Deprecated: preg_replace(): The /e modifier is deprecated, use preg_replace_callback instead in /var/www/vhosts/coolpink.net/httpdocs/lib/vendor/symfony-1.4.6/lib/util/sfToolkit.class.php on line 362

Deprecated: preg_replace(): The /e modifier is deprecated, use preg_replace_callback instead in /var/www/vhosts/coolpink.net/httpdocs/lib/vendor/symfony-1.4.6/lib/util/sfToolkit.class.php on line 362

Deprecated: preg_replace(): The /e modifier is deprecated, use preg_replace_callback instead in /var/www/vhosts/coolpink.net/httpdocs/lib/vendor/symfony-1.4.6/lib/util/sfToolkit.class.php on line 362

Deprecated: preg_replace(): The /e modifier is deprecated, use preg_replace_callback instead in /var/www/vhosts/coolpink.net/httpdocs/lib/vendor/symfony-1.4.6/lib/util/sfToolkit.class.php on line 362

Does (Not Provided) render SEO useless?

Back in October 2011, Google started a process of encrypting search data by redirecting its users to the https:// version of their Google domain, ensuring the user performs an SSL encrypted search.

For the user, this is fantastically reassuring that our search data is private.

However, for marketers, it was the beginning of a long-standing nightmare.

What does Not Provided mean?

The result was that, for encrypted search, the keyword data appeared as ‘not provided’ in Google Analytics.

At first, this was only rolled out to users logged into a Google account and only affected a small percentage of a website’s organic traffic.

Google initially claimed its effect would only, even at full roll out, still be in the single digit percentages of all Google searchers.

However, over the last couple of months, we have seen the percent of (not provided) traffic increase dramatically as Google flipped a switch to encrypt ALL searches.

The general effect of ‘not provided’ across the web can be seen on http://www.notprovidedcount.com/ which reports data from 60 websites to show the rise of ‘not provided’:

not provided SEO data

Why is Not Provided so worrying?

The average has just eclipsed 80%, though already we’ve seen some nearly reach 100%.

At its current rate of increase, ‘not provided’ is expected to reach 100% by mid-November.

This sudden increase has sparked panic throughout the SEO industry, as it did when it was first introduced.

However, there is no real great need to panic. It is more of an inconvenience, than the death of SEO as many marketers are claiming!

Many are acting like this sudden increase is a surprise; when really, it was inevitable.

SEO must adapt to the evolving landscape

For the 18 months prior to the increase, we had seen a pretty slow and steady rise, which meant marketers had learned ways around it, adapting their reporting and insight methods.

Though the 100% not provided seems like a frightening loss of data, it’s not much worse than we’ve had recently anyway.

All that has been taken away is the granularity of being able to report on traffic and sales at a key-phrase level – data we’ve only had 50% of for a while anyway.

How to determne Not provided data

Here are a few methods of coping with the change:

1. Take a step back

It’s time to take a step back from keyword granularity, and focus reporting on organic traffic as a whole.

The ultimate aim of any SEO campaign is to see an uplift of traffic and conversions for overall organic traffic. So, ultimately these are the metrics that derive success or failure.

I fear marketers will revert to obsessing about as many and as high volume a ranking as they can get, rather than focus on the end game of increasing conversions, so it’s important to still focus on quality over quantity.

2. Landing pages can reveal a lot about your campaign

Vital data can be found  by looking at exactly where organic traffic is landing on your site.

Our research indicates that traffic landing on a homepage is around 65% branded and 35% non-branded, whereas traffic landing deeper into a site is around 15% branded and 85% non-branded.

So if you start to see increases in traffic to deeper pages, it can be an indication that your campaign is producing results.

The exact landing page responsible for the increase can also tell you a lot about what key-phrases might be used

3. Buckets is the way forward

Rather than focusing on individual key-phrases, focus on buckets of key-phrases. More specifically, and leading on from the above, create buckets of key-phrases that target the same landing page to help further identify increases or decreases in performance.

4. Benchmark, if you can

If you’re lucky enough to have data spanning back a couple of years, you should be able to take benchmarks of brand and non-brand traffic, as well as individual or buckets of key-phrases, which will be useful to map against in the coming months.

5. Utilise Google Trends

It’s important to still research the trending of key-phrases. Marry up your organic data with tools like Google Trends and Keyword Planner to help explain any rise and falls.

It is certainly useful to monitor the trending of brand terms. If your brand is increasing in search volume, then it’s likely responsible for the increase in organic traffic.

 6. Use PPC data

The encrypted search only affects organic search, not paid search, meaning all PPC key-phrase data is available.

So we can still use PPC data to analyse the change in behaviour and conversion of certain key-phrases. Going forward, this data can help you to plan the expansion of your SEO campaign.

A small campaign targeting brand terms can also help to monitor the trend in search volumes of your brand, which you can team with your organic data in Analytics to see if there’s any relation.

SEO is not dead, it’s just evolving.

So if you use all the above, with ranking data, you might just be onto a winner.

Written by Mark Richards


Leave a Reply