Not Everything That Counts Should Become a KPI

Most marketing teams are not suffering from a shortage of data. They are suffering from a surplus of meaning assigned to data that does not deserve it. The word "key" in Key Performance Indicator has been stretched so thin it barely holds any weight. This post argues that the real measurement problem in modern marketing is not an inability to track things, but an unwillingness to decide which things actually matter. What follows is a walk through the original logic behind KPIs, where that logic gets misapplied, what made the problem worse, and what a functioning measurement system actually looks like in practice.

Key Takeaways

• A metric and a KPI are not the same thing. Every KPI is a metric, but most metrics should never become KPIs.
• Strategic planning experts recommend holding no more than 5 to 7 KPIs at any given time. Most marketing teams track multiples of that.
• What qualifies as a KPI depends heavily on context, campaign goals, and business objectives. There is no universal list.
• KPI inflation dilutes focus. When everything is a priority, nothing is.
• A tiered measurement system, separating diagnostic metrics from reporting metrics from true KPIs, is the practical solution to measurement overload.

The Classic Idea: Measure What Matters, Manage What You Measure

The underlying logic of KPIs is clean. Set a goal. Identify the clearest numerical signal that tells you whether you're moving toward it. Track it. Adjust accordingly.

That logic is sound. The problem is not the logic. The problem is what happened to the word "key."

Management theory has long held that organizations perform better when they commit to a limited set of clearly defined targets. The principle is not mysterious. When you try to watch everything, you end up watching nothing closely enough to act on it. KPIs were meant to be the short list, the few measures that sit above all others because they most directly reflect whether strategy is working.

The marketing version of this principle is simple: define your business objective, identify the metrics most causally connected to it, set a target, track progress, and adjust strategy when the number moves the wrong way. Done correctly, this process produces clarity. Done incorrectly, which is most of the time, it produces a spreadsheet with 40 rows and the illusion of rigor.

What Everyone Gets Wrong: Confusing "Measurable" with "Important"

Here is the real confusion, and it is worth stating plainly: the existence of a number does not make it a KPI. Measurability is not the same as importance.

Marketing teams regularly elevate social media follower counts, email open rates, page views, bounce rates, and impression volume to KPI status. Not because these numbers connect directly to business outcomes, but because the dashboard already has them and someone asked for a slide.

This is not a technology problem or a reporting problem. It is a thinking problem.

The distinction between a metric and a KPI is not subtle. Metrics track and inform. KPIs evaluate strategic progress. A bounce rate tells you something about user behavior on a specific page. Whether that behavior relates to your business goal depends entirely on what your goal is and what that page is supposed to do. A bounce rate on a contact form page is diagnostic intelligence. A bounce rate on a brand awareness landing page is close to irrelevant. The number is the same number. The organizational weight you assign to it should not be.

OnStrategy, which has worked with organizations across sectors on strategic planning, recommends holding between 5 and 7 KPIs per strategic plan. Most marketing teams exceed this by a significant margin before the quarter has started. The instinct to track more is not irrational, but the instinct to call everything a KPI is.

There is also the vanity metric problem, which is worth addressing because it is frequently misdiagnosed. Vanity metrics are not a fixed category. What looks like a vanity metric in one context is legitimate intelligence in another. Impressions mean almost nothing in a conversion campaign. In a brand awareness campaign measured over 18 months, impressions are one of the only quantifiable proxies you have. The issue is not the metric. The issue is whether the metric has been honestly connected to the actual goal or whether it is there because it makes the report look good.

What Changed: When Measurement Got Cheap, KPI Lists Got Long

Tracking anything used to cost something. Setting up measurement infrastructure required technical resources, time, and deliberate choices about what was worth monitoring. That friction was annoying, but it had an accidental benefit: it forced selectivity.

That friction is largely gone. Modern marketing teams operate with access to dozens of platforms, each generating its own performance data in real time. Analytics dashboards populate automatically. Reports can be scheduled, automated, and shared across the organization without anyone having to decide whether the data in them is worth acting on.

