Marketing Metrics Glossary for Beginners

#01Search metrics

Search Console's Performance reports show four core metrics (Google):

Fictional example: 2,000 impressions and 60 clicks give a CTR of 3%. Google notes that for some result types a link must be scrolled or expanded into view to count as an impression (Google). Google also says appearances in AI Overviews and AI Mode are counted in the Performance report under the "Web" search type (Google), so they aren't shown separately there.

#02Website analytics metrics

In Google Analytics 4 (Google):

Fictional example: 1,000 sessions, 620 of them engaged, give a 62% engagement rate and a 38% bounce rate.

Key event: an event you mark as especially important to your business, such as a form submission or purchase. See Google Analytics 4 basics and Google's key events help.

Channel group: a rule-based grouping of traffic sources such as Organic Search or Referral (Google). See tracking AI traffic in GA4.

Attribution model: the rule for giving credit for a key event to the touchpoints before it (Google). See UTM parameters and attribution basics.

#03Business and campaign metrics

These are standard formulas used across marketing; check how your company defines each before reporting.

Our view: the most common beginner error is mixing time periods, for example taking spend from one month and customers from another.

#04AI visibility metrics

There's no official standard for these yet. These are the definitions MentionX Learn uses; always state yours when reporting.

Fictional example: across 50 prompts run in one assistant, you're mentioned in 15 answers (30% mention rate) and cited in 6 (12% citation rate). If competitors get 35 mentions in total, your share of voice is 15 ÷ 50 = 30%.

Read the glossary entries for mention rate, recommendation rate and AI share of voice, and the guide to measuring AI share of voice.

#05How should beginners report metrics?

Our view:

• Pair a volume with a rate. "60 clicks at 3% CTR" says more than either alone.
• Always give the period and comparison. "Up 12% vs September" beats a bare number.
• Write the definition once in your report so readers know what each metric means, and keep it unchanged from month to month so trends stay comparable.
• Don't over-read small samples. AI answers vary between runs, so a few prompts can swing a rate a lot.

#06Where next?

Read spreadsheets for marketers and use the executive GEO dashboard template.

#07From MentionX data: AI visibility metrics in practiceOriginal data

Method. Read-only aggregate queries over MentionX scans from 22 August to 29 October 2026 (UTC): 2,217 engine runs (2,154 completed) across 18 anonymised projects and six engines.

Here's what two AI visibility metrics look like in practice. In the MentionX sample, by prompt intent:

Our view: "Brand mentioned" is a mention rate: the share of answers naming a brand. Share of voice compares a brand's prominence with others in the same answers. Both depend heavily on which prompts you choose.

Limits. A convenience sample of mostly B2B software projects, run through engine APIs, which can behave differently from the consumer apps. Perplexity, Copilot and DeepSeek figures come from a single day (26 October 2026), so treat them as a snapshot, not a trend.

#08Worked example: Ledgerline's first metrics report

Ledgerline is a fictional accounts-payable tool. Figures are illustrative.

A new Ledgerline marketer prepares her first monthly report. She picks five metrics: organic sessions, demo requests, conversion rate, AI mention rate and cited pages. Each comes with last month's figure, this month's and a one-line explanation of the change.

Her manager says it's the clearest report they've had. The figures in her report are fictional, but the format is worth copying.

#09Which email and social metrics should beginners know?

Open rate is the share of delivered emails that were opened, though privacy features in some email apps make it unreliable. Click-through rate is the share of recipients who clicked a link. Unsubscribe rate shows how many people opted out. On social platforms, impressions count how often a post was shown, reach counts unique people and engagement rate divides interactions by impressions or followers. Our view: engagement metrics are useful for comparing posts, but they rarely prove business impact alone.

#10Which paid media metrics come up most?

Cost per click (CPC) is spend divided by clicks. Cost per thousand impressions (CPM) is the cost of showing an ad 1,000 times. Cost per acquisition (CPA) is spend divided by conversions. Return on ad spend (ROAS) is revenue from ads divided by ad spend. Always check how a platform defines a conversion before comparing figures.

