Google Analytics Not Provided Keywords Analysis – Ultimate Guide

 

Google has been hiding keyword referral data from searches made by logged-in users since October of 2011. With every passing day, more and more searches are made by logged-in Google users. We therefore can no longer ignore the impact of ‘not provided’ keywords on various marketing channels.

We should also embrace this fact that one day we will lose majority (if not all) of the keyword referral data from searches made by logged-in Google users:

not-provided-world

‘Not provided’ keywords don’t fall into the category of either branded keywords or non-branded keywords. Therefore they deserve their own separate category and analysis called the ‘not provided’ keywords analysis.

Related Post: How to optimize Organic Search Campaigns without Keyword Referral Data

In a world of multi-channel marketing, people are exposed to multiple marketing channels (organic search, paid search, display, social etc) before they make a purchase or complete a conversion. Therefore it is important that we calculate the impact of ‘not provided’ keywords on all the marketing channels and not just on organic search.

In this post you will learn to calculate the impact of ‘not provided’ keywords on:

  1. Organic Branded Keywords
  2. Organic Non-Branded Keywords
  3. Paid Branded Keywords
  4. Paid Non-Branded Keywords
  5. Direct Traffic
  6. Social Media Traffic
  7. Email Traffic
  8. Affiliate Traffic
  9. Display Traffic
  10. Assisted Conversions
  11. Last Interaction Conversion
  12. Total economic value

Before we start our calculations, we need to create a new ‘custom channel grouping’ in Multi-Channel funnel reports in Google Analytics. For this to take place we should first all be on the same page about how channel labels and channel grouping are defined and used in Google Analytics.

 

Some background information about Channel Labels and Channel Grouping

A channel label is the label applied to a digital marketing channel.  For example ‘paid search’, ‘organic search’, ‘social’, ‘display’ etc are all examples of channel labels. There are two types of channel labels in Google Analytics: Default Channel Labels and Custom Channel Labels.

The default channel labels are the predefined channel labels. For example:  ‘paid search’, ‘organic search’, ‘referral’, ‘display’, ‘email’, ‘social’, ‘direct’ and ‘other advertising’ are default channel labels. The custom channel labels are the labels defined by a user. Branded keywords and non-branded keywords are examples of custom channel labels.

channel-labels

Channel Grouping is a set of channel labels. There are two types of channel grouping in Google Analytics: Basic Channel Grouping and Custom Channel Grouping. The ‘Basic Channel grouping’ is the set of predefined channel labels. The ‘custom channel grouping’ is the channel grouping created by a user.

 

Creating Super Useful Custom Channel Grouping

Follow the steps below to create a new custom channel grouping named ‘Super Advanced Keywords Analysis’:

Step-1: Go to Top Conversions Paths report (under Conversions > Multi Channel Funnels) in Google Analytics.

Step-2: Click on the ‘Channel Grouping’ link and then click on the ‘create a custom channel grouping’:

custom-channel-1

Step-3: In the ‘create or edit channel grouping’ dialog box, enter the name of the channel grouping as ‘Super Advanced Keywords Analysis’ and then click on the ‘Add New Rule’ button:

 edit-channel-grouping

Note: We are going to create 5 rules in total for this custom channel grouping.

Step-4: Enter ‘Organic B Keywords’ (which stands for organic branded keywords) as the name of the new rule.  Now set up the conditions for the rule as shown in the screenshot below:

custom-channel-3

In the ‘Matching RegExp’ text box you enter the regular expression which matches all of your branded keywords (the regex has been hidden in the screenshot above). Here we are defining all those keywords as branded organic keywords which contain your brand name in the keyword phrase and the medium of the traffic is organic.

Related Post: Regular Expressions Guide for SEO & Analytics

Since organic reminds me of vegetables, I have selected the color ‘dark green’ for this channel label. Once you have set up the conditions for the rule, click on the ‘save rule’ button (not shown in the screenshot above because of space constraints). You will now see a screen similar to the one below:

custom-channel-4

Step-5: Now create 2nd rule by clicking on the ‘add new Rule’ button. Enter ‘Paid B Keywords’ (which stands for Paid Branded Keywords) as the name of the new rule.  Now set up the conditions for the rule as shown in the screenshot below:

custom-channel-5

In the ‘Matching RegExp’ text box you enter the regular expression which matches all of your branded keywords (the regex has been hidden in the screenshot above). Here we are defining all those keywords as branded paid keywords which contain your brand name in the keyword phrase and the medium of the traffic is either cpc or ppc.

Since paid search traffic creates a hole in the pocket, it reminds me of fire on the low flame. So I have selected the color of this channel label accordingly. Once you have set up the conditions for the rule click on the ‘save rule’ button.

