Book a consultation
Resources

KPIs and metrics: KPI formulas, examples and a free template

KPIs, OKRs and metric frameworks

What it is

A KPI is one of the few numbers leadership chooses to steer by

A KPI, or key performance indicator, is a measurable value that shows progress toward a specific goal. Every key indicator is a metric, but most metrics are not key indicators.

GoalGrow profitable recurring revenue
  1. Win new clients
    • Sourced pipeline
    • Pipeline coverage
    • Win rate
  2. Keep and grow clientsReview first
    • Net revenue retention
    • Gross revenue retention
  3. Protect margin and cash
    • Gross margin
    • Cash conversion
    • Forecast accuracy
Goal, drivers, metrics. The highlighted driver is the one to review first.

A good indicator has five parts: a written formula, a data source, a target, an owner and a review rhythm. Without them, the same number shows different values in different meetings. A business can track hundreds of metrics, yet it should steer by a small set tied to its organizational goals.

Comparison

KPIs track progress, OKRs set direction, metrics measure everything else

The three terms are often mixed up. They do different jobs. Each row compares a metric, a key indicator and an OKR, in that order.

MetricKPIOKR
Know what each one is forMeasures any activityTracks a strategic goalSets an objective with key results
Know how long each one lastsContinuousOngoing, reviewed regularlyUsually a quarter or a year
See each one in practiceWebsite sessionsNet revenue retentionEnter a new segment, measured by pipeline
KPI formulas

Analytics KPIs list: 12 KPI formulas, what each one tells you and who owns it

These 12 analytics KPIs follow the path from first visit to retained revenue, plus two that measure the data team itself. Each row gives the KPI formula, the question it answers and the one role that usually owns it.

FormulaWhat it tells youOwner
Conversion rateConversions / sessions x 100How well the site turns visits into the action you want, such as a demo requestCMO
Engagement rateEngaged sessions / sessions x 100Whether visitors find the content worth their time; GA4 counts a session as engaged when it lasts longer than 10 seconds, has a key event or has 2 or more page viewsCMO
Customer acquisition cost (CAC)Sales and marketing spend / new customers won in the periodWhat it costs to win one customer, and whether growth is getting cheaper or dearerCMO
Lead-to-opportunity rateOpportunities created / leads created x 100Whether marketing sends leads that sales accepts and worksCMO
Win rateDeals won / deals closed (won plus lost) x 100How often the sales team turns a qualified opportunity into revenueHead of sales
Pipeline coverageOpen pipeline due in the period / revenue target for the periodWhether there is enough pipeline to reach the target before the quarter startsHead of sales
Net revenue retention (NRR)(Starting recurring revenue + expansion - contraction - churned revenue) / starting recurring revenue x 100Whether existing customers grow or shrink as a group, before any new salesCEO
Customer churn rateCustomers lost in the period / customers at the start of the period x 100How many customers you lose, and whether retention work is paying offHead of customer success
Gross margin(Revenue - cost of goods sold) / revenue x 100How much of each sale is left to pay for overheads and growthCFO
Forecast accuracy(1 - absolute value of (actual - forecast) / actual) x 100How far leaders can trust the forecast when they plan hiring, spend and cashCFO
Dashboard adoption rateActive dashboard users in the period / intended users x 100Whether the reports the data team builds are used to make decisionsHead of data
Data quality pass rateRecords that pass the agreed quality rules / records checked x 100Whether the numbers behind every other KPI can be trustedHead of data

Treat each formula as a starting point. Agree the filters, data source and period with the owner, then record them in your metric dictionary so every report calculates the KPI the same way.

Targets and benchmarks vary by industry, size and business model. Set them from your own baseline rather than from a published average whose definition may not match yours.

Download the KPI formulas template (Excel, free) The same 12 KPIs in one Excel file: each formula, a worked Excel example on demo inputs and the usual owner. No sign-up.

Sample dashboards

A KPI dashboard shows target, trend and owner on one page

Our sample executive dashboard uses demo data. It shows revenue, margin, pipeline coverage and retention rate, each with its target, trend and owner.

Executive KPI dashboard: target, trend and owner on one page

Demo data
Revenue (CEO)$1.32M▲ +3% vs target
Gross margin (CFO)44%▲ On target
Coverage (Head of sales)2.9x▼ -0.1x vs target
Net retention (CEO)101%▼ -4 pts vs target

Margin is on target; retention is slipping

Net revenue retention vs target, by month, %

100%102%104%106%AprMayJunJulAugSepNRRTarget104% in May: Retention drops below target100%102%104%106%AprJunAug104% in May: Retention drops below target
NRRTarget

So what: Look first at the tile off target and trending the wrong way, then ask its owner why.

KPIActualTargetOwner
Revenue$1.32M$1.28MCEO
Gross margin44%44%CFO
Pipeline coverage2.9x3.0xHead of sales
Net revenue retention101%105%CEO

Use it to decide where to look first: the tile that is off target and trending the wrong way. Then ask its owner why.

