountable for each ARR line (new-logo, expansion, renewal, downsell, and churn), and the resulting ARR split between the two.
4. Professional-services attach rate (if applicable). Services investment as a share of new-customer contract value across the past 12 months.
5. Methodology and Enablement investment. What was spent, per GTM team, on training, certification, and outcome-delivery methodology as a distinct budget line, not headcount, not tools.

Qualitative: three conversations

1. The CEO. Has the mediation cycle been a recurring pattern? How often? What is the CEO’s current explanation for why it repeats?
2. The CFO. Has NDR ever been presented to the board in decomposed form? Does Finance have a model connecting contract-length decisions to churn outcomes?
3. The CS and Sales leaders. Asked whose fault the NDR performance is, do they point at each other or at the system? The answer shows whether the defensive routine is entrenched, or whether at least one leader is prepared to make the system argument.

Phase 1: Build What the System Can See (Weeks 1–4)

The reporting redesign comes first for two reasons. It is the lowest-friction change available (no restructuring, no new headcount, no compensation change), and it creates the information environment that makes every subsequent intervention easier to justify. Inference

Board reporting redesign

Beginning with the next board cycle, NDR is presented decomposed: expansion, gross churn, and downsell as separate lines, each as a percentage of beginning ARR, with three input metrics alongside:

• the contract-length mix distribution;
• the professional-services attach rate, tracked against the 5–15%-of-ARR investment band;
• methodology and enablement investment as a distinct budget line.

The first time that slide appears, it generates questions the leadership team has not been asked before, and the questions are the objective. “Why is downsell that large a share of our loss?” and “why are three-quarters of our accounts on annual terms when multi-year contracts are associated with roughly 79% lower churn?” are the conversations that justify everything after.

One condition determines whether that slide ever gets built, and it deserves to be stated as policy rather than left to courage. A CFO who decomposes NDR for the first time is voluntarily creating visibility into dynamics no one has been accountable for, in a culture where surfacing problems can read as confessing them; the rational response to that exposure, as the report’s systems chapter documents, is not to look. So the redesign begins with an explicit amnesty: the new reporting is dated from the day the instrument changed, prior periods are not restated, and nothing surfaced by the decomposition is treated as a finding against any function or any prior decision. The first slide reports what the previous instrument was built to net away, which is a statement about the instrument, not about the people who used it. Said once by the CEO, in those terms, this is what makes honest numbers a safe thing to produce, and it costs nothing. Inference

The retention-economics model

In parallel, Finance builds the model on the company’s own ARR, average ACV, gross margin, and churn performance. It produces three numbers: the annual churn tax on sales-and-marketing spend, the return on each available retention lever at current churn levels, and the enterprise-value impact of a four-point NDR improvement. That model becomes the investment case for Phases 2 and 3.

Phase 2: Build What the System Values (Weeks 5–12)

With visibility established and the investment case built, Phase 2 moves capital to the levers the data validates and begins the outcome record. Four workstreams run in parallel. Inference

The contract-length program

Finance approves multi-year incentive pricing (on the order of 15% for two-year terms and 20% for three-year), and Sales frames multi-year commitment on what the customer gets, predictable pricing and guaranteed capacity, not on the size of the discount. The working target: shift 30% of new ARR to two-year-plus terms within 12 months. The expected effect, a blended gross-churn reduction of two to four points as the first multi-year cohort reaches renewal, is a design hypothesis. It rests on the certified differential (multi-year terms are associated with roughly 79% lower churn), not a forecast.Priced against the alternative, the discount is in most cases the small side of the trade: it costs points of margin, while a churn event forfeits the entire 37-month CAC payback.

The program is priced against a deliberately discounted effect, not the headline gradient. The certified 79% differential blends selection, renewal-window deferral, and the true effect of commitment, and no cross-sectional survey can separate the three, so the stress test assumes the worst defensible case: only a third of the differential is real treatment effect. The arithmetic, per hundred dollars of annual contract value: a 15% two-year discount costs 15 points a year. At the full gradient, moving an account from annual-term churn (roughly 14%) to multi-year churn (roughly 3%) avoids 11 points of annual loss, and at the survey’s 37-month new-only payback, every avoided point also avoids roughly two and a half points of the sales-and-marketing spend that replacing it would consume. Full effect: about 27 points of avoided replacement cost a year against 15 of discount, before counting the retained margin itself. The trade clears easily. At one-third effect, roughly 9 points of avoided replacement against the same 15 of discount: the naive case fails on replacement cost alone, and survives only on the retained account’s own margin, its later expansion, and the prepay working capital the contracts chapter prices. The stress test does not kill the program; it prices the discount. A company that believes selection carries most of the gradient should open at 10% for two-year terms rather than 15%, and let the first cohort’s renewal data set the ceiling. Calculated

One structural condition travels with the program: the multi-year instrument is a committed term around a value metric that survives AI, not a three-year lock on per-seat pricing. A seat-priced multi-year deal defers every seat-compression conversation to a single renewal where the customer arrives holding three years of accumulated leverage. Structure the term as a committed base with a usage or outcome component inside it, per Term and Metric, One Decision and The Pricing Hedge.

The professional-services program

Finance sizes services capacity into the 5–15%-of-ARR band where attach lowers churn rather than draining margin, whether through added services headcount, implementation-partner capacity, or restructuring existing services away from custom engagements toward standard onboarding packages. Attach is treated as a capital-allocation decision with a measurable return, not a cost center.

Building the outcome record

Product leads, in collaboration with CS, working from outcomes already documented in the existing base. The record is organized by customer segment, runs from product capability through business impact, and states an implementation threshold and a verification mechanism for each entry. The first version is deliberately narrow: the three to five outcomes most commonly committed in the sales process and cited in renewal conversations. The capability-gap log, which records outcomes currently being committed that the product does not reliably deliver, is built in parallel; without it the record states only what the product produces, not where it falls short.

The retention-signal intake

Product stands up a formal quarterly intake for CS- and Support-sourced retention signals: CS synthesizes capability-gap patterns into a product input, and Product formally acknowledges each gap against its roadmap status. The first cycle surfaces the gaps that have been driving churn invisibly; subsequent cycles keep the roadmap responsive to retention outcomes, not only acquisition pressure (see Product as a System Member).

Phase 3: Build What the System Rewards (Weeks 13–24)

The compensation redesign is the intervention most likely to meet resistance, because it most directly challenges the rational self-interest of functional leaders. It comes last by design: by Week 13 the board reporting has surfaced the composition, the financial model has established the investment case, and the first contract-length and services interventions have produced early evidence that the levers work. By then the redesign follows a pattern the organization can already see, rather than arriving as an untested theory. Inference

Customer success

CS compensation adds two leading indicators alongside the lagging renewal rate: outcome achievement rate (the share of customers with an outcome documented in the record by the midpoint of their contract term) and expansion revenue tied to documented value achievement, not to relationship management or discounting. The operational precondition is strict: the outcome record must exist, with a verification mechanism for each entry, before CS is measured against it. Measuring before enabling is pressure without purpose.

Sales

A portion of commission, typically 10–15%, is tied to 90-day onboarding completion and the six-month rate of achievement on outcomes documented in the record. This ties Sales pay to the loop between what Sales commits and what CS can deliver. A team whose commission is partly contingent on early outcomes has a financial interest in qualifying deals carefully and committing to value the product can deliver.

Product

Product performance measurement incorporates two retention-connected metrics for the first time: the outcome achievement rate for the segments each product area serves, and the capability-gap attribution rate (the share of churned accounts where a documented product gap contributed). This does not make Product solely accountable for NDR; it makes Product accountable for the component it controls. The answer to a CPO who objects: the current system already concentrates accountability in CS for outcomes multiple functions determine. Distributing it across those functions corrects that.

Marketing

The ICP definition is updated with retention and expansion data from the existing base (which segments, from which sources, on which value themes, retain and expand best), and pipeline quality is measured by downstream cohort performance, not conversion alone. Volume accountability stays; a quality dimension is added that aligns Marketing’s optimization with the system’s retention outcomes.

The Steady State

The Orchestrated Revenue System is a permanent operating model, and its steady state has four structural features. Inference

• A single shared record of what the product produces. The outcome record states what the company delivers, for which customer types, under which conditions. It is maintained by Product, contributed to by CS, referenced by Marketing and Sales, and reviewed quarterly as outcomes are documented and the gap log is revised against roadmap progress. Marketing describes from it, Sales commits from it, CS delivers against it, and Finance measures against it.
• A unified NDR accountability framework. The decomposition is reported to the board quarterly with each function’s contribution visible: Marketing’s ICP accuracy in the gross-churn composition, Sales’ contract discipline in the term mix, CS’s outcome delivery in the downsell trend, Finance’s allocation in the attach and methodology lines, Product’s quality in the gap attribution rate. No function can succeed by its own metrics while the system fails, because the system metrics are the board metrics.
• A product–retention feedback loop. CS documents outcome achievement and capability gaps through the record’s verification mechanism; Product acknowledges each gap against roadmap status through the quarterly intake. When the gap log records a capability the product does not reliably deliver, Sales is notified and commits to the documented timeline, not to a feature the product does not have. The gap is recorded before the sale rather than surfaced after it.
• A closing loop across the value chain. Support data flows to CS as a required health input and to Product as a required roadmap signal. Marketing tracks post-close cohort performance by source, segment, and value theme. Sales adjusts deal qualification from post-sale feedback. Product balances acquisition and retention signals in prioritization.

With those four features in place, retention signal feeds back into the decisions that set next year’s churn.

What the whole program requires of the CEO and CFO is the argument of The Imperative, which applies here with full force: the work can be designed with outside expertise, but it has to be owned internally. For the operator’s short list of first moves, see Top Takeaways.