10 Ways to Get Better Advertising Data (and Why You Need Them)

Where do the gaps in your advertising data come from?

Just a few years ago, marketers could assume that the data in Google Analytics, Google Ads, or other advertising tools reflected reality with reasonably good accuracy. Today, that assumption is increasingly risky. The tools still collect data — but a growing share of traffic and user behavior simply slips through the cracks, resulting not only in gaps in reports but also in flawed decisions about advertising spend.

There are several reasons for this, and each one operates independently.

Users don’t stick to a single device. Someone clicks an ad on their phone on the way to work, then completes the purchase that evening on a laptop. Traditional cookie-based tracking can’t connect these two visits into a single journey. For Google Analytics, these are two different users; for Google Ads, only one of the site visits is linked to the ad click. The result: the conversion is either attributed incorrectly (GA4) or not attributed at all (Google Ads, Meta Ads).

Users are increasingly rejecting cookies. Since GDPR came into effect in 2018, every website has had to ask for tracking consent — and a significant share of visitors don’t give it. In theory, this is good news for privacy; in practice, it’s another data gap. Research from data.rocks shows that cookie banner rejections often result in roughly 30% of sessions going unrecorded in GA4.

Ad blockers block analytics too. It’s estimated that over 30% of users in Poland use an ad blocker (more information ➡️), particularly among younger age groups and desktop users. Most of these tools block not just ads, but also analytics and tracking scripts — including Google Analytics, the Meta pixel, and Google Ads conversion tags. Part of your traffic is simply invisible to all of these systems at once.

Browsers are increasingly restricting tracking. Apple’s Safari uses a mechanism called ITP (Intelligent Tracking Prevention), which blocks third-party cookies and shortens the lifespan of first-party cookies — often to 7 days, sometimes to just 24 hours. Firefox does the same thing with ETP (Enhanced Tracking Protection), blocking known tracking scripts and cross-site cookies. The effect in both cases is similar: Analytics loses the user’s history, Google Ads can’t attribute a conversion to a click from a week ago, and remarketing systems stop recognizing returning visitors.

The result? The data your marketing decisions are based on may be incomplete — sometimes by tens of percent.

Incomplete data, bad decisions

These problems all boil down to one thing: poor data quality translates directly into poor decision-making. In data analysis, there’s a principle called GIGO — Garbage In, Garbage Out — which states plainly: if you feed a system incorrect or incomplete data, you’ll get flawed conclusions out of it. This applies both to Google Ads algorithms, which optimize campaigns based on conversion signals, and to your own business decisions based on Analytics reports.

If the data says a campaign isn’t performing — when in reality its conversions simply aren’t visible — you might turn it off or cut its budget, losing real revenue. If Analytics isn’t recording half of your visits, it’s hard to tell which channels are actually bringing in customers. Data is the foundation of a marketing strategy — and a foundation full of holes will eventually catch up with the whole company’s results.

How to reduce tracking gaps

The truth is that these problems can’t be fully solved — data quality has declined in recent years, and that trend isn’t likely to reverse. Still, there are many solutions worth implementing, and their effects reinforce one another.

User-provided data

Enhanced conversions (EC) is a Google Ads feature that enriches standard tracking with data provided directly by users — such as an email address or phone number entered in an order form. This data is encrypted (hashed) on the browser side before it reaches Google, so no sensitive information leaves your site in readable form. Google then matches it against signed-in user accounts, allowing a conversion to be attributed to a specific campaign and click — even if cookies were blocked or had expired.

In practice, this means conversions that were previously invisible — because a user switched devices between the click and the purchase — can now be correctly recorded and attributed.

At data.rocks, we’ve been using this feature for years, and we can confidently say it has real potential. Google reports that implementing EC recovers an average of 5% of conversions in text-based campaigns and around 17% on YouTube. Data from data.rocks client accounts shows a typical Search conversion increase of about 2–3%. It’s worth remembering, though, that the feature performs better over time — it reaches its full potential only after a few months, while our analyses covered the period shortly after implementation.