See executive dashboards
Benefits

Well-defined indicators turn meetings from debate into decisions

Key indicators turn strategy into a few numbers everyone can follow.

Defined well, they align teams, show progress and flag problems early enough to act.

With agreed indicators, a 60-minute leadership meeting spends 40 minutes on causes and actions instead of 15, and 5 minutes on debating numbers instead of 30Two stacked bars of a 60-minute meeting: before, 30 minutes debate, 15 review, 15 causes and actions; after, 5, 15 and 40.Agreed indicators move meeting time from debate to actionHow a 60-minute leadership meeting is spent, minutesDemo dataDebating whose number is rightReviewing resultsCauses and actionsBefore:no shared definitions30 min15 min15 minAfter:one formula per indicator515 min40 min+25 minutes on causes and actions015304560minutesWith agreed indicators, a 60-minute leadership meeting spends 40 minutes on causes and actions instead of 15, and 5 minutes on debating numbers instead of 30Two stacked bars of a 60-minute meeting: before, 30 minutes debate, 15 review, 15 causes and actions; after, 5, 15 and 40.Demo dataAgreed indicators move meetingtime from debate to actionHow a 60-minute leadership meeting is spent,minutesDebating whose number is rightReviewing resultsCauses and actionsBefore: no shared definitions301515After: one formula per indicator51540+25 minutes on causes and actions

They also change meetings. Leaders stop arguing about whose figure is right. They discuss why a number moved and what to do next, so the team makes informed decisions instead of guesses. Over time, historical data on each indicator helps show which initiatives worked.

All of this depends on governance. An indicator with an unclear formula or a weak data source is worse than none, because it builds confidence in the wrong number.

Good indicators also make accountability fair. When each one has one owner and one formula, people are judged on the same basis.

They support continuous improvement too. Leaders read a few numbers, see which ones moved and spend the meeting on causes and actions.

Best practices

Choosing key indicators is a governance decision, not just an analytics one

Five best practices keep an indicator set small and useful.

  1. 1

    Measure a real goal

    Pick measures that track a stated goal, not what is easiest to report.

  2. 2

    Write it down

    Record formula, source, filters and refresh in a metric dictionary.

  3. 3

    Give it an owner

    One person is accountable for each measure and its target.

  4. 4

    Keep the set small

    Only the few measures that drive leadership decisions.

  5. 5

    Review and retire

    Drop measures that no longer drive action; revisit targets each planning cycle.

Follow them in order before any measure reaches a dashboard.

Types

Leading indicators warn you early; lagging indicators confirm the result

Leading and lagging measures work as a pair.

A leading indicator dips six weeks before the lagging revenue result doesLine chart of a leading indicator (pipeline coverage) and a lagging indicator (revenue), indexed, over 16 weeks; the leading dip comes six weeks earlier.Leading indicators move weeks before resultsPipeline coverage and revenue, indexed to 100, weeklyDemo data8090100110Leading:pipeline coverageLagging:revenueEarly warning, 6 weeks aheadW1W6W11W16A leading indicator dips six weeks before the lagging revenue result doesLine chart of a leading indicator (pipeline coverage) and a lagging indicator (revenue), indexed, over 16 weeks; the leading dip comes six weeks earlier.Leading indicators moveweeks before resultsDemo dataPipeline coverage and revenue,indexed to 100, weeklyLeading: pipeline coverageLagging: revenue8090100110Early warning, 6 weeks aheadW1W6W11W16
Act before results landLeading indicators signal outcomes early, such as pipeline coverageLagging indicators confirm them, such as revenue
Match the measure to the horizonStrategic measures track goals over quartersOperational measures track daily or weekly execution
Measure what people feel tooQuantitative measures use numbers, such as marginQualitative measures capture views, such as satisfaction
Common challenges

KPI mistakes start with missing definitions

The common mistakes: too many indicators, measuring activity instead of outcomes, undocumented formulas and each team calculating the same number its own way. Setting targets without a baseline is another. Each one erodes trust in the dashboard, and a metric dictionary entry prevents each of them.

A metric dictionary entry with one formula, tier, type, source, owner, baseline and target prevents competing formulas, too many indicators, activity metrics and targets without a baselineAnnotated metric dictionary card for NRR: formula, tier, type, source, owner, baseline and target, each linked to the KPI mistake it prevents.One dictionary entry prevents the common KPI mistakesA metric dictionary entry for net revenue retention, with the mistake each field preventsDemo dataNet revenue retention (NRR)Mistake it preventsFormula(Start ARR + expansion − contraction − churn) ÷ start ARRUndocumented or competing formulasTierLeadership KPI, 1 of 12Too many indicatorsTypeOutcome, laggingMeasuring activity, not outcomesSourceBilling system, month-end snapshotOwnerCEOBaseline and target104% over the last 4 quarters; target 105%Targets set without a baselineWrite each formula once, in one place, and every report reads from it.A metric dictionary entry with one formula, tier, type, source, owner, baseline and target prevents competing formulas, too many indicators, activity metrics and targets without a baselineAnnotated metric dictionary card for NRR: formula, tier, type, source, owner, baseline and target, each linked to the KPI mistake it prevents.Demo dataOne dictionary entry preventsthe common KPI mistakesA metric dictionary entry for net revenueretention, with the mistake each fieldpreventsNet revenue retention (NRR)Formula(Start ARR + expansion − contraction −churn) ÷ start ARRPrevents: Undocumented or competingformulasTierLeadership KPI, 1 of 12Prevents: Too many indicatorsTypeOutcome, laggingPrevents: Measuring activity, notoutcomesSourceBilling system, month-end snapshotOwnerCEOBaseline and target104% over the last 4 quarters; target105%Prevents: Targets set without a baselineWrite each formula once, in one place, andevery report reads from it.
KPIs by executive

