Strategy

The Marketing KPIs That Matter (and the Vanity Metrics That Don't)

Rafal ChojnackiBy Rafal Chojnacki11 min

A marketing KPI is a metric that measures progress toward a business goal — and the test of whether a number is a real KPI or a vanity metric is simple: does it change a decision? Impressions, follower counts, raw traffic and email open rates fill most dashboards, but they usually move without telling anyone what to do differently. The KPIs that matter connect to revenue, cost and growth, sit in a clear hierarchy from a single north-star metric down to diagnostic numbers, and each earns its place by informing an action.

The Marketing KPIs That Matter (and the Vanity Metrics That Don't)

TL;DR

  • A KPI is only useful if it changes a decision. If a number can go up or down and nobody does anything differently, it is a vanity metric.
  • Vanity metrics (impressions, followers, raw traffic, likes, open rate in isolation) move without meaning and crowd out signal.
  • Build a hierarchy: one north-star metric, a few business KPIs (revenue, CAC, LTV, MER), channel KPIs beneath, and diagnostic metrics at the bottom.
  • Separate leading from lagging indicators. Lagging KPIs (revenue) confirm; leading ones (pipeline, qualified leads) predict and let you act in time.
  • KPIs differ by funnel stage and business model — ecommerce, B2B and services track different things.
  • Fewer KPIs, watched closely, beat many watched loosely. A crowded dashboard hides the signal.
  • Every KPI should map to an owner and an action, or it does not belong on the report.

The test: does it change a decision?

The fastest way to clean up a marketing report is to ask of every number: if this moved, what would we do? If the answer is "nothing" or "not sure", it is not a KPI — it is decoration. This one test separates the metrics worth tracking from the ones that merely accumulate.

Vanity metrics fail the test because they are unattached to a decision. Impressions rise when spend rises, and tell you nothing about whether the spend worked. Follower counts grow slowly and steadily and almost never trigger an action. Raw website traffic can double from a low-quality source and mean less revenue, not more. These numbers feel like progress because they usually go up, but "went up" is not the same as "mattered".

The metrics that pass the test are tied to money and movement: revenue, cost of acquisition, conversion rate, pipeline, retention. When one of these moves, a decision follows — spend more, spend less, fix the funnel, change the offer. That link to action is what makes a metric a KPI.

Vanity metrics versus KPIs that actually drive decisions.

Vanity metrics vs KPIs

The same raw number can be vanity or signal depending on whether it is connected to an outcome.

Vanity metric The KPI it should become
Impressions Cost per acquisition, conversions
Followers Revenue or leads attributable to the audience
Raw traffic Traffic-to-conversion rate, revenue per visit
Email open rate Revenue per email, click-to-conversion
Likes / engagement Assisted conversions, brand search lift
Ad clicks Conversion rate, cost per qualified outcome

The pattern: a vanity metric measures activity; a KPI measures outcome or efficiency. Clicks are activity; conversion rate is outcome. Impressions are activity; cost per acquisition is efficiency. The fix is rarely to delete the raw number — it can stay as a diagnostic — but to promote the outcome metric to the KPI slot and demote the activity metric to a supporting role.

A KPI hierarchy: north-star metric, drivers and inputs.

Build a KPI hierarchy

Good reporting is layered, not flat. A crowded dashboard where every metric has equal weight hides the signal; a hierarchy makes the important numbers obvious.

  • North-star metric (1). The single number that best captures the value the business delivers and predicts long-term growth — for example, revenue, active customers, or qualified pipeline. Everything ladders up to it.
  • Business KPIs (3–5). Revenue, CAC, LTV, MER, retention — the numbers leadership judges the whole function on.
  • Channel KPIs. Per-channel efficiency (ROAS, cost per lead, conversion rate) used by the team to steer each channel.
  • Diagnostic metrics. The activity numbers (impressions, clicks, CTR) that explain why a KPI moved, consulted when something needs investigating — not reported as headlines.

The hierarchy answers different questions at different altitudes: the board reads the north-star and business KPIs, the team works the channel KPIs, and diagnostics are pulled when a KPI needs explaining. Presenting all four layers as equals — the typical crowded dashboard — is why so many reports are long and useless. The visualisation of this is covered in the companion marketing dashboard growth teams should track; this piece is about which numbers, that one is about how to show them.

Leading versus lagging indicators on a timeline.

Leading vs lagging indicators

A KPI set built only on lagging indicators can confirm what happened but cannot help you act in time. The strongest sets balance both.

  • Lagging indicators report the outcome after the fact — revenue, closed deals, MER. They are the truth, but by the time they move, the period is over.
  • Leading indicators predict the outcome early — qualified pipeline, add-to-cart rate, trial starts, demo bookings. They move first, so they let you intervene while it still matters.

For a business with a long sales cycle, leading indicators are essential: waiting for closed revenue to react means reacting a quarter too late. A healthy report pairs each lagging KPI with the leading indicator that predicts it, so the team sees trouble (or opportunity) coming rather than confirming it afterwards.

Glossary

  • KPI — a metric measuring progress toward a business goal, useful only if it changes a decision.
  • Vanity metric — a number that moves without informing an action (impressions, followers, raw traffic).
  • North-star metric — the single number that best captures delivered value and predicts growth.
  • Leading indicator — a metric that predicts an outcome early (pipeline, add-to-cart).
  • Lagging indicator — a metric that confirms an outcome after the fact (revenue, closed deals).
  • Diagnostic metric — an activity number used to explain why a KPI moved, not reported as a headline.

KPIs by business model

The right KPI set depends on how the business makes money.

Model Core KPIs Common vanity trap
Ecommerce Revenue, ROAS/MER, CAC, AOV, repeat rate, conversion rate Traffic and impressions without conversion
B2B / SaaS Qualified pipeline, CAC, payback, win rate, LTV MQL volume with no pipeline link
Services / lead gen Cost per qualified lead, close rate, revenue per lead Raw lead count regardless of quality
Content / media Revenue per visit, subscriber value, retention Pageviews and time-on-site alone

Across all of them, the trap is the same shape: a volume metric (traffic, MQLs, raw leads, pageviews) treated as success without the quality or revenue metric that gives it meaning. The discipline is to always pair a volume number with the outcome number that qualifies it.

Every KPI needs a measurement contract

A metric name is not a definition. “CAC,” “marketing-sourced revenue,” and “qualified lead” often mean different things to marketing, sales, and finance. Each KPI needs a written formula, source system, included costs, time zone, currency convention, refresh cadence, owner, and decision threshold.

The dashboard should expose data quality as well as performance. Useful controls include the share of events with a stable ID, deduplication rate, percentage of leads with a current CRM stage, and last validation date. Missing data must not silently become zero.

One KPI should map to one class of decision. If a movement does not trigger an action, threshold, and accountable owner, it is an observation rather than a key performance indicator. This turns reporting from a scorecard into an early-warning and resource-allocation system.

How Space Ads approaches marketing KPIs

Across the accounts we audit, the reporting problem is rarely too little data — it is too much of the wrong kind. Dashboards are crowded with impressions, clicks, followers and open rates, and the numbers that would actually change a decision — CAC, payback, qualified pipeline, MER — are buried or missing. Everyone looks at the report, nobody acts on it, because it does not point to an action.

Our approach is to cut the report to a hierarchy: one north-star, a handful of business KPIs tied to revenue, channel KPIs for steering, and diagnostics kept for investigation. Every KPI gets an owner and a decision it informs; anything that fails the "would this change what we do?" test comes off the headline view. We pair lagging KPIs with the leading indicators that predict them, so the team acts in time. That reporting backbone is the core of web analytics and how we run performance marketing; when a leadership team needs marketing measured in numbers it can act on, a fractional CMO brings that discipline.

Stop doing / Do instead

Stop doing Do instead
Reporting impressions, followers and raw traffic as success Promote the outcome metric (CAC, conversion, revenue)
Treating every metric with equal weight Build a hierarchy: north-star, business, channel, diagnostic
Relying only on lagging indicators Pair each with a leading indicator that predicts it
Celebrating volume without quality Pair every volume metric with an outcome metric
Crowding the dashboard Track fewer KPIs closely; keep diagnostics for investigation
Reporting numbers with no owner Give every KPI an owner and a decision it informs

FAQ

What is a marketing KPI?

A marketing KPI is a metric that measures progress toward a business goal, such as revenue, customer acquisition cost, conversion rate or qualified pipeline. The practical test of a real KPI is whether it changes a decision — if a number can move and nobody acts differently, it is a vanity metric, not a KPI.

What are vanity metrics in marketing?

Vanity metrics are numbers that move without informing an action — impressions, follower counts, raw website traffic, likes and email open rates in isolation. They feel like progress because they usually rise, but they do not tell anyone what to do differently, so they crowd out the metrics that do.

What marketing KPIs should a company track?

Build a hierarchy: one north-star metric, three to five business KPIs (revenue, CAC, LTV, MER, retention), channel KPIs for steering each channel, and diagnostic metrics kept for investigation. The exact set depends on the business model — ecommerce, B2B and services track different core numbers.

What is the difference between leading and lagging indicators?

Lagging indicators confirm an outcome after it happens (revenue, closed deals); leading indicators predict it early (qualified pipeline, add-to-cart rate, trial starts). Lagging KPIs are the truth but arrive late; leading ones let you act while it still matters. A good KPI set pairs each lagging metric with a leading one.

How many marketing KPIs should you track?

Fewer than most dashboards show. A crowded report with dozens of equal-weight metrics hides the signal. A workable structure is one north-star, three to five business KPIs at the headline level, channel KPIs for the team, and diagnostics available on demand rather than reported as headlines.

Is website traffic a vanity metric?

On its own, usually yes — traffic can rise from low-quality sources and produce less revenue, not more. It becomes a KPI when paired with an outcome: traffic-to-conversion rate, revenue per visit, or conversions from that traffic. Measure what the traffic does, not just how much of it there is.

Key takeaways

  • A KPI is only useful if it changes a decision; if nothing follows from it moving, it is vanity.
  • Vanity metrics measure activity; KPIs measure outcome or efficiency.
  • Build a hierarchy: north-star, business KPIs, channel KPIs, diagnostics.
  • Pair lagging indicators with the leading indicators that predict them.
  • Track fewer KPIs closely, give each an owner and a decision, and keep diagnostics for investigation.

Sources and further reading

Continue learning

Continue reading

Success Stories

The same operating standard, across different models