But here’s what those numbers aren’t telling you: how many people tried to use your website and couldn’t. How many potential donors gave up halfway through your form. How many people needing crisis support left because they couldn’t find your phone number.

Analytics tools like Google Analytics are brilliant at measuring what happens. They’re terrible at measuring what doesn’t happen.

The people who don’t appear in your data

When someone visits your website and immediately leaves because they can't navigate it with their screen reader, that registers as a bounce. Same as someone who clicked the wrong link by accident.

When someone tries to complete your donation form but can't because it doesn't work with their keyboard, they might show up in your analytics as "time on page". You might even think they were engaged with your content.

When someone on a slow mobile connection waits for your page to load, then gives up and closes the tab, they probably won't register at all. The page didn't finish loading, so the analytics script never fired.

Your analytics show you the people who could use your website. They don't show you the people who couldn't.

What "good" conversion rates might be hiding

Let's say your donation page has a 2% conversion rate. Two out of every 100 visitors complete a donation.

Is that good? Bad? Average?

Here's the more important question: what if it should be 4% or more?

What if half the people who want to donate can't actually complete your form because error messages don't explain what's wrong, required fields aren't marked clearly, the form times out before they finish, the submit button doesn't work on their mobile device, or their screen reader can't navigate the fields?

You'd never know from your analytics. You'd just see a 2% conversion rate and assume that's normal.

Through our Impact Optimiser assessments with 95 charities in 2025, we repeatedly found donation forms with barriers that prevented completion. The charities had no idea - their analytics just showed them the successful donations, not the attempted ones.

The mobile data deception

Your analytics might tell you that 60% of your traffic comes from mobile devices. That sounds like your mobile experience is working.

But consider this: what if those mobile visitors are spending less time on your site than desktop visitors, bouncing more frequently, and never completing key actions like donations or contact forms?

Your analytics will show you these patterns if you dig into the data. What they won't tell you is why. They won't tell you that your text is too small to read without zooming, that your buttons are too close together to tap accurately, that your forms require constant pinching and scrolling, or that your navigation doesn't work properly with touch.

Time on page: the misleading metric

Someone spending five minutes on your "How to access our services" page - that looks good, right? They must be really engaged with your content.

Or maybe they're struggling to understand overly complex information, looking for a phone number that's buried in paragraphs of text, trying to work out how to actually access your services from a wall of policy information, or re-reading content multiple times because it's not written clearly.

Someone spending 30 seconds on the same page might have found your phone number immediately and called, downloaded your simple clear PDF guide, or got the answer they needed from well-structured content.

Analytics can't tell you the difference between productive time and frustrated time.

The silent majority who just leave

The most dangerous thing about analytics is what they normalise.

You see patterns - certain pages have high bounce rates, certain forms have low completion rates, certain user flows see lots of drop-off. You start to accept these as normal.

But what if they're not normal? What if they're symptoms of serious problems?

When someone can't use your website, they typically don't fill out a feedback form explaining the problem, email you to complain, call to tell you your site doesn't work, or leave a review saying they couldn't access your services.

They just leave. Silently. And your analytics record them as just another bounced visitor.

What you should actually be measuring

Analytics become useful when you start measuring the right things.

Instead of page views, track successful completions. Instead of time on page, measure task completion rates. Instead of just traffic sources, see which sources actually convert. Instead of bounce rates alone, identify where people exit on critical journeys.

The most revealing metric is often comparison data. If your mobile conversion rate is half your desktop rate, you don't have different audiences - you have a mobile usability problem. If people consistently drop off at the same point in a form, that step has a barrier. If returning visitors behave identically to new ones, they're not building familiarity - they're encountering the same problems again.

The data you’re not collecting

Standard analytics don't capture accessibility tool usage like screen readers, keyboard navigation, or voice control. They don't track attempts to complete actions that failed, error states and form validation failures, page load failures on slow connections, or interactive element failures when buttons don't work or forms don't submit.

You need different tools to find these problems - accessibility testing, user testing, form analytics, error tracking.

This is exactly what our Impact Optimiser assessment does. It's a free, no-obligation review that identifies the barriers your analytics can't see - the ones preventing people from accessing your services or completing donations. We look at your website through the lens of the people you serve and show you what's actually stopping them.

Why your analytics are lying to you