Not Provided Keywords: How to Get Your Search Data Back

Not provided keywords appear in Google Analytics when organic search traffic arrives without keyword data attached, because Google encrypts search queries for logged-in users. The result is that the “organic keywords” report in GA4 shows almost nothing useful. To recover that data, you need to connect Google Search Console to your Analytics account, use third-party tools, or build a proxy methodology using landing page data.

This is one of the most common frustrations I hear from marketing teams who expect their analytics platform to tell them what people searched for before landing on their site. It can, but not in the way most people assume, and the workarounds require a bit of setup.

Why Does “(Not Provided)” Exist in the First Place?

In 2011, Google began encrypting search queries for users who were signed into a Google account. By 2013, that encryption had been extended to all searches, regardless of whether the user was logged in. The referrer data that used to pass keyword information to Google Analytics was removed. What you see instead is “(not provided)” in the organic keywords dimension.

Google’s stated reason was user privacy. The practical effect was that organic search, which had been one of the most transparent channels in digital marketing, became one of the least. Paid search data was preserved because Google had a commercial interest in keeping it visible to advertisers. Organic search data was not.

I remember the shift clearly. We were running a large content and SEO programme at the time, and the keyword-level data we had been using to inform editorial decisions disappeared almost overnight. Teams that had built reporting workflows around organic keyword performance had to rebuild from scratch. The frustration was real, and some of that frustration is still sitting in analytics dashboards today, unresolved.

If you want a broader view of how Analytics has evolved and where it fits in your measurement stack, the Marketing Analytics hub at The Marketing Juice covers GA4, attribution, and performance reporting in depth.

What Does GA4 Actually Show for Organic Keywords?

In GA4, if you go to Reports, then Acquisition, then Traffic Acquisition, and filter by organic search, you will see session data but almost no keyword-level detail. The “Session default channel group” dimension confirms organic search as a source, but it does not tell you what the user searched for.

If you look at the “First user source” or “Session source/medium” dimensions, you will see “google / organic” but the keyword field remains empty or shows “(not set)” in GA4’s terminology. This is not a configuration error. It is the expected behaviour given how Google handles search data.

There are some exceptions. Certain paid search keywords will appear if you have auto-tagging enabled and are running Google Ads campaigns. Some traffic from other search engines, including Bing, may still pass keyword data depending on the referrer configuration. But for the bulk of organic Google traffic, the keyword dimension is simply unavailable inside GA4 on its own.

The SEMrush blog has a useful explanation of how Google Analytics handles keyword data and why the gap exists, which is worth reading if you want the full technical picture.

How to Connect Google Search Console to GA4

The most direct solution is to connect Google Search Console to your GA4 property. Search Console records the queries that drove impressions and clicks to your site directly from Google Search. It is not a perfect dataset, but it is the closest thing to real keyword data available for organic traffic.

To set this up, you need to have a verified Search Console property for your domain, and you need admin access to your GA4 property. In GA4, go to Admin, then Property Settings, then Search Console Links. Click Link, select your Search Console property, and confirm the connection. Once linked, a new set of reports becomes available under Reports, then Acquisition, then Search Console.

The Search Console reports in GA4 show four dimensions that are genuinely useful: Google organic search queries, Google organic search traffic, Google organic search landing pages, and Google organic search countries. The queries report is the one most people are looking for. It shows the search terms users entered before clicking through to your site, along with clicks, impressions, click-through rate, and average position.

There are limitations worth knowing. Search Console data is sampled for queries with very low volume. Queries that Google considers privacy-sensitive may be aggregated or withheld. The data typically goes back 16 months. And the Search Console reports in GA4 do not allow you to cross-reference keyword data with on-site behaviour like time on page or conversion rate in the same row, which is a significant constraint.