Each executive needs a different short list of KPIs

Treat each list as a KPI sample to adapt, not a standard.

Of 48 metrics a company tracks, 12 are leadership KPIs, three per executive, and 36 are supporting metricsOne bar of 48 metrics: CEO, CFO, CMO and COO hold 3 KPIs each; 36 are supporting metrics.Leadership steers by 12 of the 48 metrics it tracksMetrics in a mid-size company's reports: KPIs per executive vs supporting metricsDemo data12 KPIs, 3 per executiveCEO3CFO3CMO3COO336 supporting metrics: they explain why a KPI movedReviewed everyleadership meetingReviewed by the teams that own themTreat each list as a sample to adapt; the role cards below show which three each executive keeps.Of 48 metrics a company tracks, 12 are leadership KPIs, three per executive, and 36 are supporting metricsOne bar of 48 metrics: CEO, CFO, CMO and COO hold 3 KPIs each; 36 are supporting metrics.Demo dataLeadership steers by 12 of the48 metrics it tracksMetrics in a mid-size company's reports:KPIs per executive vs supporting metrics36 supporting metrics12 KPIs, 3 per executiveCEO 3 · CFO 3 · CMO 3 · COO 3Supporting metrics explain why a KPI moved;the teams that own them review them.Treat each list as a sample to adapt; the rolecards below show which three each executivekeeps.

CEO: steer growth

Revenue growth, net revenue retention, north star metric.

CFO: protect cash

Gross margin, cash conversion, forecast accuracy.

CMO: fund what converts

Sourced pipeline, cost per opportunity, conversion rate.

COO: deliver on time

On-time delivery, cycle time, capacity utilization.

Analytics for SaaS companies
Frequently asked questions

KPI formulas and analytics KPIs: common questions

Questions leaders ask when they define or review their KPIs.

The basics4

Analytics KPIs are the few measures a business uses to judge whether its marketing, sales, finance, customer and data work is reaching its goals. Each one has a written formula, a data source, a target and an owner. Common examples are conversion rate, customer acquisition cost, win rate, net revenue retention and forecast accuracy.

There is no fixed number. A leadership team usually steers by a short list it can review in one meeting, and each department adds a few of its own. If a KPI has not led to a decision or an action for several review cycles, retire it or move it back to the supporting metrics.

Use benchmarks with care. Published figures often mix company sizes, business models and definitions, so they rarely compare like with like. Start from your own baseline over the last few periods, then set a target the owner agrees is ambitious and achievable. Use an external benchmark only when its source and definition match yours.

Leading KPIs move before the result, such as pipeline coverage or engagement rate, so there is still time to act. Lagging KPIs confirm the result after the period, such as revenue, gross margin or churn. A useful KPI set pairs each lagging KPI with one or two leading KPIs that drive it.

Defining and owning KPIs2

Name the numerator, the denominator, the period and the filters. For conversion rate, that means which conversions count, whether you divide by sessions or users, which dates apply and whether internal traffic is excluded. Record the formula and its source system in a metric dictionary, so every report calculates it the same way.

One named person per KPI, usually the leader whose team moves it most. The business owner sets the target and decides the actions when the KPI is off track. The data team owns the calculation, the data source and the data quality checks, so the owner can trust the number in every review.

Related resources

Guides 4

Solutions 1

Next-step finder

Where should your KPI work start?

1 question. We suggest one next step. Nothing is sent anywhere.

Illustrative
1. What do you need right now?
0 of 1 answered

Get a KPI dashboard your leadership team will use

Start with a consultation to define your indicators; we build the dashboard once you are ready.

See executive dashboards
See how we work
A small KPI tree links one goal to three drivers and six metricsKPI tree: profitable growth splits into revenue, margin and retention, each with two metrics; margin is highlighted.Profitable growthRevenueNew bookingsWin rateMarginGross marginCost to serveRetentionNRRChurnA small KPI tree links one goal to three drivers and six metricsKPI tree: profitable growth splits into revenue, margin and retention, each with two metrics; margin is highlighted.Profitable growthRevenueNew bookingsWin rateMarginGross marginCost to serveRetentionNRRChurn