# GTM Metric Definitions — worksheet

Take this to each department. One row per definition. **One named human per row — not a team.**

Thirty definitions is the ceiling, not the target. If your list runs past forty you are modelling
the business instead of the questions your exec team actually asks.

Three columns to fill in, and the third is the one people skip:

| Column | What goes in it |
|---|---|
| **Owner** | One person's name. "RevOps" is not an owner. Two owners produce two definitions inside a quarter. |
| **Source of truth** | System, object, field. Not "Salesforce" — `Opportunity.ACV__c`. |
| **Agreed definition** | Plain English, in your words, in one or two sentences. |

Each group below states **the one fight** that decides the rest of it, and a **defensible default**
you can adopt today and revisit later. The failure is not picking the wrong default — it is having
two of them in circulation at once.

---

## 1 · Company-wide — Identity

**The fight:** does a subsidiary roll up — and does it roll up the same way for reporting and for comp?

**Default:** domain is the spine. Hierarchy is an attribute sitting on top of identity, never identity
itself. Decide reporting and comp separately; they are rarely the same answer.

| Definition | Owner | Source of truth | Agreed definition |
|---|---|---|---|
| Account | | | |
| Person / contact | | | |
| Parent–child relationship | | | |
| Duplicate and merge rule | | | |
| Account ownership | | | |

---

## 2 · Company-wide — Time

**The fight:** who stamps the date a deal entered a stage?

**Default:** system-stamped entry and exit, always. If a human can type it, cycle time is unmeasurable —
this is the one prerequisite with no workaround. You cannot backfill a timestamp that was never written.

| Definition | Owner | Source of truth | Agreed definition |
|---|---|---|---|
| Clock start / clock stop | | | |
| Cycle time | | | |
| Stage aging | | | |
| Fiscal calendar | | | |
| As-of vs period | | | |
| Snapshot cadence | | | |

> **Check this first.** In Salesforce: is field history tracking on for `Opportunity.StageName`, and are
> your stage-date fields writable by a rep? In HubSpot: are `hs_date_entered_*` populated? It is a
> two-hour question and roughly half of what AI can do for you depends on the answer.

---

## 3 · Company-wide — Money

**The fight:** what counts as a customer, and who decides?

**Default:** define it on revenue, not on status. "Has active contracted revenue" is computable; a status
field is maintained by whoever remembers.

| Definition | Owner | Source of truth | Agreed definition |
|---|---|---|---|
| Bookings | | | |
| ARR / MRR / TCV | | | |
| New vs expansion vs renewal | | | |
| Discounts, ramps, multi-year | | | |
| Churn and contraction | | | |
| Customer | | | |

> **Count your currency fields before this meeting.** Most orgs have ten to fifteen on the opportunity —
> `Amount`, ARR, ACV, TCV, MRR, Net ARR, plus qualification-framework amounts that are not revenue at
> all. Exactly one of them means bookings. Nothing in your CRM says which.

---

## 4 · Company-wide — Segmentation

**The fight:** whose definition of Enterprise wins — sales', marketing's or finance's?

**Default:** one segment field, computed, not typed. Derived from firmographics on the resolved account —
never a picklist three teams maintain differently.

| Definition | Owner | Source of truth | Agreed definition |
|---|---|---|---|
| Segment | | | |
| ICP tier | | | |
| Industry | | | |
| Employee band | | | |
| Region | | | |
| Product line | | | |

---

## 5 · Sales — Funnel & lifecycle

**The fight:** what makes an opportunity qualified?

**Default:** an observable event, never a judgment. "Accepted by sales and a discovery meeting occurred."
Anything needing a rep's opinion drifts within a quarter.

| Definition | Owner | Source of truth | Agreed definition |
|---|---|---|---|
| Lead | | | |
| MQL | | | |
| Qualified opportunity | | | |
| Stage entry / exit | | | |
| Pipeline created | | | |
| Re-opened deals | | | |
| Disqualified vs no-decision | | | |

> **Check your stage picklist for more than one motion.** A single list routinely carries new business
> *and* renewal or expansion — stages like `Open Renewal` or `Expansion Identified` sitting after
> `Closed Won`. A win rate across both blends motions that convert at completely different rates.
> Closed Lost is not a second motion; it is the same funnel's terminal outcome.

---

## 6 · Marketing — Source & channel

**The fight:** are these one field or two? Most teams have one.

**Default:** two fields. Source is immutable first touch. Channel is what marketing spends against and is
mutable. Conflate them and you can never restate history.

| Definition | Owner | Source of truth | Agreed definition |
|---|---|---|---|
| Source | | | |
| Channel | | | |
| Campaign | | | |
| Self- vs marketing-sourced | | | |
| Attribution model | | | |

---

## 7 · Customer Success — Retention

**The fight:** when does an account count as at risk, and who is allowed to say so?

**Default:** risk is computed from observable signals — usage trend, support volume, sentiment, delivery
against what was promised at kickoff — not from a rep's health-score dropdown.

| Definition | Owner | Source of truth | Agreed definition |
|---|---|---|---|
| Renewal date | | | |
| At-risk | | | |
| Health score inputs | | | |
| Expansion opportunity | | | |
| Contraction vs churn | | | |

---

## Before you start building

Three things, none of which need budget:

1. **Run the four-person test.** Ask your CRO, CMO, CFO and Head of Sales — separately, no conferring —
   to write down qualified opportunities created last quarter. Twenty minutes. It will tell you more
   than a vendor evaluation.
2. **Check whether stage transitions are system-stamped.** Section 3 above. Two hours to answer.
3. **Write the twenty questions your exec team actually asks.** That list is the scope of the entire
   programme, and it is what stops this becoming a two-year data project.

## The six-week checkpoint

Ask for one number, from your new definitions, that **contradicts** something in your current board deck.

If the work is real, there will be one — and finding it is the point. If everything matches perfectly,
nothing has actually been defined yet.

---

*LeanScale · prepared for the Updata CEO Summit · leanscale.team*