KPI Metrics: Most of Them Are Measuring the Wrong Things

KPI metrics are the numbers a business agrees to track in order to judge whether its marketing is working. The problem is that most organisations pick them badly, track them in isolation, and then wonder why the data never quite explains what happened in the market.

A KPI is only as useful as the decision it supports. Strip away that context and you are left with a number that looks important, gets reported in a slide deck, and changes nobody’s behaviour.

Why Most KPI Frameworks Are Built Backwards

The conventional approach to KPIs goes like this: a team launches a campaign or a channel, the platform spits out a list of available metrics, someone picks the ones that look relevant, and those become the KPIs. It is entirely backwards.

The right sequence is the opposite. Start with the commercial outcome the business needs. Work backwards to the marketing behaviour that would drive it. Then identify the metric that most accurately signals whether that behaviour is happening at scale. That is a KPI. Everything else is a supporting data point at best, and a distraction at worst.

I have sat in enough quarterly business reviews, across enough sectors, to know that most KPI dashboards are built by people who are more comfortable with data than with commercial accountability. The dashboard looks impressive. The metrics are technically accurate. And the business is no clearer about whether marketing is earning its budget.

If you want to understand how analytics tools fit into a broader measurement framework, the Marketing Analytics and GA4 hub covers the full picture, from attribution to custom reporting to making GA4 actually work for your business.

What Is a KPI Metric and What Is Just a Metric?

This distinction gets collapsed constantly, and it causes real problems. A metric is any measurable data point: page views, email open rates, impressions, bounce rate. A KPI, a Key Performance Indicator, is a metric that has been selected because it directly signals progress toward a specific business objective.

Not every metric is a KPI. Not every KPI needs to be a metric you track daily. The confusion between the two is one of the main reasons marketing teams end up reporting on twenty numbers when three would be more useful and more honest.

When I was running the performance division at iProspect, we had clients who wanted weekly reports covering every available data point from every channel. The reports were long. They took hours to produce. And in almost every case, the clients made decisions based on two or three numbers buried inside them. We eventually pushed back and rebuilt the reporting around those two or three numbers. The clients were initially sceptical. Within a quarter, most of them were making faster, better decisions. The noise had been masking the signal.

The Crazy Egg breakdown of website KPIs is a useful reference for anyone trying to separate signal from noise in their site analytics. It does not solve the selection problem, but it gives you a structured way to think about which site metrics are worth elevating to KPI status.

The Four Categories of Marketing KPI Metrics

There is no single correct KPI framework, but there is a useful way to organise the options. Marketing KPIs broadly fall into four categories, and a well-designed measurement system will have representation from each, weighted toward the ones that matter most for the current business objective.

Awareness and Reach Metrics

These measure how many people are being exposed to the brand or message. Impressions, reach, share of voice, branded search volume. They are important for businesses in growth mode or launching into new markets. They are nearly useless as primary KPIs for businesses trying to optimise short-term revenue. The mistake is treating them as universally important regardless of what the business is actually trying to do.

Engagement Metrics

Click-through rates, time on page, scroll depth, email open rates, social engagement. These tell you whether the audience is responding to the content or creative. They are useful diagnostic metrics. They are rarely useful as primary KPIs because engagement does not reliably predict conversion or revenue. A piece of content can generate enormous engagement and drive zero commercial value. I have seen it happen repeatedly, particularly in content marketing programmes where the team was optimising for shares and comments rather than for downstream pipeline.

The Crazy Egg guide to email marketing metrics is worth reading if you are trying to figure out which engagement signals in email actually correlate with commercial outcomes and which ones are just flattering noise.

Conversion Metrics

Conversion rate, cost per acquisition, lead volume, qualified lead rate, return on ad spend. These are the metrics that most closely connect marketing activity to commercial outcomes. They are also the most commonly gamed, because they respond to short-term tactics (heavy discounting, low-quality lead sources, broad match keywords) that inflate the number without improving the underlying business.

