What Are Digital Marketing KPIs? The Metrics Worth Tracking to Measure Performance

The sentence I hear most often in a first meeting with a new client goes like this: "We measure everything, but we don't know what to look at." In this guide I put digital marketing KPIs into one catalog, organized by channel and funnel stage. For each indicator you also get the formula, the question it answers and a practical way to set a target. Reading a finished report is a separate topic, so here my goal is simpler: help you pick the right indicators and calculate them correctly.

What are digital marketing KPIs?

Digital marketing KPIs are a few chosen measures that show how close your ads, SEO, social media and email work gets to a business goal. For example, cost per acquisition, return on ad spend, organic clicks and customer lifetime value belong here. In short, every KPI is a metric, but not every metric deserves KPI status.

I like to keep the definition narrow, so a number earns KPI status only when it meets three conditions. First, it connects to a business goal. Second, one person owns it. Third, someone makes a decision when it moves. Otherwise, any number that fails this test can live on a dashboard, but it should stay off the meeting agenda.

I also recommend limiting the count on purpose. My field observation is simple: when a manager sees five to eight indicators on one page, they make decisions. Then again, when they see twenty, they look at the biggest number and move on. So pick a few indicators from the catalog below instead of all of them.

What is the difference between a KPI and a metric?

A metric is anything you can count: impressions, clicks, sessions, likes, time on page. A KPI, in contrast, is one of those metrics that you tie to a goal and track on a fixed rhythm. In other words, the difference is not in the number itself. Instead, it is in the job you give the number.

For example, organic sessions can be the main KPI for a publisher. On the other hand, the same number is only a supporting metric for a service firm that wins clients through a quote form. That firm's real KPI is qualified leads from organic search.

In practice, there is a quick test you can run on any number. Write this question next to it: "If this halves next month, what will I do?" If the answer is a clear action, you have a KPI. If the answer is "we'll keep an eye on it", it is probably a metric, and it belongs on page two.

• Metrics: impressions, clicks, sessions, followers, opens.
• KPIs: cost per acquisition, qualified leads, ROAS, customer lifetime value.
• Context indicators: market share, seasonality, competitor ad pressure.

How do you group digital marketing KPIs by funnel stage?

In my work, the most useful grouping follows the customer journey. I work with four stages: awareness, consideration, conversion and retention. Each stage asks its own question, so each stage also needs its own KPI set.

1. Awareness: How many people saw you, and were they the right people?
2. Consideration: Did those people show interest and spend time on your site?
3. Conversion: Did interested visitors buy, fill in a form or call?
4. Retention: Did customers come back, buy again or refer someone?

The biggest benefit of this split is fairness, because it stops you from judging a channel by the wrong yardstick. For instance, if you expect direct sales from a video campaign, you will switch it off too early. That campaign's job is awareness, so awareness KPIs should judge it. Likewise, if you count branded search traffic as new customer acquisition, you are fooling yourself.

If you want to connect funnel stages to a retargeting setup, my post on strengthening the sales funnel with remarketing shows which message I use at each stage.

Which KPIs matter at the awareness stage?

At the awareness stage the question is simple: did you reach the right people often enough? So volume and cost indicators lead here. However, reading volume on its own is risky, because cheap impressions often go to the wrong audience.

• Reach: The number of unique people who saw your ad at least once.
• Frequency: Impressions divided by reach. It also tells you how often the average person saw the ad.
• CPM: Spend divided by impressions, times 1,000. Put simply, it is the cost of a thousand impressions.
• Branded search volume: Impressions for queries that contain your brand name in Search Console.
• Video view rate: People who watched a set length of the video divided by impressions.

Of these, I trust branded search volume the most, because it reflects real memory. When people see your ad and then type your name into Google, awareness has really formed. It is also free, and it does not depend on the ad platform's own reporting.

Watch frequency closely too. In practice, very high frequency means you keep showing the same ad to the same person and burn budget. Then again, very low frequency can mean the message never sticks. In the end, the right range depends on the industry and the creative. So instead of copying a fixed number, compare against your own history.

How do you measure interest at the consideration stage?

At the consideration stage the person knows you and compares options. So the KPIs here measure the quality of interest. Click-through rate, engaged sessions and product page views are the core indicators.

Google Analytics 4 gives a clear definition for this stage. According to Google's GA4 metric definitions, an engaged session lasts longer than 10 seconds, has a key event, or has at least two page or screen views. Engagement rate, then, is the share of engaged sessions. Bounce rate is now simply the opposite: the share of sessions that did not engage.

