Team performance metrics: 18 KPIs and how to measure them

Most teams are measured by what they ship: revenue, tickets closed, deadlines met. Those numbers are lagging indicators. By the time output drops, the causes have been building for months: unclear priorities, a role nobody owns, meetings that decide nothing, a quiet loss of trust. Team performance metrics exist to make those causes visible while there is still time to act.

This guide defines 18 KPIs that describe how a team works, explains how to measure each with a short survey, shows how PEAKTEAMS rolls them up into one Team Net Value score from 0 to 100, and covers the part most dashboards skip: why an average can look fine while half the team disagrees with it.

What makes a good team performance metric

A metric earns its place when the team can influence it, it can be measured the same way again in three months, it can be compared across teams, and a low score points to something you could change. Output numbers such as revenue or velocity fail the last test: a drop tells you something is wrong, not what. Survey-based metrics about how the team works are diagnostic. Use both.

Input, process and output: the model behind the KPIs

Team research has organised team effectiveness around an input-process-output model for decades. Inputs are what the team starts with: a purpose, people, resources, leadership. Processes are what the team does with them: how it communicates, decides, resolves conflict and learns. Outputs are what comes out: results and the effect they have on the wider organisation. Ilgen, Hollenbeck, Johnson and Jundt (2005) extended this into the IMOI model: today's outputs feed back into tomorrow's inputs, and many "processes" are really emergent states such as trust. Hackman (2002) adds that a leader's main lever is setting the conditions a team works in.

PEAKTEAMS uses this structure directly: five input KPIs, ten process KPIs, two output KPIs, and one friction KPI for the energy the organisation drains from the team through bureaucracy, politics and instability. Friction stands apart because a high score is bad, and because it takes a share of performance away rather than adding to it.

The 18 team performance KPIs

Input KPIs: what the team starts with

• Purpose & Objectives. Whether every member can say what the team is for, what it is trying to achieve this quarter, and for whom. Low scores show up later as rework and priority fights.
• Values & Vision. Agreement on the longer direction and on the principles that decide trade-offs. Teams can be clear on this quarter's goals and still pull apart over the year.
• Environment & Resources. Whether the team has the budget, tools, access and organisational support it needs. Often the KPI the team controls least, which is why leaders need to see it.
• Team Organization. Role clarity, skill coverage, workload balance and onboarding. Two people believing they own the same decision is a Team Organization problem, not a personality clash.
• Leadership Principles. How decisions get made and explained, whether accountability is explicit, and whether good work is noticed: leadership as the team experiences it, whoever provides it.

Process KPIs: how the team works

• Social Cohesion. The quality of relationships: belonging, mutual respect, willingness to help. Beal and colleagues' 2003 meta-analysis found the link to performance is stronger when performance is measured as behaviour and the task needs close coordination.
• Innovation & Growth. Openness to new approaches and permission to try things that might not work.
• Delivery Methodology. Whether the way the team plans, coordinates and documents work actually helps, or whether delivery depends on a few people carrying it.
• Inner Drive. Individual ownership: members doing what they said they would, without being chased.
• Identification & Engagement. Pride in the team and investment in its mission. Gallup's Q12 meta-analysis reports that business units in the top quartile of engagement show 23% higher profitability than the bottom quartile, and lower turnover: 51% lower in low-turnover organisations, 21% lower in high-turnover ones.
• Feedback & Development. Whether feedback flows, failures are examined instead of buried, and people get better at their jobs.
• Motivation & Focus. Collective energy and the ability to keep it on the priorities.
• Trust & Safety. Psychological safety and interpersonal trust: the belief that you can take an interpersonal risk here without being punished. Google's Project Aristotle named psychological safety the most important of the five dynamics it found in effective teams. See the psychological safety survey guide.
• Communication Culture. Information sharing, listening, and whether concerns can be raised in the open.
• Conflict Navigation. Whether disagreement is addressed early and used to improve, or avoided until it leaks out sideways.

Output KPIs and friction

• Team Output. Delivery against commitments: on time, at the agreed quality, meeting stakeholder expectations. This is the team's own view; place it next to your hard delivery data.
• Organizational Impact. Value beyond the team's boundaries: alignment with strategy, collaboration with other teams, contribution to results the organisation cares about.
• Friction & Consumption. Energy lost to bureaucracy, politics, unproductive meetings and churn. Scored in reverse: agreement means more friction. Ask about it directly; do not infer it from low scores elsewhere.

How to measure each KPI

Every KPI is measured the same way: a few statements that each member rates from 1 (strongly disagree) to 5 (strongly agree). The KPI score is the team average of its items. Simple on purpose: a score the team does not understand will not be acted on.

Who answers. Every member, including the lead, anonymously; a metric only the manager fills in measures the manager's perception. Show results only once a minimum number of responses is in, three by default, so no individual answer can be inferred.

How many items. Two to three per KPI. The PEAKTEAMS team survey takes about 12 minutes per member for all 18 KPIs plus the four team identity dimensions. The team survey questions guide has three items per KPI if you write your own.

How often. Quarterly. More often and people stop answering carefully; less than twice a year and you cannot tell whether an intervention worked. Add a round about eight weeks after a new lead or a reorganisation.

Rolling 18 KPIs into one number: Team Net Value

Eighteen scores are too many for a summary, so PEAKTEAMS computes a Team Net Value (TNV) from 0 to 100: the value a team creates through inputs, processes and outputs, less the share friction takes away.

1. Average the input, process and output KPIs into three scores from 1 to 5, then average those three.
2. Convert to 0 to 100: 1.0 becomes 0, 5.0 becomes 100. These are the value points.
3. Convert friction to a load from 0 to 1 the same way and square it: friction of 3.0 costs about a quarter of the value, 4.0 more than half, 5.0 everything.
4. TNV equals value points times (1 minus the friction tax).

Friction multiplies rather than subtracts, so a team creating little value cannot look average just because it has no friction. A team averaging 4.0 on input, process and output with friction at 1.5 lands at about 74; with friction at 4.0 it drops to around 33, though its own work is unchanged.

Why the average hides the real story

A KPI score of 3.6 can describe two very different teams. In one, everybody answered 3 or 4: a shared, mildly positive view. In the other, half answered 5 and half answered 2: two camps with opposite experiences of the same team. The second team has the more urgent problem.

This is why dispersion belongs next to the average. PEAKTEAMS reports it as team cohesion: how much members agree with each other across all questions, calculated from the variance of their answers. High cohesion means the team shares one picture of itself, good or bad. Low cohesion means the first conversation is not "how do we improve X" but "why do we see X so differently", and on Leadership Principles or Purpose & Objectives it usually means some members get information others do not.

Which metrics to start with

If 18 at once is too much, start with five: Purpose & Objectives, Team Organization, Trust & Safety, Communication Culture and Friction & Consumption. Clear goals and roles, a safe climate, information that flows and an honest account of friction cover most of what goes wrong. Add Team Output to check the self-assessment against delivery.

• Every member answers anonymously; nothing is shown below the minimum response count
• At least two items per KPI on one 1 to 5 agreement scale
• Spread reported next to the average
• Same instrument repeated quarterly; results discussed within two weeks with one or two actions agreed

Measurement is the easy half. What changes performance is the conversation that follows and the one or two things the team commits to before the next survey. A 30-minute team health check is a good format for it.