Conversion tracking has come a long way since the early days of digital advertising. Search Engine Land’s early coverage of Google’s conversion tracking improvements is a useful reminder of how far the tooling has evolved, and how much more precise the measurement can now be if you set it up correctly.

Revenue and Retention Metrics

Customer lifetime value, revenue attributed to marketing, churn rate, net revenue retention. These are the metrics that actually tell you whether marketing is building a business or just buying short-term numbers. They are also the hardest to measure cleanly, because they require integration between marketing data and financial data that most organisations have not done the work to establish.

The Benchmark Problem: How Do You Know If the Number Is Good?

This is the question that exposes weak KPI frameworks faster than anything else. A 3.2% conversion rate. Is that good? It depends entirely on the industry, the channel, the price point, the traffic quality, and a dozen other variables. Without a benchmark, the number is an observation, not a performance signal.

There are three types of benchmarks worth having. The first is historical: how does this metric compare to the same period last year, or to the last campaign of the same type? The second is competitive: how does it compare to the industry average or to your direct competitors? The third is internal: how does it compare to the target you set when you planned the campaign?

Most teams have the third. Fewer have the first. Almost none have the second in any reliable form. That is a problem, because without competitive context, you can be improving consistently and still be losing ground in the market.

I spent time judging the Effie Awards, and one of the most consistent patterns in the entries that failed to impress was this: strong absolute numbers presented without any competitive or market context. A brand could show impressive growth in awareness, but if the category was growing faster, the brand was actually losing share. The numbers looked good. The story they told was the opposite of good.

GA4 and the KPI Selection Problem

GA4 is a more powerful analytics platform than Universal Analytics was. It is also significantly more complex, and that complexity creates a specific risk for KPI selection: more available data makes it easier to find a metric that tells the story you want to tell.

The platform gives you event-based tracking, cross-device measurement, predictive metrics, and custom reporting capabilities that Universal Analytics never offered. Used well, those features let you build a KPI framework that is genuinely connected to user behaviour and commercial outcomes. Used poorly, they give you more ways to produce impressive-looking dashboards that do not actually explain anything.

The Moz guide to GA4 custom reports is practical if you are trying to configure the platform to surface the metrics that actually matter for your specific business, rather than defaulting to whatever GA4 shows you out of the box.

The Semrush overview of Google Analytics is also worth reading for context on how GA4 fits into a broader measurement stack, particularly if you are integrating it with paid search or SEO data.

One specific GA4 feature worth understanding in the context of KPIs is A/B testing integration. If you are running experiments to improve conversion rates or engagement, you need your testing framework and your KPI framework to be aligned. Semrush’s guide to A/B testing in GA4 covers the mechanics of setting this up in a way that produces reliable data rather than misleading results.

Vanity Metrics and Why They Persist

A vanity metric is one that looks good in a report but does not connect to any commercial outcome. Social media followers. Total page views. Email list size. App downloads. None of these are inherently useless, but all of them become vanity metrics the moment they are tracked without reference to what they are supposed to drive.

The reason vanity metrics persist in organisations is not stupidity. It is incentive structure. If a marketing team is measured on metrics it can control, it will optimise for those metrics. Followers are easier to grow than revenue. Page views are easier to inflate than qualified leads. The team looks productive. The business is not necessarily better off.

Early in my career, I inherited a content programme that was reporting record traffic numbers every month. The client was delighted. When I dug into the data, the traffic was almost entirely from informational searches with no purchase intent. The bounce rate on those pages was high, the time on site was low, and the conversion rate from content traffic to any kind of commercial action was effectively zero. We had been optimising a programme that was generating activity with no commercial value. It took a difficult conversation to reframe what success looked like, but the programme that came out of that conversation was half the size and three times as commercially productive.

The HubSpot argument for marketing analytics over web analytics makes this point well. Web analytics tells you what happened on your site. Marketing analytics tells you whether what happened on your site is connected to anything that matters commercially. Those are different questions, and most teams are only asking the first one.