• Click-through rate (CTR): Clicks divided by impressions, times 100.
• Engagement rate: Engaged sessions divided by total sessions.
• Views per session: Page views divided by sessions.
• Micro conversion rate: Steps such as add to cart, brochure download or pricing page visit.

Micro conversions save the day for businesses with long sales cycles. A B2B firm, for example, might get only a handful of quote requests a month. You also cannot optimize a campaign on that little data. Still, if you track steps like pricing page visits or catalog downloads, you catch signals much earlier.

Conversion KPIs and their formulas

The conversion stage is where leadership pays the most attention. Therefore every indicator here needs a link to money. The formulas below work for both paid and organic channels.

• Conversion rate: Conversions divided by clicks (or sessions), times 100.
• CPA (cost per acquisition): Total spend divided by conversions.
• CPL (cost per lead): Total spend divided by form fills or calls.
• ROAS: Revenue from ads divided by ad spend.
• ROI: (Profit minus investment) divided by investment, times 100.
• Average order value: Total revenue divided by number of orders.

Do not mix up ROAS and ROI. The reason is simple: ROAS divides revenue by spend and ignores product cost and shipping. As a result, a ROAS of 4 can mean profit for one business and a loss for a low margin one. You can run the numbers quickly with my ROAS calculator.

For service firms, CPL alone is not enough either. After all, cheap leads are often weak leads. That is why I always add a qualified lead rate next to CPL. Put simply: how many of the incoming forms actually reached the proposal stage? Making ad decisions without pulling this rate from the CRM is like driving with your eyes closed.

Which KPIs track retention and loyalty?

Most businesses forget the last stage of the funnel, yet a large part of profit sits there. Winning a new customer is expensive, while convincing an existing one to buy again is usually far cheaper. Therefore you should treat retention KPIs as seriously as conversion KPIs.

• Repeat purchase rate: Customers with a second order divided by all customers.
• Customer lifetime value (CLV): Average order value times yearly order frequency times average customer lifespan in years.
• Churn rate: Customers lost in the period divided by customers at the start of the period.
• CLV to CAC ratio: Customer lifetime value divided by customer acquisition cost.

The CLV to CAC ratio is my strongest argument in budget talks. A customer may look expensive to win, but if they keep buying for years, the math changes completely. The reverse is also true. For example, a customer you won cheaply who never returns can reveal that a campaign that looked good on paper actually lost money.

If you want to see how I join order data with ad data on the store side, my e-commerce consulting page explains the setup.

Which KPIs should you track for SEO?

The most common SEO mistake is making rankings the only KPI. Rankings matter, but personalization and location mean every user sees a slightly different page. In addition, AI summaries now absorb part of the clicks. So I measure SEO in four layers.

1. Visibility: Search Console impressions and average position.
2. Interest: Organic clicks and organic CTR (clicks divided by impressions).
3. Quality: Engagement rate of organic sessions.
4. Value: Key events, leads and revenue from organic search.

In practice, layered reading pays off fast. If impressions rise while clicks fall, the problem sits in your titles and descriptions. On the other hand, if clicks rise while conversions fall, the problem sits on the page itself. That way you get a precise diagnosis instead of a vague "SEO isn't working".

To lift CTR, test your title in the Google SERP preview tool before you publish. I covered searches that show you without sending a click in my post on zero click searches. If you want me to set up SEO measurement for you, have a look at my SEO consulting page.

Google Ads KPIs: from Quality Score to ROAS

Google Ads offers the richest measurement of any channel. That richness is also a trap, because the interface has hundreds of columns. When I manage an account, I sort indicators into three groups: efficiency, quality and results.

The efficiency group holds CPC (cost per click), CTR and impression share. Impression share tells you how often your ad actually showed out of all the searches it qualified for. The report splits lost share into budget and rank, so looking at the two separately shows whether you face a budget problem or a quality problem.

Quality Score sits at the center of the quality group. According to Google's Quality Score help page, the score runs from 1 to 10 and rests on three components: expected CTR, ad relevance and landing page experience. Specifically, Google describes it as a diagnostic tool, not a direct input to the auction. Even so, a low score points to the weak component.

The results group covers CPA, conversion rate, conversion value and ROAS. I explained how I plan budgets around these results in how to set a Google Ads budget. If you are thinking about outside help, my Google Ads management page explains how I work.

What should you measure on social media and email?

