Growth metrics are operating controls, not wall art.
They should help leaders decide what to scale, stop, repair, or sequence next. Most dashboards do something weaker: they report activity, defend functions, and create the illusion of control. Traffic, leads, MQLs, signups, opportunities, pipeline, conversion rates, and revenue can all be useful. None are automatically honest.
The best growth metrics are hard to game without improving the business.
Metrics should expose system health
A growth system needs metrics that answer operating questions:
- Are we acquiring the right customers?
- Are they activating into real value?
- Do they retain?
- Do they expand?
- Which channels produce quality, not just volume?
- Which loop is getting stronger?
- Which constraint is binding?
- Where are we saturating?
- What burden are we creating for sales, onboarding, support, and product?
If the dashboard cannot answer those questions, it is not a growth dashboard. It is reporting decor.
The growth dashboard
Build the dashboard around decisions, not departments.
| Decision area | Metric family | Useful signals |
|---|---|---|
| Customer quality | Segment and source quality | ICP match, retained revenue by source, support burden by source |
| Activation | Activation quality | Activation event completion, time to value, activation-to-retention rate |
| Retention | Cohort health | Logo retention, revenue retention, usage retention, churn reasons |
| Expansion | Account growth | Expansion by cohort, seat/use-case growth, product-qualified expansion |
| Acquisition economics | CAC quality | CAC, payback, sales cycle, discounting, channel margin |
| Pipeline quality | Sales reality | Stage conversion, no-show rate, win rate by source, loss reasons |
| Channel health | Saturation and efficiency | CAC trend, conversion decay, creative fatigue, audience overlap |
| Loop health | Compounding signals | Invite-to-activation, referral quality, content-assisted retained revenue, partner repeatability, loop health by cohort/source |
| Operating burden | Capacity cost | Onboarding hours, tickets per account, support burden by source, implementation backlog, sales capacity used |
| Constraint movement | Bottleneck progress | The metric tied to the current constraint |
The last row is important. Every growth review should know the current constraint and whether it moved.
Attribution humility checklist
Attribution is useful until it becomes fake precision.
Multi-touch growth is messy. Buyers see content, hear from peers, attend events, talk to sales, read reviews, compare vendors, ask communities, and return months later through search or outbound. Pretending one model captures all of that perfectly leads to bad decisions.
Use this checklist:
- Are we treating attribution as evidence rather than truth?
- Are we reviewing cohort quality by source, not just credited pipeline?
- Are we asking sales and CS what actually influenced the deal?
- Are we separating demand creation from demand capture?
- Are we accounting for brand, word-of-mouth, partners, and dark social where relevant?
- Are we avoiding overfunding the most measurable channel at the expense of the most effective one?
- Are we looking at retained revenue, not just sourced revenue?
The goal is not perfect attribution. The goal is better allocation decisions.
Metrics that lie by omission
Some metrics are not wrong; they are incomplete.
Lead volume omits fit. CAC omits retention if viewed too early. Conversion rate omits discounting and support burden. Pipeline omits probability and quality. Revenue omits churn risk. Activation omits whether the behavior predicts retention. NPS omits usage reality. Traffic omits intent.
A metric lies when leaders use it to make a decision it cannot support.
Metrics that resist gaming
Prefer metrics connected to customer value and economic reality:
- cohort retention by segment and source;
- activation-to-retention conversion;
- retained revenue by acquisition source;
- CAC payback by customer quality band;
- expansion by use case and cohort;
- support burden per retained dollar;
- pipeline conversion by source and segment;
- channel saturation trends;
- loop conversion, such as invite-to-activated-account or referral-to-retained-customer;
- constraint movement over time.
These metrics are not perfect. They are just harder to improve cosmetically.
The operating cadence
A dashboard is only useful if it changes decisions.
In the growth review, ask:
- What did we learn?
- What scaled and stayed healthy?
- What grew but degraded quality?
- What should stop?
- What constraint moved?
- What is the next constraint?
- What artifact or operating rule needs to change?
If the meeting ends with commentary but no decision, the metrics did not do their job.
The operator's rule
Do not build dashboards to observe growth.
Build dashboards to manage the growth system.