Our best result was a 15% increase in Search conversions. Since text-based campaigns accounted for about 70% of revenue on that account, this translated into real growth. In months when Google Ads generated 1 million PLN in revenue, that meant roughly 100,000 PLN in additional reported revenue — without increasing the ad budget. That volume of data is a genuine boost for campaign algorithms, which perform best with as many signals as possible to analyze — ultimately translating into real performance gains, not just better numbers in reports.

Meta Ads and Microsoft Ads offer comparable solutions, working on the same principle and delivering similar benefits.

Customer Match in Google Ads

Customer Match is a feature that lets you use first-party data to build audience lists in Google Ads. In practice, it involves uploading hashed customer contact data — email addresses, phone numbers, or postal addresses — which Google matches against signed-in user accounts. This lets you target your existing customers directly or exclude them from specific campaigns.

Based on these lists, Google can also build lookalike audiences — reaching new users whose behavior and profile resemble your best customers.

The feature also has an automated version — Customer Match with enhanced conversions. Instead of manually uploading lists, Google builds them automatically based on data collected at each conversion. This means the lists are continuously updated and don’t require regular maintenance from the advertiser.

Offline conversion import in Google Ads

Offline conversion import is particularly useful for lead-generation businesses — anywhere a user fills out a form on the website, but the actual transaction happens off-site: over the phone, by email, or during a meeting with a sales rep. Under standard tracking, Google Ads only sees the moment a lead is captured — it has no idea whether that lead actually turned into a customer. Offline conversion import fills that gap by sending Google Ads information about which clicks actually led to real sales.

In practice, this requires creating a conversion list file containing the GCLID click identifier, which is saved in a cookie the moment a user arrives on the site from an ad. This is the most precise way to link the data, but it relies entirely on cookies — if the user has blocked them, the identifier never gets saved in the first place. Alternatively, matching can be done by hashed email address: if the user provided it in a form and later made a purchase, Google can match that information against its database of signed-in users.

Google has announced the upcoming Data Manager API — a new interface designed to streamline and automate the offline conversion import process. The goal is to connect CRM systems directly to Google Ads, removing the need for manual file uploads. At this point, the feature is still in an early rollout phase — it’s worth keeping an eye on Google’s updates in this area.

Google Consent Mode

Google Consent Mode (CoMo) is a well-known feature among marketers — though few understand exactly how it works under the hood. When a user rejects tracking, Google doesn’t give up on observation entirely: it collects anonymized interaction signals, then — based on data from users who did give consent — models the behavior of those who declined, and attributes it to traffic sources, devices, or campaigns. According to Google, this recovers up to 65% of data that would otherwise be lost.

At data.rocks, we tested CoMo’s effectiveness in practice. The results were promising: thanks to Consent Mode, purchase reporting in GA4 became about 23% more accurate, and session reporting improved by about 31%.

➡️ Read more in our case study:

How Consent Mode Really Works

The feature does have its limitations, though. Modeling only works when GA4 has a sufficiently large data sample — including at least 1,000 daily events with rejected consent over a minimum of 7 consecutive days. In practice, this means sites with fewer than roughly 10,000 monthly users generally won’t benefit from this feature — they simply don’t generate enough data for the modeling to be reliable.

Implementing Google Consent Mode should go hand in hand with properly configuring consent management for other tools as well, such as Meta Ads. A common mistake is registering consents in the wrong order — this can result in Meta Ads conversions not being recorded even when the user did give consent. After implementation, be sure to check each tool’s panel to confirm data is being collected correctly — in particular, make sure data volume hasn’t dropped sharply compared to the period before implementation.

💡 Tip: Check if data modeling is active in your GA4: 

Admin > Data display > Reporting identity

Check if the selected identity is set to “Blended” — if so, Analytics can use modeled data.

Expand the blended identity and check if modeled data is active.

If modeling isn’t working, check your monthly user count in GA4. If it’s approaching 10,000, consider increasing traffic — for example, through reach campaigns on Google Ads or Meta Ads. You might be close to unlocking modeling, and enabling it could mean a significant improvement in data quality.

Optimizing the cookie banner’s design

Did you know that around ⅓ of users click a random button on the cookie banner? These are exactly the people worth fighting for by optimizing the banner’s layout.