That last point matters more than people realise. Knowing that a keyword drove 400 clicks is useful. Knowing that those 400 clicks produced zero conversions is what tells you whether the keyword is actually worth pursuing. You can approximate this by comparing landing page performance in GA4 against the landing pages shown in Search Console, but it requires manual correlation rather than a single joined view.

Using Landing Page Data as a Keyword Proxy

Before Search Console integration existed, and still today when you want more granular behavioural data, landing page analysis is the most practical workaround for understanding keyword intent from organic traffic.

The logic is straightforward. If a page is ranking for a specific keyword, organic traffic to that page is likely arriving via that keyword or a close variant. If you know what your pages are optimised for, and you can see which pages are receiving organic traffic, you can make a reasonable inference about what users searched for.

In GA4, go to Reports, then Engagement, then Landing Page. Filter the session source/medium to “google / organic”. This shows you which pages organic visitors are entering on, along with sessions, engagement rate, and conversions. You now have a list of your highest-traffic organic entry points.

Cross-reference that list with Search Console. In Search Console, go to Performance, then Pages, and click on each URL to see which queries are driving traffic to it. You now have a reasonably complete picture: which pages are receiving organic traffic, which queries are driving that traffic, and, back in GA4, how those visitors behave once they arrive.

This is not a smooth workflow. It requires switching between two platforms and doing some manual matching. But it is more honest than hoping GA4 will show you keyword data it was never designed to surface. I have run this process for clients across retail, financial services, and travel, and it consistently surfaces insights that a straight Analytics report would miss.

Third-Party Tools That Fill the Gap

For teams that need richer keyword data than Search Console provides, third-party SEO platforms offer another layer of visibility. These tools do not access your actual Analytics data. Instead, they estimate keyword rankings and traffic volumes based on their own crawl data and clickstream modelling. The data is an approximation, but it is often a useful one.

SEMrush, Ahrefs, and Moz are the main platforms in this category. Each allows you to enter your domain and see which keywords the tool believes your pages are ranking for, along with estimated traffic, position, and keyword difficulty. This data can be used alongside your GA4 and Search Console data to build a fuller picture of your organic keyword footprint.

Moz has written about Google Analytics alternatives and supplementary tools that are worth exploring if you want to understand what else is available beyond the standard GA4 setup.

One thing I would caution against is treating third-party keyword estimates as ground truth. Early in my career, I made the mistake of building a keyword performance report almost entirely from an SEO tool’s traffic estimates, without validating against actual Analytics data. The estimates were directionally useful but materially wrong in several categories. When the client asked why the tool showed 8,000 monthly visits from a particular keyword cluster but Analytics showed 1,200 sessions, I did not have a clean answer. Since then, I treat third-party keyword data as a hypothesis to be tested against actual traffic, not a replacement for it.

Can Behaviour Analytics Tools Help?

Tools like Hotjar and Crazy Egg sit alongside GA4 rather than replacing it, but they can add useful context to your organic keyword analysis. If you know from Search Console that a particular page is receiving significant organic traffic from informational queries, behaviour analytics can show you whether visitors are reading the content, where they drop off, and whether the page is actually satisfying the intent behind those searches.

Hotjar integrates with Google Analytics and allows you to segment session recordings and heatmaps by traffic source, so you can isolate organic visitors and observe how they interact with specific landing pages. The Hotjar and Google Analytics integration is straightforward to set up and adds a qualitative layer that pure Analytics data cannot provide.

Similarly, Crazy Egg offers Google Analytics integration for A/B testing, which can be useful if you want to test different content approaches for pages that are receiving organic traffic from specific keyword clusters. If Search Console shows a page ranking well for a commercial keyword but converting poorly, you can use that combination of data to design and measure a content experiment.

These tools will not recover the keyword data that Google removed. What they do is help you understand whether the traffic you are getting from organic search is actually engaging with your content in the way you intended. That is a different question, but often a more actionable one.

Building a Practical Keyword Reporting Workflow

The honest answer is that there is no single report in any tool that gives you a complete view of not provided keywords. What you can build is a workflow that combines multiple data sources into a picture that is good enough to make decisions from. That is a different standard than “complete”, but it is the right standard for practical marketing work.

When I was running the performance marketing function at a large agency, we had clients who would ask for a single dashboard showing keyword-level ROI across all organic traffic. The honest answer was that it did not exist, and anyone who told them otherwise was either selling them something or had not thought it through. What we could give them was a methodology that was transparent about its assumptions and consistent enough to track over time.

Here is a workflow that works in practice. First, connect Search Console to GA4 if you have not already done so. Second, export the Search Console queries report weekly and tag each query by intent category: informational, navigational, commercial, transactional. Third, in GA4, pull the organic landing page report and match it against your Search Console landing page data to identify which pages are over-performing or under-performing relative to their search visibility. Fourth, use a third-party tool to fill gaps, particularly for keywords where you have organic impressions but low click-through rates, since that often signals a ranking or meta-data problem rather than a content problem.

The HubSpot blog makes a useful distinction between marketing analytics and web analytics, and it is relevant here. Web analytics tells you what happened on your site. Marketing analytics tells you why it happened and what to do about it. Keyword analysis sits squarely in the marketing analytics category. The goal is not to count visits. It is to understand what people were looking for, whether you gave them what they needed, and what that means for your content and SEO strategy.

What Most Teams Get Wrong About Not Provided

The most common mistake I see is teams treating not provided as a problem to be solved once, rather than a constraint to be managed continuously. They connect Search Console, look at the queries report once, and then go back to ignoring keyword data entirely. The value comes from treating this as a recurring workflow, not a one-time fix.

The second mistake is over-indexing on keyword volume at the expense of keyword intent. Search Console will show you queries with high impression counts and low click-through rates. The instinct is to optimise title tags and meta descriptions to improve CTR on those queries. Sometimes that is right. But sometimes low CTR on a high-impression query is telling you that the page does not match what the user is actually looking for, and no amount of meta-data optimisation will fix a content mismatch.

The third mistake is treating not provided as a Google Analytics problem specifically, when it is really a data strategy problem. If your entire understanding of organic search performance depends on a single report in a single tool, you are one platform change away from losing visibility entirely. The teams that handle this well are the ones who triangulate across multiple sources, understand the limitations of each, and build reporting that is honest about uncertainty.

MarketingProfs has written about the power of web analytics for marketers in a way that still holds up: the value of analytics is not in the data itself but in the decisions it enables. That framing is useful when you are working with an imperfect dataset like organic keyword data. You do not need perfect information. You need enough information to make a better decision than you would make without it.

If you are building out a broader analytics capability beyond keyword tracking, the Marketing Analytics section of The Marketing Juice covers measurement frameworks, GA4 setup, and attribution in more detail.

A Note on GA4 Explorations for Keyword Analysis

GA4’s Explorations feature offers more flexibility than the standard reports, and it is worth using for keyword-adjacent analysis even if it cannot surface the keyword data itself. You can build a free-form exploration that shows organic landing page performance segmented by geography, device, and user type. You can create funnel explorations that show how organic visitors progress from entry page to conversion. You can use path analysis to see where organic visitors go after their initial landing page.

None of this tells you what keyword drove the visit. But it tells you a great deal about the quality of the organic traffic you are receiving, and that is often the more important question for conversion-focused teams. A keyword that drives 2,000 visits with a 0.2% conversion rate is less valuable than a keyword that drives 400 visits with a 3% conversion rate, and GA4 Explorations is where you surface that distinction.

The early days of paid search taught me something that applies equally to organic. At lastminute.com, when I launched a paid search campaign for a music festival and watched six figures of revenue come in within roughly a day, the keyword data was the starting point, but the conversion data was what told us whether the campaign was actually working. The same principle applies to organic. Keywords drive traffic. What matters commercially is what that traffic does when it arrives.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions