Most businesses have analytics installed and still fly blind — because they're watching the wrong numbers. They see traffic going up and feel good, while having no idea whether any of those visitors became customers. The problem isn't a lack of data; it's that the seductive, feel-good metrics drown out the few that actually tell you if your website is working. Let's fix that: the metrics that matter, the vanity ones to ignore, and how to set up analytics that drive decisions instead of decorating a dashboard.
The test for whether a metric matters
Here's the one rule that cuts through everything: a metric matters only if a change in it would change what you do. If your pageviews doubled, would you do anything differently? Probably not. If your conversion rate or cost-per-lead doubled, you absolutely would. That test instantly sorts the signal from the noise — and most dashboards are mostly noise.
Metrics that matter vs vanity metrics
Here's the sort, side by side. One column drives decisions; the other makes you feel good and tells you nothing:
Metrics that matter vs vanity metrics
One drives decisions. The other makes you feel good and tells you nothing.
Metrics that matter
Conversion rate
The % of visitors who take your key action (lead, sale, booking). The single most important number — it tells you if the site actually works.
Cost per lead / acquisition
What it costs to get a lead or customer from each channel. Tells you where to spend and where you're wasting money.
Traffic by source & quality
Not just how many visitors, but which channels send people who convert. 100 buyers beat 10,000 bouncers.
Top converting pages & paths
Which pages and journeys actually produce leads or sales — so you double down on what works and fix what leaks.
Vanity metrics
Raw pageviews / 'hits'
A big number that feels good and means little on its own. Traffic without conversion is just expensive noise.
Total followers
Followers aren't customers. A huge audience that never buys is a vanity trophy, not a business result.
Time on site (alone)
High time-on-site can mean engaged — or confused and lost. Meaningless without conversion context.
Bounce rate, misread
A 'high bounce' on a page that answered the question fully can be fine. Judged in isolation, it misleads.
The pattern: the metrics that matter all connect to money or decisions — conversions, cost, quality, what's working. The vanity metrics are big numbers disconnected from outcomes. Both go up and to the right when things are going well, but only one tells you why or what to do next.
Conversion rate: the number to obsess over
If you track one thing, track conversion rate — the percentage of visitors who take your key action (lead, sale, booking). It's the single most important number because it tells you whether the site actually works, and because it compounds with everything else: improving it makes every visitor, every ad dollar, and every SEO effort more valuable at once. Most business sites convert 1–3%; going from 1.5% to 2.5% is a 67% lift in results from the same traffic. That's why we treat conversion as the master metric — and why the CRO checklist and why websites don't convert exist. Chasing more traffic while ignoring conversion is filling a leaky bucket.
Cost per lead and traffic quality
Two more that separate winners from guessers:
- Cost per lead / acquisition by channel. What it actually costs to get a lead or customer from each source. This tells you where to spend more and where you're burning money — and it's impossible to know without it.
- Traffic quality, not just quantity. A channel sending 10,000 visitors who bounce is worth less than one sending 100 who buy. Look at which sources send people who convert, and weight your effort there.
Together, these turn "we're getting traffic" into "we know which traffic makes us money" — a completely different basis for decisions.
Why vanity metrics are dangerous
Vanity metrics aren't just useless — they're actively risky, because they tempt you to optimize for them. Chase raw pageviews and you'll produce traffic that doesn't convert. Chase follower count and you'll build an audience that never buys. Misread bounce rate and you'll "fix" pages that were working fine. The feel-good number going up gives a false sense of progress while the business doesn't grow. The antidote is discipline: judge every metric by the test above, and ignore the ones that wouldn't change a decision.
Setting up analytics that actually help
The tool matters less than the setup. Most businesses are fine with Google Analytics 4 (free), or a privacy-friendly option like Plausible if you prefer simpler, cookie-less analytics. But an analytics tool with no conversion tracking just measures vanity. So the one essential setup task:
- Define your key actions — form submit, purchase, booking, call.
- Track each as a conversion in your analytics tool.
- Feed conversions back to your ad platforms (Google, Meta) so they optimize toward real outcomes, not clicks.
Do that and your analytics start answering the only questions that matter: is the site working, where do customers come from, and what should I change? (Page speed and Core Web Vitals belong here too — PageSpeed Insights — because speed quietly affects all of the above.)
Look monthly, decide, repeat
Finally, cadence. Check the metrics that matter monthly (weekly if you're actively testing), and resist daily check-ins — daily data is mostly noise that breeds anxiety and knee-jerk reactions. Analytics is for decisions, not surveillance. A focused monthly review of conversion rate, cost per lead, traffic quality, and top pages beats a daily glance at a wall of vanity numbers every time.
Want analytics that tell you what to do, not just what happened? Explore our analytics service and business website builds, see the results we've driven, or book a call — we'll set up conversion tracking and a simple dashboard of the numbers that actually matter for your business, and cut the vanity noise.
The goal of analytics was never a prettier dashboard. It's better decisions. Track the handful of metrics that connect to money, ignore the ones that just feel good, and you'll finally know whether your website is working — and exactly what to do about it.
FAQ
Questions, answered.
What businesses ask about measuring their website.
The ones tied to money and decisions: conversion rate (the % of visitors who take your key action), cost per lead or acquisition by channel, traffic quality by source (which channels send people who actually convert, not just visit), and your top converting pages and paths. Page speed / Core Web Vitals matter too because they affect all of the above. These tell you whether the site works, where to spend, and what to fix. Everything else is context at best and distraction at worst — track what connects to leads, sales, and decisions.
Vanity metrics are numbers that look impressive but don't inform a decision — raw pageviews, total followers, generic 'hits', time-on-site in isolation. They're not 'bad' exactly; they're seductive. They make you feel good (the number went up!) while telling you nothing about whether you're getting more customers. The danger is optimizing for them — chasing traffic or followers that never convert — instead of the metrics that actually move revenue. A metric is only useful if a change in it would change what you do; if it wouldn't, it's vanity.
It varies by industry and traffic type, but most business sites convert 1–3% of visitors into leads or sales, good ones 3–5%, and well-optimized ones higher. Rather than chasing a universal benchmark, measure your own baseline and improve it — going from 1.5% to 2.5% is a 67% increase in results from the same traffic. Conversion rate is the metric to obsess over because it compounds with everything else: improving it makes every visitor, every ad dollar, and every SEO effort worth more. Compare yourself to your past, not a generic number.
For most businesses, Google Analytics 4 (free) covers the essentials — traffic, sources, conversions, and behaviour — especially once you set up conversion tracking for your key actions. Privacy-friendly alternatives (like Plausible or Fathom) are simpler and lighter if you want cleaner, cookie-less analytics. Product-led businesses might add a product-analytics tool (like PostHog) for deeper funnel analysis. The tool matters less than the setup: an analytics tool with no conversion tracking just measures vanity metrics. Pick one, then configure it to track what actually matters.
Define your key actions first (form submit, purchase, booking, call), then configure your analytics tool to record each as a conversion event — in GA4 these are 'events' you mark as key conversions, often fired when someone reaches a thank-you page or clicks a call/submit. Connect the same conversions back to your ad platforms (Google, Meta) so they can optimize toward real outcomes. Without conversion tracking you're flying blind — you can see traffic but not whether it does anything. This setup is the single highest-value analytics task; do it before worrying about anything else.
Not necessarily — it depends on the page's job. A high bounce rate on a page that fully answered the visitor's question (say, a contact page with your phone number, or a blog post that resolved their query) can be perfectly fine. A high bounce on a landing page meant to drive a next step is a problem. Bounce rate is meaningless in isolation; it only means something in the context of what the page was supposed to do. This is exactly why raw bounce rate is closer to a vanity metric — judged alone, it misleads as often as it informs.
Set a regular, low-frequency cadence — monthly for most small businesses, weekly if you're actively running campaigns or testing changes — and resist daily check-ins, which mostly produce noise and anxiety from normal fluctuation. The goal of analytics is decisions, not surveillance: look often enough to spot real trends and the impact of changes you've made, not so often that you react to random daily wiggles. A focused monthly review of the metrics that matter beats a daily glance at a dashboard of vanity numbers every time.
Yes — but you need the *right* analytics, not a complicated dashboard. At minimum, track your conversions (leads, sales, bookings) and where they come from, so you know what's working and where to spend. Without that, you're guessing — pouring effort into channels that might do nothing and unable to tell if changes help. The setup is free and modest; the payoff is making decisions on evidence instead of hunches. You don't need enterprise analytics; you need conversion tracking and the discipline to look at the few numbers that matter.