How Many KPIs Should a Marketing Team Actually Track?

There is no universal answer, but there is a useful rule of thumb: if you cannot explain every KPI on your dashboard to a non-marketer in one sentence, you have too many. The purpose of a KPI is to create shared understanding of what success looks like. That purpose is defeated by complexity.

For most marketing functions, a primary KPI set of three to five metrics is sufficient. These are the numbers that go into the board report, that get reviewed in every leadership meeting, and that the team is genuinely held accountable for. Below that, you can have a broader set of diagnostic metrics that the marketing team tracks internally to understand the levers behind the primary KPIs. But those diagnostic metrics should not be elevated to the same status as the primary set.

When I was growing the iProspect team from around twenty people to over a hundred, one of the things I was most deliberate about was keeping the commercial KPIs simple and consistent. Revenue, margin, client retention, new business wins. Everything else was context. The team knew what they were being measured on. That clarity changed how they made decisions, and it changed how I could hold people accountable without micromanaging the activity beneath those numbers.

The Attribution Layer: Why KPIs Lie When Attribution Is Broken

A KPI is only as accurate as the data feeding it. And the data feeding most marketing KPIs is running through an attribution model that is, at best, an approximation of reality. Last-click attribution, which most platforms still default to, assigns all credit for a conversion to the last touchpoint before purchase. That systematically undervalues upper-funnel activity and overvalues channels that capture intent rather than creating it.

The practical consequence is that your KPIs for paid search will look strong because paid search sits at the bottom of the funnel and collects credit for conversions that were driven by brand activity, content, or social exposure weeks earlier. Your brand and content KPIs will look weak by comparison. You will cut the brand budget. Paid search performance will then decline, because you have removed the upper-funnel activity that was creating the demand paid search was capturing. This cycle plays out in organisation after organisation, and it is driven entirely by a broken attribution model feeding misleading KPIs.

The Moz integration of GA4 and Moz Pro is relevant here because one of the persistent attribution problems is the disconnection between organic search performance and downstream commercial outcomes. Getting those data sources talking to each other is a precondition for building KPIs that accurately reflect the contribution of SEO to the business.

Building a KPI Framework That Actually Works

The process is not complicated, but it requires discipline that most teams do not apply consistently. These are the steps that have worked for me across a range of clients and internal marketing functions.

Start with the business objective. Not the marketing objective. The business objective. Revenue growth, market share, customer retention, margin improvement. Whatever the business is trying to achieve in the next twelve months. Marketing KPIs should be in service of that objective, not parallel to it.

Identify the marketing behaviours that most directly contribute to that objective. If the objective is revenue growth, what does marketing need to do to drive it? Generate more qualified leads? Improve conversion rates? Increase average order value? Reduce churn? Each of those is a different behaviour, and each requires different KPIs.

Select the metric that most accurately signals whether the behaviour is happening. Not the metric that is easiest to track. Not the metric that looks best in a report. The metric that most honestly reflects whether the behaviour is occurring at the scale and quality required.

Set a benchmark before you start tracking. What does good look like? What is the baseline? What is the target? Without those reference points, the number you track is an observation, not a performance signal.

Review the KPI set quarterly, not annually. Business objectives shift. Channel performance shifts. A KPI that was the right measure six months ago may no longer be the right measure today. The framework should be stable enough to allow for trend analysis but flexible enough to reflect changes in the commercial context.

One thing I learned early, when I could not get budget for a new website and ended up building it myself instead, is that the most useful skill in marketing is not the ability to use the tools. It is the ability to ask the right question before you reach for any tool at all. KPI selection is fundamentally a question-asking exercise. What are we trying to achieve? What would tell us we are achieving it? What would tell us we are failing? Answer those questions honestly and the metrics follow naturally.

If you want to go deeper on the analytics infrastructure that makes good KPI tracking possible, the Marketing Analytics and GA4 hub covers attribution, GA4 setup, custom reporting, and the broader measurement questions that sit behind any KPI framework worth building.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions