Go to Blog

How to extract the data from “not provided” segments

SEMrush Blog
How to extract the data from “not provided” segments

If you are a website owner or SEO specialist, you know how annoying the “not provided” segment of keyword research is. It conceals from us a considerable part of the data we’re seeking, and these days, when almost everyone is somehow involved with Google services, the share of “not provided” data is only growing. Luckily, SEO consultant companies never stop looking for good ways to lock out this part of information, and in our post we are going to give you some bright ideas on how to use Google Analytics’ “not provided” data successfully.

 Applying filters in Google Analytics

A famous post by Avinash Kaushik tells us about a great method for extracting information about landing pages for search queries that are hidden in the “not provided” segment.

It suggests creating in Google website analytics software custom filter the main idea of each is to substitute the not provided entry with name of landing page which the user has been referred to. The filter itself should look like this:

What actually is set there is that whenever Google Analytics sees “not provided” as a search query, it should use its landing page address as the name of the search term. As a result, instead of one single entry displaying “not provided,” you will receive a list of the pages users have landed on with all the corresponding information, such as Visits, Bounce Rate, Avg. Time on Site, etc.

Even though you will not know the exact keywords, you will get comprehensive data on the most popular landing pages, which will help you figure out what pages to primarily focus on.

Use the data from Webmaster Tools

This Google web analytics tool can also give you some useful data. Navigate to the Traffic menu on the left side of the page and select “Search Queries”. Below the graph, you will find a list of the best performing keywords in terms of the number of times they have been shown in the search results and number of clicks they have received. In this table, you will also be able to check the click through rate for these keywords as well as their average position in the search results.

With a Search Queries report, you will receive a good amount of useful information, which will tell you what keywords are responsible for most of the traffic coming from organic search results. When you know the best keywords for SEO, you can go back to Google Analytics and check whether these words give you the best conversions as well. If not, analyze the Bounce Rate, Avg. Time on Site, and other relevant metrics.

Check AdWords keyword report

This method will require some investment, but for those who are already buying traffic from paid search results, it will be an additional useful feature from AdWords. The thing is that, while Google Analytics doesn’t tell us keywords from not provided segment, in the AdWords report you can actually see the whole data on search queries.

Find in the AdWords menu the Matched Search Queries report. It will show the keyword referring data – what keywords have triggered the ad and what pages have been visited. Of course, results you will get from paid search data can’t be fully applied to the organic search, but nevertheless these two spheres overlap and you will get some hints on how to find keywords for SEO.

Make the use of geographical segmentation

Luckily, we can segment the not provided data by the secondary dimension, for example, geographical location, which, in turn, will help you understand at least from what region such users are coming. To obtain this information, choose Country/Territory (or even City) as a secondary dimension for “not provided” entries.

A final note

Of course, an essential share of a “not provided” segment can be a great issue for website owners who are trying to make the most of Google data in order to improve their businesses. However, there are already several ways how to go about this problem (Webmaster Tools and AdWords reports, filtering the data in Google Analytics), and ever-evolving SEO will surely suggest new methods and approaches.

Like this post? Follow us on RSS and read more interesting posts:

This profile is specifically designed for our old blog posts.
Share this post


2000 symbols remain