How to report Organic Keywords Performance in the world of ‘Not Provided’



According to NotProvidedCount.com, the website which tracks the data of not provided keywords, the average percentage of ‘not provided’ Google traffic is around 60% at the time of writing this post. 

average-percent-not-provided

And according to the same website, the not provided traffic will reach 100% on 29th April, 2014.

While I take these numbers and calculations with a pinch of salt (and expect the same from you because of the small sample size used in these calculations), it is not really a “news” for anybody in the digital marketing industry now that the ‘not provided’ keywords traffic is rising each month.  

The rise of ‘not provided’ keywords is gradually eating away the contributions made by both your branded and non-branded keywords in generating traffic and conversions.  

As a result the value of your SEO efforts is being diminished by Google. Today I am going to fix this problem for you. I will show you how you can report your organic keywords performance in the world of ‘not provided’ keywords.

The first thing that you need to remember is that ‘not provided’ traffic is the traffic coming from some keywords.  It can not be the traffic coming from some other referral.

The second important thing to remember is that the not provided keywords are organic keywords.

So we have got two valuable insight from ‘not provided’:

  1. Not provided traffic is keyword traffic
  2. Not provided traffic is organic traffic.

What we don’t know about not provided traffic is how much of the traffic is from branded keywords and how much is from non-branded keywords (and of course the actual keyword referrals) .

So it would be safe to conclude (in theory) that not provided traffic is neither branded nor non-branded. It is simply a different type of organic traffic called the ‘not provided traffic’.

So we have three different types of organic traffic:

  1. Branded Keywords Traffic
  2. Non Branded Keywords Traffic
  3. Not Provided Keywords Traffic

As the traffic from ‘not provided’ keywords will increase, there is a strong possibility that the traffic and conversions from both branded and non-branded keywords will decrease over time.

So if you don’t report the traffic and conversions from ‘not provided’ keywords, you won’t be able to report the accurate picture of your SEO efforts.

Not provided traffic will continue to eat your branded and non-branded keywords traffic and conversions, and over time, you may give the impression of running a poorly optimized SEO campaign to your clients.

So here is my solution.

Create three different column charts which shows the traffic and conversions coming from branded keywords, non-branded keywords and not provided keywords.

Create one more column chart which shows the traffic coming from all of the keywords (branded + non-branded + not provided).

 For example:

branded-nonBranded

not-provided-total-keywords

To get the number of branded, non-branded and not provided keywords you need to create and apply advance segments/filters to your ‘organic search traffic’ report in Google Analytics.

Once you have applied these advanced segments/filters (which can filter out branded, non-branded and not provided keywords), you just need to count the number of keywords sending traffic for each month and note it down in your excel spreadsheet.

For example:

 count

In order to count the number of ‘not provided’ keywords sending traffic to your website, you need to apply ‘landing page’ as the secondary dimension.

For example: 

not-provided-count

I have saved these three different customization of organic search traffic report as ‘shortcut’ reports so that I don’t need to apply advanced segments and filters every time.

Note: You can learn more about the various shortcuts in Google Analytics from this post: Google Analytics Shortcuts: Tricks, Tools, keyboard & APIs 

Ideally the total number of keywords which send traffic to your website must increase every month. This is a sign of a healthy SEO campaign.

And the last chart on ‘total keywords sending traffic’ exactly solves this very problem.

No matter how much the volume of not provided keywords increases over time, you can always analyse and report the true value of your SEO efforts if you can report the relative changes in branded, non-branded and not provided keywords as well as the total number of keywords sending traffic and conversions to your website. 

So if branded/non-branded keywords traffic is going down then look at the volume of not provided keywords.

Is there a correspondence increase in the volume of not provided keywords? If there is, you know where the branded/non-branded traffic ended up. If there is not, then you know the SEO campaigns are not performing well and it is not the fault of rise in not provided keywords. 

To learn more about ‘not provided’ keywords analysis, check out these two posts:

  1. Google Analytics Not Provided Keywords Analysis – Ultimate Guide
  2. Google Analytics Not Provided Keywords – workarounds

 

Subscribe by Email

Comments

  1. says

    Himanshu,

    Unless I misread your post, I think you’ve got a problem. Adding a landing page secondary dimension to (not provided) doesn’t tell you how many keywords drove traffic to your site – it tells you how many landing pages received traffic from the (not provided) keyword.

    You can’t get the keywords back out of (not provided), they simply aren’t there for us to get at – no matter how clever we think we are.

    I think the best we can muster right now is to apportion the volume of (not provided) traffic back into each actual keyword driving traffic into a site in the same ratio that each known keyword represents of the known total keyword volume. You could improve that potentially by doing the same thing on a landing page level, which would remove the relatively unlikely scenario of a landing page receiving only actual keywords and no (not provided) keyword traffic.

    Thoughts or am I barking up the wrong tree?

    Al.

  2. says

    why not stacked barchart? this way you use the same scale (your charts are a bit confusing because of different y axis) and use colors to highlight variations. Plus, you don’t need a total chart :)

  3. Dario Ricci says

    Nice try!
    So the conclusion is that if you can just focus on the number of keywords is generating traffic ( i mean is a tricky metric, if you cannot verify the quality of keywords), don’t you think that the cruel reality is: that’s not so important. I mean if the focus is SEO and you cannot rely on some missing data, keep on focus on available data an keep on create good contents and (i supposo) good traffic.

    What do you think?

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>