Make sure your banner’s layout encourages consent to tracking while still giving users a real choice. In the first view, the banner shouldn’t offer more than one option for rejecting cookies. A common mistake is a “Customize consent” button that, in practice, acts like a rejection button — the user thinks they’re making a choice, when in fact they’re rejecting all consents by default. A well-optimized banner lets users choose individual consents — for example, marketing only — but only in a second-step view.

Example of a non-optimized banner:

Example of an optimized cookie banner:

At data.rocks, we actively study how changing the cookie banner’s layout affects consent rates. One example from a client: before the changes, GA4 recorded 73% of purchases; after the changes, 76%. That’s a modest improvement, but the layout change itself takes only a few minutes, so it’s definitely worth making.

GTM Server-Side

GTM Server-Side (GTM SS) is a solution that moves the collection and transmission of analytics data from the user’s browser to your own server. This is a more technically advanced implementation — we recommend it for companies with access to development resources or a technical agency partner — but the results are proportional to the effort invested.

GTM Server-Side can significantly improve the quality of data sent to analytics tools. It reduces the impact of the Safari and Firefox blocking mechanisms discussed earlier, which can cut available data by as much as 30–40%. Once implemented, data is sent directly from the server, reducing the risk of loss and recovering anywhere from 10% to 25% of events that might otherwise have been blocked.

For example:

• If your website generates 1 million events per month, GTM Server-Side could improve reporting by up to 250,000 additional events.
• For e-commerce campaigns, this can mean more accurate conversion tracking and more precise ad ROI data.

Implementation requires getting your analytics and advertising tags in order, developer support, and covering a monthly server hosting cost (for example, on a platform like Stape, starting from $20/month). It’s both a time and financial investment — worth considering once the scale of your advertising activity justifies the additional cost.

Google Tag Gateway

Google Tag Gateway (GTG) is the younger sibling of GTM Server-Side — a feature that offers some of the same benefits, but is much simpler to implement.

GTG is a free Google feature that lets you route Google tag requests (e.g., GA4, Google Ads) through your own domain. As a result, data that was previously treated as coming from a third-party service becomes first-party data — fully compliant with GDPR. This change in cookie context effectively reduces the likelihood of them being blocked by browsers and ad blockers.

If a website uses a content delivery network (CDN), such as Cloudflare, implementation is usually quite simple: it comes down to connecting GTM with the CDN, making minor tag configuration changes, and verifying that data flow is correct.

According to Google, advertisers who implemented this solution saw an average of 14% more measured conversions. Our own testing shows that results improve over time — so it’s worth evaluating outcomes after a few weeks rather than immediately after implementation.

Meta Conversions API

Meta Conversions API (CAPI) is the server-side counterpart to the Meta pixel — a solution that moves conversion tracking from the user’s browser to the server, similar to what GTM Server-Side does for Google Analytics and Google Ads. Standard Meta tracking relies on the pixel, which runs in the user’s browser and is vulnerable to all the issues discussed earlier: ad blockers, browser restrictions, cookie rejection. Conversions API moves part of this tracking to the server — conversion data is sent directly from your server to Meta, bypassing the browser.

CAPI can be implemented in several ways. The simplest is integration through GTM Server-Side or a direct integration with your e-commerce platform — Shopify, WooCommerce, and Magento all offer native CAPI connections that don’t require developer work.

The benefits are similar to those of enhanced conversions: better attribution of conversions to campaigns, more accurate optimization of Meta’s algorithms, and recovery of events the pixel alone would have missed. Meta recommends using the pixel and CAPI together — the two data sources complement each other, and the system automatically deduplicates events to avoid double-counting events.

According to Meta’s own data, implementing CAPI recovers an average of 10 to 20% of conversions that would otherwise be invisible to the pixel alone.

Connecting online and offline data

Google Analytics is excellent for analyzing on-site user behavior — but for lead-generation businesses, that’s usually only part of the story. A user fills out a contact form, and the actual sale happens days or weeks later, over the phone or by email. Without connecting these two worlds, you can’t tell which campaigns are generating not just leads, but revenue.

The solution is integrating Google Analytics data with your CRM. There are a few ways to do this:

• Using the Measurement Protocol — a GA4 feature for sending offline events directly to Analytics
• Combining data in a data warehouse, such as BigQuery
• Combining both methods, which offers the greatest flexibility and completeness
• And in the future, via the Data Manager API — Google’s announced tool for direct CRM integration with its ecosystem

None of these methods are immune to cookie rejection — if a user didn’t consent to tracking, linking their GA4 session to a CRM record may be impossible. Even so, the benefits are significant: instead of evaluating campaigns solely by the number of forms filled out, you can analyze their actual impact on sales. That’s a fundamental shift in approach — from measuring activity to measuring results.

➡️ Read our article about the Measurement Protocol:

What is Measurement Protocol, and When Should You Use It in GA4?

Data enrichment

Integrating GA4 with your CRM is just the beginning. The possibilities that come from combining data across sources go much further — and the more complex the business, the more there is to gain. A few examples:

• Optimizing for profit, not revenue. Combining shopping campaign performance data with margin data lets you allocate your budget to where you’re actually making money — not just where you’re selling a lot.
• Weather-dependent campaigns. Integrating advertising data with weather data for the user’s location works well anywhere demand for a product or service depends on weather conditions.
• Customer lifetime value (LTV) analysis. CRM data lets you assess which customers are most valuable over months or years — and optimize campaigns to acquire more customers like them, not just the cheapest conversions.

Much of this data counts as sensitive information that can’t be sent through standard scripts visible in the website’s code. These kinds of analyses typically require GTM Server-Side — which allows you to enrich data safely on the server side — or combining data sources in a warehouse like BigQuery.

Getting started can be demanding: you’ll need a developer, time, and the right infrastructure. The scope of work depends heavily on which data you want to analyze. This approach is especially recommended for companies with more advanced analytics needs that want to move beyond standard reporting and base their decisions on a complete view of the business.

Don’t forget the basics — configuration errors

All of these solutions only pay off if the basics are already covered. Before implementing advanced features, make sure your tools are working correctly — configuration errors can wreak real havoc on your data and cancel out the benefits of even the best-planned initiatives. Here are a few examples of situations that can lead to data loss:

• A website that loads too slowly — if a page is overloaded with scripts, analytics tools may not have time to collect data before the user leaves. For one of our clients, this led to the loss of as much as 26% of traffic.
  ➡️ Read the full story: https://data.rocks/en/blog/session-to-click-ratio/ 
• Consent errors — incorrect cookie banner configuration may lead to advertising tools like Meta Ads collecting no data at all, even when the user did give consent.
• Issues in purchase tracking — if the order confirmation page is shown after the user returns from an external payment page, some purchases may go unrecorded entirely.
• Incorrect UTM traffic source tags — misconfigured tracking links mean you don’t actually know where your users are coming from.

💡 Tip: Make it a habit to double-check your Google Analytics data — especially after a site update, a change to your store settings, or a plugin change. These are the moments when errors tend to creep in, since analytics tools are closely tied to how the rest of the site runs. If you’re not confident your data is accurate, it may be worth bringing in specialists for an audit. 

Which Solution Is Right for You?

Here’s a quick rundown of everything covered above — who each solution is best suited for, and how much effort it takes to implement.

Feature	Recommended for	Difficulty
User-provided data / enhanced conversions	Google Ads, Bing Ads, and Meta Ads advertisers	🟡 Medium
Customer Match (Google Ads)	Companies with a customer database	🟡 Medium
Offline conversion import (Google Ads)	Lead-generation businesses	🟡 Medium
Google Consent Mode	Everyone	🟢 Low
Cookie banner optimization	Everyone	🟢 Low
GTM Server-Side	Companies with access to an IT team	🔴 High
Google Tag Gateway	Companies using a CDN (e.g., Cloudflare)	🟢 Low
Meta Conversions API	Meta Ads advertisers	🟡 Medium
Blending online and offline data	Lead-generation businesses with a CRM	🔴 High
Data enrichment	Companies with advanced analytics needs	🔴 High