On social media, likes and follower counts are the most visible numbers, and also the least meaningful. Followers are easy to buy, and likes swing with the algorithm. That is why I focus on engagement rate and site traffic for organic social.

• Engagement rate: (Likes plus comments plus saves plus shares) divided by reach or followers.
• Save and share rate: The strongest sign that people found real value in a post.
• Social sessions and conversions: Data that reaches GA4 through UTM tagged links.
• Email click rate: People who clicked divided by emails delivered.
• Unsubscribe rate: People who unsubscribed divided by emails delivered.

I no longer trust email open rate as much as I used to. The reason: some email clients preload content for privacy reasons, which can inflate opens. Consequently, click rate and conversions from email are far more reliable signals.

You can check an account's engagement with my Instagram engagement rate tool. For tagging social links correctly, the UTM builder does the job. Untagged links often land in the wrong channel in GA4, and then your channel KPIs lie to you.

Website experience KPIs: Core Web Vitals and engagement

No matter how good your ads and SEO are, traffic that lands on a slow or jumpy site goes to waste. For this reason, website experience belongs in your KPI set too. Google's Core Web Vitals are also a sensible starting point.

The Core Web Vitals guide on web.dev lists three metrics with these good thresholds. LCP (Largest Contentful Paint) should be 2.5 seconds or less. INP (Interaction to Next Paint) should be 200 milliseconds or less. CLS (Cumulative Layout Shift) should be 0.1 or less. Google recommends judging them at the 75th percentile of page loads, separately for mobile and desktop.

These metrics are KPIs for the technical team and a health check for the marketing team. In other words, you do not need to debate them in every weekly marketing meeting. That said, if mobile conversion rate suddenly drops, this is one of the first places to look.

If your site structure holds conversions back, my web design service treats speed and conversion as one job.

A KPI table by channel and funnel stage

The table below pulls the indicators above into one view. Each row then shows the channel, the stage, the main KPI and its formula. Pick the rows that fit your business and leave the rest for page two.

Notice one detail while you read the table. In short, the same channel deserves different KPIs at different stages. SEO, for example, serves both consideration and conversion, so it appears in two rows.

How do you set targets for digital marketing KPIs?

Choosing indicators is half the work. The other half is giving each one a realistic target. When I set targets, I draw on three sources: the business goal, historical data and unit economics.

1. Start from the business goal: How many sales or new clients do you want this quarter?
2. Check historical data: What was the conversion rate at each stage over the last 6 to 12 months?
3. Work out unit economics: What is the most you can spend to win one customer?
4. Spread the target backward: Move from the sales target down to lead, click and budget targets.
5. Add checkpoints: Review targets monthly and quarterly, not weekly.

I do not recommend adopting industry averages as targets. The reason is that averages blend different prices, brand strength and tracking setups into one number. Your own history is worth more than any average you find online.

If you have no history, treat the first three months as a measurement period. During that time you set a baseline, not a target. Then you build the target on top of that baseline.

How does a backward target calculation work in practice?

The numbers below are an illustrative calculation only. They do not come from any client, because the point is the method.

Say a service firm wants 30 new clients this quarter. Historically, 25% of proposals turn into sales, so it needs 120 proposals. Next, if half of qualified leads reach the proposal stage, it needs 240 qualified leads. If 60% of incoming leads qualify, the target becomes 400 leads in total.

Now move to the ad side. With a 4% site conversion rate, 400 leads then need 10,000 clicks. At an average CPC of 3 US dollars, the quarterly budget comes to about 30,000 US dollars. Cost per client is then 30,000 divided by 30, which is 1,000 US dollars.

The last step matters most: is that cost acceptable? If one client leaves several times that amount in gross profit over their lifetime, the plan holds up. If not, you either raise the conversion rate or shrink the target. My percentage calculator helps with the quick math.

What are a north star metric and leading indicators?

A north star metric is the single indicator that best sums up long term value for the business. In e-commerce it is usually monthly net revenue or active customers. For a service firm, on the other hand, the best choice may be signed contract value.

However, a north star metric moves slowly. For example, SEO work can take months to show up in revenue. So you need leading indicators next to it. A leading indicator is an early step that predicts the result.

• Leading indicator for revenue: add to cart rate and checkout reach rate.
• For organic growth: impression growth on new queries.
• Service sales: watch pricing page visits and qualified leads.
• For loyalty, the lead signal is days until the second order.

In practice I tell teams this: talk about leading indicators in the weekly meeting and the north star in the monthly one. That way you avoid panic moves over weekly noise, yet you still spot problems early.

Which KPIs can mislead you?

Some indicators look great while the business gets worse. I call them vanity metrics. The problem is not the numbers themselves; instead, it is reading them without context.

• Total traffic: Irrelevant visits grow the number but not the sales.
• Follower count: Followers who never engage can even reduce your reach.
• Platform ROAS: Each platform tends to claim the sale for itself, so the sum can look inflated.
• Low CPL: Cheap leads without a quality check only push the cost further down the pipeline.
• Average position: A few branded queries can make the average look better than reality.

Be especially careful with platform ROAS. That is because Google, Meta and others can each claim the same sale. Because of this, every month I compare the sum of platform reports with the real order system. When the gap is large, I trust the order system rather than the platforms.

If you want to see how organic and paid channels affect each other, read my post on organic growth or paid ads.

How do digital marketing KPIs change by business type?

In practice, every business picks a different set of digital marketing KPIs from the same catalog. Below I share my starting set for the three business types I work with most. These are starting points based on field experience, not guarantees, and you should adjust them to your own data.

For e-commerce, five indicators are enough: ROAS, average order value, conversion rate, repeat purchase rate and CLV to CAC. During sale periods in particular, read ROAS together with margin.

However, B2B service firms look different. Here qualified leads, proposal rate, close rate, cost per client and sales cycle length lead the way. Since data arrives slowly, track micro conversions as well.

Local businesses such as clinics, restaurants or repair services get more from phone calls, direction requests and map visibility. I covered that topic in how to do Google Maps SEO. To plan paid social spend around these goals, see my guide on calculating a social media advertising budget.

How do you make KPI definitions clear across the team?

Above all, the same word means different things to different people. Marketing may say "conversion" and mean a form fill, while sales means a signed contract. That is why I suggest a short definition card for every KPI.

• Name and a one sentence definition.
• Formula and data source.
• Owner and update frequency.
• Target value and the reasoning behind it.

Keep these cards in one shared document and ask every new team member to read it first. Then meetings discuss the decision a number calls for, not whether the number is right. In my experience this small habit ends most reporting arguments before they start.

Which tools do you need to track KPIs?

The good news: you do not need expensive software for a core KPI set. In fact, free tools cover most of the work. The bad news: connecting them takes time and care.

• Google Analytics 4: Sessions, engagement and key event tracking.
• Search Console: Organic impressions, clicks and query data.
• Google Ads and Meta ads: Spend, impressions and platform conversions.
• CRM or order system: Qualified leads, sales and real revenue.
• Looker Studio or a spreadsheet: One screen that brings it all together.

Mind one terminology change in GA4. In its article on key events in GA4, Google now calls the old Analytics "conversions" key events. It keeps the word "conversion" for actions shared with Google Ads to measure campaign performance. As a result, if your reports blur the two, the numbers in both systems will not match and confusion follows.

In my own work the CRM sits at the center. Whatever the ad platform says, the sales record has the final word. This approach also keeps reports simple and arguments short.

How often should you review your KPI set?

A KPI set is not something you build once and forget. As the business grows, the product changes or a new channel opens, the indicators need updates too. So I suggest three rhythms.

On a weekly rhythm, look only at leading indicators and spend. Put simply, the aim is to catch anomalies, not to change strategy. On a monthly rhythm, compare conversion and cost indicators with targets. On a quarterly rhythm, question the KPI set itself: are you still measuring the right things?

For instance, awareness KPIs come first for a brand new business. A few years later, however, retention and CLV take the lead. Keeping the same set for years means running a changing business with an old map.

Also, log every change in writing. If you do not know when you changed a definition, you will draw the wrong conclusion when you compare two reports six months later. Consequently, a simple change log solves this completely.

Where should you start?

After this many indicators you may wonder where to begin. My advice: this week, pick one business goal and sketch the funnel that leads to it. Then write down one KPI for each stage.

Next, check that the data behind those four indicators is actually correct. Are key events set up properly in GA4? Does the ad account receive conversions correctly? Is the source field filled in your CRM? After all, if tracking is broken, even the best set of digital marketing KPIs produces bad decisions.

Finally, treat the first month as a baseline and set the target in month two. In my experience, this plain approach beats a complex dashboard with hundreds of numbers.

If you would like to build this setup together, write to me through the contact page. I will review your current tracking and tell you plainly which indicators are missing or misleading.