 

Step-6: Create 3rd rule by clicking on the ‘add new Rule’ button. Enter ‘Organic NB Keywords’ (which stands for Organic Non Branded Keywords) as the name of the new rule.  Now set up the conditions for the rule as shown in the screenshot below:

custom-channel-6

In the ‘Matching RegExp’ text box you enter the regular expression which matches all of your branded keywords (the regex has been hidden in the screenshot above). Here we are defining all those keywords as organic non-branded keywords which do not contain your brand name in the keyword phrase, which do not contain ‘not provided’ keywords and the medium of the traffic is organic. Once you have set up the conditions for the rule click on the ‘save rule’ button.

 

Step-7: Create 4th rule by clicking on the ‘add new Rule’ button. Enter ‘Paid NB Keywords’ (which stands for Paid Non Branded Keywords) as the name of the new rule.  Now set up the conditions for the rule as shown in the screenshot below:

custom-channel-7

In the ‘Matching RegExp’ text box you enter the regular expression which matches all of your branded keywords (the regex has been hidden in the screenshot above). Here we are defining all those keywords as paid non-branded keywords which do not contain your brand name in the keyword phrase, which do not contain ‘not provided’ keywords and the medium of the traffic is either cpc or ppc. Once you have set up the conditions for the rule click on the ‘save rule’ button.

 

Step-8: Create 5th rule by clicking on the ‘add new Rule’ button. Enter ‘Not Provided Keywords’ as the name of the new rule. Now set up the conditions for the rule as shown in the screenshot below:

custom-channel-8

Once you have set up the conditions for the rule click on the ‘save rule’ button. You will now see a screen similar to the one below:

custom-channel-9

Note: Make sure you assign different colors to channel labels so that it is to spot them in reports.

Step-9: Click on the ‘save and apply’ button and you have now created your first super useful custom channel grouping. You will now see a screen similar to the one below:

custom-channel-10

 

Calculating the Impact of Not Provided keywords on Organic Branded Keywords

Step-1: Make sure that the custom channel grouping ‘Super Advanced Keywords Analysis’ is selected and you are viewing the ‘Top Conversions Paths’ report (under Conversions > Multi Channel Funnels) in Google Analytics.

Step-2: Select the time period for which you want to do the analysis.

Step-3: Click on the ‘advanced’ filter in the reporting interface and then set the conditions for it as shown in the screenshot below:

not-provided-organic-b

Click on the ‘apply’ button. You can now see the impact of ‘not provided’ keywords on organic branded keywords:

not-provided-organic-b2

From the report you can determine all the conversion paths in which ‘not provided’ keywords played an important role along with the ‘organic branded keywords’ in initiating, assisting or completing a conversion. Without the role of the ‘not provided’ keywords in the conversion process, these conversions would not have occurred in the first place. The number of such conversions is 2718 and the value of these conversions is $286,919.51

 

Calculating the Impact of Not Provided keywords on Organic Non-Branded Keywords

Step-1: Make sure that the custom channel grouping ‘Super Advanced Keywords Analysis’ is selected and you are viewing the ‘Top Conversions Paths’ report (under Conversions > Multi Channel Funnels) in Google Analytics.

Step-2: Select the time period for which you want to do the analysis.

Step-3: Click on the ‘advanced’ filter in the reporting interface and then set the conditions for it as shown in the screenshot below:

not-provided-organic-nb

Click on the ‘apply’ button. You can now see the impact of ‘not provided’ keywords on organic non-branded keywords:

not-provided-organic-nb2

From the report you can determine all the conversion paths in which ‘not provided’ keywords played an important role along with the ‘organic non-branded keywords’ in initiating, assisting or completing a conversion. Without the role of the ‘not provided’ keywords in the conversion process, these conversions would not have occurred in the first place. The number of such conversions is 32 and the value of these conversions is $3017.02

 

Calculating the Impact of Not Provided keywords on Direct Traffic

Follow the steps 1 and 2 above

Step-3: Click on the ‘advanced’ filter in the reporting interface and then set the conditions for it as shown in the screenshot below:

not-provided-direct

Click on the ‘apply’ button. You can now see the impact of ‘not provided’ keywords on ‘direct traffic’:

not-provided-direct2

From the report you can determine all the conversion paths in which ‘not provided’ keywords played an important role along with the ‘direct traffic’ in initiating, assisting or completing a conversion. Without the role of the ‘not provided’ keywords in the conversion process, these conversions would not have occurred in the first place. The number of such conversions is 20,770 and the value of these conversions is $2,504,336.35

 

Similarly you can calculate Impact of Not Provided keywords on Social Media Traffic, Email Traffic, Affiliate Traffic and Display Traffic. All you have to do is to make some changes in your advanced filter.

 

Calculating the total economic value generated by ‘not provided’ keywords

The total economic value generated by ‘not provided’ keywords is calculated as

Assisted Conversion Value generated by ‘not provided’ keywords + Last Interaction Conversion Value generated by ‘not provided’ keywords

We can determine the assisted conversion and last interaction conversion values through ‘Assisted conversions’ report (under Conversions > Multi Channel Funnels) in Google Analytics.

 

Follow the steps below to calculate the total economic value generated by ‘not provided’ keywords:

Step-1: Go to the ‘Assisted conversions’ report in your Google Analytics account.

Step-2: Select the custom channel grouping ‘Super Advanced Keywords Analysis’

Step-3: Select the time period for which you want to do the analysis.

Step-4: Apply the on page filter ‘keywords’ so that you can see all of your custom channels in your report. You will now see a screen something like this:

 total-economic-value

You can now determine the number of assisted conversions and last interaction conversions generated by ‘not provided’ keywords along with their conversion value. Just sum up the assisted conversion value and last interaction conversion value generated by ‘not provided’ keywords and you will get the total economic value generated by the ‘not provided’ keywords. In our example it is:

$1,115,343.93 + $1,207,332.49 = $2,322,676.42

Who would have imagined that the total economic value of ‘not provided’ keywords could be $2.32 million?

Bonus Insight => You can now also determine the assisted conversion values and last interaction conversion values of: organic branded keywords, organic non branded keywords, paid branded keywords and paid non branded keywords. 

Update

I have been recently asked a good question on Linkedin related to this Post which I think is worth sharing:

Adam Cranfield • Hi Himanshu – could you explain what the point is of measuring the impact of “not provided” keywords? How would you use this info to optimise your website or gain customer insight?

My Answer:

First of all, the point of any analysis is to get an insight. Once we get the insight, we need to determine whether this insight is useful or not. Insights are generally useful because you get to know your business/campaign better. But what is even more useful is the ‘actionable insight’. Can we take any action on the basis of the insight we have gathered?

In our case, the answer is ‘yes’.  It is a common practice that before we fix a problem we assess the size of the problem. Yes ‘not provided’ is a problem. It is a big problem for marketers who rely on keywords to optimize their campaigns.  The first step toward assessing the size of the ‘not provided’ problem is to measure the total economic value generated by ‘not provided’ keywords. If this economic value is small or negligible for your business then you don’t need to worry about ‘not provided’ keywords yet.

However if the total economic value is large or very large or the performance of your marketing campaigns are deeply affected by ‘not provided’ keywords then you have got a work to do. For example:

If only few organic branded conversions occurred because of the role of ‘not provided’ keywords in the conversion process then we can conclude that the majority of ‘not provided’ keywords are non-branded and our acquisition strategy is weak. We are targeting keywords which have low conversion potential for our business.  This conclusion is based partially on the observations of customers’ online behavior over the years and partially on the assumption that majority of people eventually convert through branded keywords.

People generally start their conversion journey by searching for non-branded keywords (provided they are not already familiar/loyal to your brand). But as their understanding of what they are looking for increases they refine their search queries. Also people rarely buy on their very first visit. They do comparison shopping, check websites for reviews; look for better deals before they make a purchase decision.  So when they return to your website and convert, the source of traffic is either direct or branded keywords (because branded keywords are easy to remember)

 

Because of this customers’ behavior, majority of conversions are attributed to direct traffic and branded keywords. So if majority of conversions on your website are occurring as a result of direct traffic and branded keywords then it is actually a sign of a healthy business. It also means that your branding is strong. Contrary to this if majority of conversions on your website are occurring as a result of non-branded keywords then it means you are either a new business or your branding is poor.

These are some of the useful insights you can get from the analysis of ‘not provided’ keywords.  We should also need to remember that one day we may not have any keyword referral data left to analyze. But that doesn’t mean that we should stop our keywords analysis. You then need to start optimizing your campaigns without keyword referral data. Check out this post for more details: How to optimize Organic Search Campaigns without Keyword Referral Data

Related Post: Overcoming Not provided keywords in Google Analytics

 

Now it is your turn. How do you analyze the impact of not provided keywords? Please share your thoughts and insight.

Other Posts you may find usefulHow to do Attribution Modeling in Google Analytics – Ultimate Guide

Bonus Post => Introducing Predictive Marketing – The next stage of Business Optimization

 

 

Himanshu Sharma About the Author: is the founder of seotakeaways.com which provides SEO Consulting, PPC Management and Analytics Consulting services to medium and large size businesses. He holds a bachelors degree in ‘Internet Science’, is a member of 'Digital Analytics Association', a Google Analytics Certified Individual and a Certified Web Analyst. He is also the founder of EventEducation.com and EventPlanningForum.net.

My business thrives on referrals, so I really appreciate recommendations to people who would benefit from my help. Please feel free to endorse/forward my LinkedIn Profile to your clients, colleagues, friends and others you feel would benefit from SEO, PPC or Web Analytics.