For the past few years, the dominant narrative in digital analytics has been clear: browser-based tracking is fundamentally broken. Between ad blockers, browser privacy features, consent banners, and network conditions, we’re told that a large share of traffic — and even conversions — is simply invisible.
Depending on who you ask, estimates of lost data range anywhere from 30% to 70% or more. These figures are often repeated, rarely questioned, and frequently used to justify increasingly complex tracking stacks.
At Able CDP, we wanted to ground this discussion in something more concrete.
Instead of estimating tracking loss indirectly, we asked a simpler question:
When a real purchase or signup happens, how often did a browser tracking event actually fire?
To answer that, we ran a study across our customer base using backend conversion data as ground truth.
What we found surprised us — and it meaningfully reframes the conversation about tracking loss.
How the study works
Able CDP tracks browser events at key conversion points such as:
-
Checkout completion
-
Account sign-ups
- Lead form completions
At the same time, we ingest authoritative conversion data directly from backend systems, including:
-
WooCommerce and Shopify
- Other e-commerce platforms and CRM systems
This gives us a rare advantage: for every confirmed purchase, we can check whether a corresponding browser event was recorded.
For more details about how we gather the data and how we collect server-side conversions to browser events, see our page about server-side tracking, as well as a more technical description in our documentation.
Importantly, our browser tracking setup is deliberately unremarkable:
-
A standard JavaScript tracking script
-
Loaded from a third-party domain (ablecdp.com)
-
Sending data to a third-party server
-
No fingerprinting
-
No server-side event recovery
-
No browser-specific workarounds
In other words, this is not a “best-case” or exotic tracking implementation. It’s representative of how many analytics tools operate today.
The results
Across all study instances, covering non-recurring purchases in online funnels that had 100 or more purchases during the full first week of 2026:
-
Average browser coverage: 96% of purchases had a corresponding browser event
-
Median: also 96%
-
Observed range: 92% – 99%
Put differently, only 1–8% of confirmed purchases lacked a browser tracking event.
That result alone runs counter to the idea that a large fraction of conversions are disappearing due to browser tracking prevention.
But the more interesting question is why the remaining gap exists.
What explains the missing 1–8%?
When we looked closer at the variance between customers, a clear pattern emerged:
Sites with better performance and cleaner implementations consistently tracked closer to 99% of conversions.
Sites at the lower end of the range (92–95%) tended to share characteristics such as:
-
Slower page loads
-
Heavier checkout flows
-
More complex client-side logic
-
Mobile-heavy traffic on weaker connections
This strongly suggests that most missing browser events are caused by technical execution issues, such as:
-
Tracking scripts not loading in time before navigation or redirect
- Heavy tracking scripts (like GTM) and website media assets blocking loading of analytics and ad platforms scripts
-
Network latency on slow or unstable connections
-
Script errors such as race conditions between form submission and script execution
In short: the dominant failure mode appears to be performance, not privacy blocking.
If widespread tracking prevention were the primary cause, we would expect to see:
-
Much lower averages
-
Larger variance
-
Sharp drops correlated with specific browsers or regions
We did not observe that.
Reconciling this with the “Tracking Is Broken” narrative
At first glance, these results may seem to contradict the broader industry narrative that browser-based tracking is increasingly unreliable. However, that apparent contradiction largely disappears once we are precise about what is being measured.
What this study does — and does not — measure
This analysis focuses exclusively on conversion events (checkouts and sign-ups) and compares browser-tracked events to authoritative backend conversion data. It does not measure:
-
Pageview or session-level tracking loss
-
Funnel completeness or user journeys
-
Attribution accuracy across channels
-
Consent acceptance or rejection rates
As a result, this study does not claim that tracking loss is concentrated at the pageview level, nor does it make assumptions about user intent or behavior earlier in the funnel.
What it does show is narrower, but important:
When a confirmed conversion occurs, a corresponding browser event is present in the vast majority of cases.
This implies that large headline figures about tracking loss should not be assumed to apply uniformly across all event types. Conversion-level tracking behaves differently from traffic-level measurement, and the two should not be conflated.
Our study quantifies this distinction, without making claims beyond what the data directly supports.
The tech audience skew
One other explanation to consider is that a lot of the most-cited research comes from tech-focused sites.
For example, there's a study alleging that 58% of The Hacker News users are blocking Google Analytics. Putting a questionable methodology aside (it wasn't measured at The Hacker News), that's a site whose audience is literally developers and tech enthusiasts—exactly the people most likely to run ad blockers, use Firefox or Brave, and have strong opinions about tracking.
If you're selling to that audience, those numbers may be relevant. (Although, the highest we've ever observed for a niche tech audience was about 15% of traffic being blocked.) If you're running a typical e-commerce store, they're not. Our customer base skews toward mainstream SaaS and e-commerce—business software, consumer goods, subscriptions—and the blocking rates reflect that.
What this means for analytics and CDPs
The most important implication of this study is not whether 4% or 8% of browser events are missing — it’s how those missing events are treated.
1. Missing conversions vs unknown conversions
There is a critical difference between:
-
Not tracking a conversion at all, and
-
Tracking 100% of conversions, while knowing that some lack browser context
Most analytics stacks only observe what happens in the browser. When a conversion event fails to fire, that conversion simply disappears from the dataset.
By contrast, Able CDP ingests conversions directly from backend systems like Stripe and WooCommerce. This means:
-
100% of confirmed conversions are present
-
Browser events are used as context, not as the source of truth
-
The remaining 1–8% are explicitly visible as conversions without browser attribution, rather than silently missing
This distinction is fundamental for downstream analysis.
2. Attribution uncertainty is more important than raw counts
From a decision-making perspective, the real problem is rarely that a small percentage of conversions are missing sources.
The bigger issue is that:
-
There's uncertainly whether conversion didn't occur or occurred but wasn't attributed
-
The uncertainty is unevenly distributed across channels
-
Traditional reports often hide this uncertainty behind partial data
Knowing which conversions lack browser context — and how many — is more actionable than assuming browser data is complete.
3. Ground truth changes how tracking loss is interpreted
When backend conversions are treated as ground truth:
-
Browser tracking gaps become measurable instead of speculative
-
Differences between sites reflect implementation and performance quality
-
Tracking loss can be reasoned about quantitatively, not rhetorically
This shifts the role of a CDP from "recovering lost data" to making uncertainty explicit and manageable. For example, Able CDP's built-in tracking and attribution reports clearly indicate which conversions don't have a known source, either because no browser tracking data exist or because cookie consent wasn't given and browsing path couldn't have been followed.
A more precise way to talk about tracking loss
Based on this study, we believe the industry conversation would benefit from more precision.
A more accurate framing is:
Browser-based tracking may be getting increasingly unreliable for traffic measurement and user journeys, but conversion-level browser tracking remains largely intact, with most observed losses attributable to technical and performance factors rather than widespread blocking.
This doesn’t minimize privacy changes or regulatory pressure. It simply reflects what the data actually shows when measured against a reliable ground truth.
Why we’re sharing this
We’re publishing these findings because exaggerated or imprecise claims about tracking loss:
-
Push teams toward unnecessarily invasive techniques of questionable legality
-
Distract from solvable engineering problems
-
Create confusion about what data can still be trusted
Our goal with Able CDP is not to “outsmart” browsers or users, but to help teams build clear, honest, and resilient measurement systems.
This study is one small step toward a more evidence-based conversation about what’s really broken — and what isn’t.
If you’re interested in the methodology, edge cases, or running a similar analysis on your own data, we’re always happy to compare notes.