#11What are leading and lagging indicators?

Leading indicators move early and suggest future results, such as mention rate in AI answers, demo requests or trial sign-ups. Lagging indicators confirm results later, such as revenue or retention. A good report shows both, so you can act early and still prove outcomes.

#12What do sample size and variance mean?

Sample size is how many observations a metric is based on. Variance describes how much results differ between observations. Small samples and high variance make figures unreliable. AI answers vary between runs, so AI visibility metrics need repeated runs before you trust a change.

#13What is attribution?

Attribution assigns credit for a conversion to the channels a buyer touched. Last-click attribution gives all credit to the final click; multi-touch models spread it. AI answers that don't send a click are hard to attribute, so add self-reported attribution, such as a form question asking how people found you.

#14What is a baseline?

A baseline is the starting value of a metric before a change. Without one, you can't tell whether a change worked. Record baselines before launching campaigns or fixes.

#15What are vanity metrics?

Metrics that look impressive but don't connect to goals, such as follower counts or raw page views without context. They aren't useless, but don't report them as success on their own.

#16How do you read a percentage change correctly?

A rise from 2% to 3% is a one-point increase but a 50% relative increase. Say which you mean. Show the underlying numbers so readers can judge the size of the change.

#17What are cohort metrics?

A cohort groups people by a shared start point, such as the month they signed up. Cohort retention shows what share of each group is still active later. Cohorts show trends that averages hide.

#18What should a metric definition include?

What's counted, how it's calculated, the data source, the time zone and any exclusions. Write definitions once and reuse them in every report.

#19What is conversion rate, and how do you read it?

Conversion rate is the share of visitors or users who complete a goal, such as a sign-up or purchase. It depends heavily on how you define both the goal and the audience. A landing page with a 10% conversion rate from a narrow, targeted audience isn't necessarily better than one with 2% from a broad audience. Always state the denominator. Our view: conversion rate is most useful for comparing versions of the same page over time.

#20What are customer acquisition cost and lifetime value?

Customer acquisition cost (CAC) is total sales and marketing spend divided by new customers in the same period. Customer lifetime value (LTV) estimates the revenue a customer brings over the relationship. The ratio between them shows whether growth is sustainable. Both rely on assumptions, so explain how you calculated them.

#21What are bounce rate and engagement rate?

Bounce rate traditionally counted single-page sessions. Google Analytics 4 instead reports engagement rate, the share of sessions that lasted a set time, viewed more than one page or triggered a conversion, and defines bounce rate as its opposite. Check which definition your tool uses before comparing numbers across reports.

#22What is citation rate in AI answers?

Citation rate is the share of answers that link to or reference your pages. It differs from mention rate, because an answer can name your brand without citing your site, or cite your page without naming you. Tracking both gives a fuller picture. As the MentionX table shows, engines also differ in how often they show citations at all.

#23How do you choose metrics for a new campaign?

Pick one primary metric tied to the campaign's goal, two or three supporting metrics and a baseline for each. Decide before launch what result would count as success. That stops you choosing the flattering metric afterwards.

#24How do you spot misleading metrics?

Watch for missing denominators, changed definitions, tiny samples, cherry-picked date ranges and averages hiding big differences between groups. Ask how a number was calculated before acting on it.

#25How should you learn these metrics?

Pick the five metrics your team reports most, find where each comes from in your tools and recalculate one yourself. Doing it once by hand makes the definitions stick and shows you where numbers can go wrong. Then explain it to a colleague in two sentences.

#26Common questions

Which metrics matter most? The ones tied to your goals, usually leads or revenue plus a few leading indicators.

What is share of voice? A measure of how prominent your brand is compared with others in the same results.

Is traffic still important? Yes, but AI answers can influence buyers without a click.

How many metrics should I report? A handful, each with context.