The result is measurement abundance paired with strategic confusion. A team might monitor organic traffic, direct traffic, referral traffic, branded search volume, social reach, social engagement, email open rates, click-to-open rates, lead volume, marketing qualified leads, sales qualified leads, pipeline value, closed revenue, CAC, CLV, ROMI, NPS, and 20 other signals simultaneously, calling all of them KPIs, and then struggle to answer the basic question: is this working?

The proliferation of tools did not cause bad measurement strategy. It amplified whatever measurement strategy already existed. Disciplined teams became more disciplined. Undisciplined teams ended up with more noise.

There is also a cultural dimension here. When leadership asks "what does marketing do," the defensive instinct is to show volume. A long list of tracked metrics looks like thoroughness. It reads as accountability. In practice, it is often the opposite because accountability requires the ability to say whether something succeeded or failed, and that requires a standard against which to judge. A 40-metric dashboard rarely provides that. It provides coverage.

What This Means Operationally: How to Build a Measurement System That Actually Works

What is the difference between a metric and a KPI in practice?

The clearest operational distinction is this: a metric tells you what happened, while a KPI tells you whether you are winning. Metrics are inputs to judgment. KPIs are the judgment itself.

In practice, that means KPIs require four things that ordinary metrics do not. They require a clearly defined measure, a specific numerical target, a named data source, and a reporting frequency with someone accountable for it. If any of those four elements is missing, what you have is a metric that has been promoted above its pay grade.

How do you decide which metrics qualify as KPIs?

The most useful test is not "can we measure this" but "does this number directly reflect whether our strategy is working." Ask what business objective this marketing effort supports. Then ask which single number, if it moved significantly in the wrong direction, would tell you the strategy has failed. That number is a candidate for KPI status.

A second useful test is the "so what" question. When the number goes up by 15 percent, so what? If the answer requires three more questions before you land on something actionable, the metric is probably diagnostic, not strategic.

A third consideration involves leading versus lagging signals. Lagging indicators, such as revenue or closed deals, tell you the outcome but arrive too late to be corrective. Leading indicators, such as pipeline volume or trial conversion rate, give you advance warning. Strong KPI sets include both. A team tracking only lagging metrics is flying with no radar. A team tracking only leading metrics can never confirm whether its predictions were right.

What does a tiered measurement system look like?

Rather than flattening all data into one level, a tiered system separates metrics by their organizational function.

The first tier contains diagnostic metrics: signals you monitor to detect anomalies and understand the mechanics of performance. Bounce rates, session durations, email open rates, and social engagement volumes typically live here. They inform optimization decisions but do not define strategic success or failure.

The second tier contains reporting metrics: data you share regularly with stakeholders to show what is happening across campaigns and channels. Customer acquisition cost, conversion rates by channel, and lead volume usually belong here. Useful, worth tracking closely, but not necessarily the measures that define whether marketing is delivering against the business.

The third tier contains true KPIs: the handful of numbers that directly reflect strategic objectives and against which the team is formally accountable. For most marketing teams, these will involve some combination of revenue contribution, customer lifetime value relative to acquisition cost, and whatever retention or growth metric the business has identified as central to its model.

The table below shows how common marketing data points typically map across these tiers, along with the conditions under which a metric earns KPI status.

This is not a ranking system. A metric in the diagnostic tier is not inferior to one in the strategic tier. It is just serving a different function. Confusing those functions is where measurement strategy breaks down.

Why do most marketing teams resist simplifying their KPI lists?

Partly because of genuine uncertainty. If you are not sure which three metrics actually capture whether marketing is working, tracking 30 of them feels safer. The risk of leaving something important out feels greater than the cost of tracking too much.

Partly because of organizational politics. Different stakeholders care about different numbers, and a long KPI list can function as a peace treaty rather than a measurement strategy.

And partly because reducing the list requires committing to a definition of success, which creates accountability. When there are 40 KPIs, it is hard to fail all of them. When there are five, failure becomes legible.

That last point is the uncomfortable one. KPI discipline is not really a data problem. It is a commitment problem. The mechanics of building a tiered system are not particularly complex. Deciding what the team will be held accountable for, and what it will not, requires organizational honesty that is harder to come by.

Stop Tracking More and Start Deciding What Counts

The measurement problem worth solving is not technical. The tools exist. The data exists. The real gap is in the decision about what the data is for, which numbers carry real organizational weight, and which are providing the comfortable illusion of rigor.

A well-constructed measurement system asks less of your dashboard and more of your judgment. It holds five to seven KPIs that are explicitly tied to strategic objectives, supported by a diagnostic layer that informs day-to-day decisions, and reviewed often enough that the team can actually act on what it learns.

Start by auditing everything currently labeled a KPI on your team. Ask, for each one, what business objective it reflects, whether there is a specific target with a deadline, who owns the number, and what would happen if it went wrong. The metrics that cannot answer those questions cleanly are not KPIs. They are metrics. That is not a criticism. That is a classification.

Measure fewer things with greater commitment, and the clarity you have been building dashboards to find will start showing up on its own.

Frequently Asked Questions

What is the difference between a marketing metric and a KPI?

A metric is any quantifiable data point that tracks marketing activity or performance. A KPI (Key Performance Indicator) is a specific subset of metrics that directly measures progress toward a defined strategic objective. Every KPI is a metric, but the reverse is not true. The distinction matters because KPIs carry organizational accountability, which means they require a defined target, a data source, a reporting cadence, and a named owner. Most metrics lack those requirements because they serve a diagnostic or informational function rather than a strategic one.

How many KPIs should a marketing team have?

Strategic planning practitioners recommend holding between 5 and 7 KPIs per strategic plan or planning period. This applies to the organization as a whole, so a marketing team's KPI list should be a subset of that, typically 3 to 5 measures that reflect marketing's contribution to the broader objectives. More than that, and the list stops functioning as a priority system and starts functioning as a tracking list, which is a different thing with a different purpose.

What makes a metric a "vanity metric" in marketing?

A vanity metric is any data point that looks good on a report but does not connect to a business outcome in any actionable way, given the specific goals of a campaign or period. The category is not fixed. Social media follower count is a vanity metric when you are running a conversion campaign, but a legitimate signal when audience growth is the explicit strategic goal. The test is simple: if the number improves significantly, can you explain why that is good for the business? If the explanation requires multiple logical leaps, the metric is probably vanity in its current context.

What is the "so what" test for KPI qualification?

The "so what" test asks what would happen if a given metric moved significantly in either direction. If the answer is immediate and actionable, the metric is a strong KPI candidate. If the answer leads to more questions before landing on something you can actually do, the metric belongs in the diagnostic tier. A 30 percent drop in email open rate, for example, triggers a clear chain of investigation: delivery issues, list hygiene, subject line quality, send time. That is useful intelligence, but it is not a strategic outcome measure. It is a signal that points you toward a problem.

How does marketing ROI differ from other marketing KPIs?

Return on Marketing Investment (ROMI) is calculated by dividing net profit from marketing efforts by the cost of those marketing investments, then multiplying by 100. Unlike most marketing metrics, which measure activity or partial outcomes, ROMI captures the full financial relationship between spend and return. This makes it one of the most strategically meaningful numbers available to marketing teams, particularly when making budget allocation decisions. Its limitation is that it is a lagging indicator: it tells you what worked after the fact, not what is likely to work next. Strong KPI sets pair ROMI with leading indicators that provide earlier warning signals.

Why do marketing teams end up with too many KPIs?

Several forces drive KPI inflation simultaneously. First, the reduction in cost and friction around data collection means more numbers are available without deliberate effort, so they accumulate by default. Second, organizational politics often mean that multiple stakeholders want their priority metrics included in reports. Third, and most fundamentally, committing to a short KPI list requires committing to a definition of success, which creates clear accountability. A long list distributes accountability so broadly that failure becomes hard to identify. KPI discipline is not a data management challenge. It is a governance challenge.