Socio360
Run the scan
BLOG REVOPS

The RevOps Maturity Model: Five Stages

SHORT ANSWER

RevOps maturity has five stages: ad hoc, defined, enforced, predictive, and optimising. Most B2B companies sit at stage two, where definitions exist on paper but nothing in the system enforces them. The move from two to three is the single largest jump in forecast accuracy available.

KEY TAKEAWAYS
  • Assess on observable tests, not on self-description. Almost everyone rates themselves one stage too high.
  • Stage 2 to 3 — from defined to enforced — is where forecast accuracy actually changes.
  • You cannot skip stages. Predictive analytics on unenforced process produces confident nonsense.
  • Most companies stall at stage 2 because enforcement is unpopular and documentation is not.
  • Stage 5 is not a destination — it is a rate of improvement, and it decays without cadence.

Why maturity models usually fail

Most maturity models describe stages in aspirational language — aligned teams, data-driven culture, single source of truth — which lets everyone place themselves comfortably in the middle. A useful model uses tests you can either pass or fail.

Each stage below has one observable test. Run it honestly and you will usually find you are a stage lower than you assumed, which is the point.

The five stages

StageCharacterised byThe test
1. Ad hocProcess lives in people's headsCan a new rep be productive without shadowing someone for a month?
2. DefinedDefinitions written, not enforcedDo two functions report the same metric identically?
3. EnforcedThe system prevents the wrong pathWhat happens in the system if a rep skips a stage?
4. PredictiveThe forecast is trusted and explains itselfCan you explain a forecast miss in under a day?
5. OptimisingContinuous, measured improvementWhat did you retire last quarter?

Stage 1 — Ad hoc

Process exists but only as tribal knowledge. The CRM is a contact list with optional fields. Forecasting is a conversation. Reporting is a spreadsheet someone rebuilds each week.

This is entirely appropriate below roughly $1M ARR. The mistake is not being here — it is staying here past the point where more than one person needs to understand the same thing. Unlock stage 2 by writing definitions down, which costs a week and no money. See RevOps for early-stage startups for what is genuinely worth building at this stage.

Stage 2 — Defined

Definitions exist. There is a document describing stages, an agreed MQL definition, a process everyone was trained on. And the system enforces none of it, so practice drifts from documentation continuously.

This is where most B2B companies sit, and where most stay, because writing definitions is popular and enforcing them is not. Enforcement means telling a rep they cannot advance a deal, which produces immediate friction against a benefit that arrives a quarter later.

Unlock stage 3 with validation rules at stage change, routing enforced in-system with fallbacks, and required fields tied to decisions. The technical work is days. The organisational work is convincing leadership to hold the line for the first month.

Stage 3 — Enforced

The system makes the wrong path impossible rather than describing the right one. Stage exit criteria are checked, routing has a fallback and an SLA alert, and data quality runs as recurring jobs rather than periodic cleanups.

This is the stage at which the forecast starts meaning something, because the stages themselves now mean something. The jump from 2 to 3 produces the largest single improvement in forecast accuracy available to most companies — typically moving from a forecast that misses by 25–40% to one within 10–15%.

Unlock stage 4 by building the reporting layer on the now-trustworthy model, deriving win probability from historical stage conversion rather than rep sentiment, and installing the operating cadence.

Stage 4 — Predictive

The forecast is trusted and, more importantly, explains itself. When the number moves, someone can trace it to a specific stage, segment, and cause within a day rather than reconstructing it manually over a week.

Cohort retention is calculable, cost per qualified opportunity is known by source, and the weekly pipeline review runs on live CRM data rather than an exported deck. Unlock stage 5 by adding the removal discipline: quarterly field and automation audits, renewal-triggered tool reviews, and an annual architecture review.

Stage 5 — Optimising

Not a destination but a rate. The system improves continuously and, crucially, shrinks as often as it grows. Fields get retired, tools get consolidated, process gets removed when it stops earning its cost.

Companies at stage 5 decay back to stage 4 within about two quarters if the cadence stops, because entropy is the default behaviour of any system in use. Maturity is maintained, never achieved.

Where companies get stuck

Stuck atBecauseThe fix
1 → 2Nobody owns writing definitions downOne week, one document, three signatures
2 → 3Enforcement creates friction now for benefit laterLeadership holding the line for one month
3 → 4Reporting built before the model was fixedRebuild reporting on the enforced model
4 → 5The request queue consumes all capacityRing-fence 30% for roadmap and removal work

The 2-to-3 row accounts for the majority of stalled functions. It is not a technical problem and it will not be solved by better tooling — validation rules take a day to write. It is solved by a leader willing to absorb a month of complaints, which is why the fix belongs to the CRO rather than to RevOps.

Using this honestly

Run the five tests as written and take the lowest stage you fully pass. A company that can explain a forecast miss in a day but cannot say what it retired last quarter is at stage 4, not 5 — and knowing that is more useful than a flattering self-assessment.

Then work on exactly one transition at a time. Attempting to enforce process while simultaneously rebuilding reporting produces two half-finished projects and no measurable change, which is the pattern described in how to build a RevOps strategy.

Want this diagnosed on your own numbers?

The RADAR™ Scan scores your revenue engine in 2 minutes — 12 questions, a 0–100 score, and your gate verdict. No email required.

Run your RADAR™ Scan
FREQUENTLY ASKED

Questions this raises.

What are the stages of RevOps maturity?
Five: ad hoc where process lives in people's heads, defined where definitions are written but not enforced, enforced where the system prevents the wrong path, predictive where the forecast is trusted and explains itself, and optimising where the system improves and shrinks continuously.
What stage are most companies at?
Stage two — defined but not enforced. Definitions exist in a document, everyone was trained, and the system enforces none of it, so practice drifts from documentation continuously. Companies stay here because writing definitions is popular and enforcing them creates immediate friction.
Which RevOps maturity transition matters most?
Stage two to three, from defined to enforced. It produces the largest single improvement in forecast accuracy available to most companies, typically moving a forecast that misses by 25–40% to one within 10–15%, because the stages themselves finally mean something consistent.
Can you skip RevOps maturity stages?
No. Predictive analytics built on unenforced process produces confident nonsense, because the underlying stage data does not mean the same thing across reps. Each stage depends on the previous one holding, which is why reporting built before enforcement has to be rebuilt afterwards.
How do you assess RevOps maturity honestly?
Use observable tests rather than descriptions. Can a new rep be productive without a month of shadowing? Do two functions report the same metric identically? What happens in the system if a rep skips a stage? Can you explain a forecast miss in a day? What did you retire last quarter? Take the lowest stage you fully pass.
WHEN READING ISN'T ENOUGH

First we build your pipeline. Then we build the machine that scales it.

Every engagement starts with the RADAR™ Reveal — a 2-week audit with a scored report, gate verdict, and roadmap. Yours to keep, whatever you do next.

Still figuring out if we can help?

Get a personalized answer from your everyday AI tool