Keep on scrollin'

Better Measures of Effectiveness: Share of Search

“If you can’t measure it, you can’t manage it”

This quote from famed Management Consultant Peter Drucker is regularly rolled-out to highlight the necessity of effectiveness measurement in marketing.

It’s a snappy quote and a useful phrase for instilling the value of good measurement in marketing teams.

However, it was written before marketing entered the digital world.

The age of abundance

The digitisation of marketing during the past 20 years has created an abundance of measurement available to brands.

That means that for marketers today, it’s often not a case of ‘what can I possibly measure,’ but more a case of ‘how the hell do I know which one of these 2,000 data points, metrics, methods, tools and systems are even important.’

It’s why at 1000heads we’re on a constant drive to distil and to simplify measurement wherever possible.

For this very reason, recent research into the usefulness of Share of Search as a trackable indicator of performance caught our eye.

Share of search and its role

The premise of Share of Search is simple – long term trends in the effectiveness (or not) of brand communications can be seen in changes to the number of people searching for the brand over time.

Now, it is worth pointing out at this stage that this approach is not about ‘search marketing’ and it’s also not about inserting a brand into people’s unbranded searches.

What it is about, however, is using the number of searches carried out within a category as an accurate indicator of 1) communication effectiveness and 2) commercial performance. Almost as a bridge between the two, validating the link between good communications and commercial impact where, more often than not, that link can often be hard to identify.

So we end up with something like…

Brand communications are resonating


Number of people searching for brand-related terms goes up


Penetration goes up

(For a more detailed explanation and exploration of the research data, watch this video.)

How could share of search be used?

First, let’s explore the idea of ‘share.’ Share is an important aspect of marketing measurement as by its nature it offers relativity to competitors… or ‘my share is bigger than your share.’

Share is already used a lot when marketers measure their communication programmes against competitors: Share of Voice, Share or Conversation, Share of Engagement, Share of Fame.

It’s also used to bring context to sales data: Share of Market.

With all of this in mind, the primary use of Share of Search data is as a contemporaneous read on the potential long-term commercial impact of communications.

So, in short – if it can be proved that a lift in searches for a brand really does indicate that more people will go on to buy in the future, then we could use that lift in searches as an almost immediate indicator of communications effectiveness.

And we’ve, therefore, created the opportunity to use historic data to predict future impact; by knowing that, for example, a 10pp swing in the share of search against competitors equates to a 5pp increase in the share of the market further down the line.

What to do with this theory

Test it. Begin mining search data for your category and start experimenting with analysing it against the metrics you are already tracking over time for your communications work. Look for cause and effect, create models that help measure and predict the impact of communications metrics and commercial metrics.

Watch outs

  • Correlation does not equal causation. Increased penetration is impacted by factors other than communications. Be careful what you attribute success to.
  • Share of Search mightn’t work for all categories. Try it on high-consideration categories like automotive and smartphones; categories where search plays a vital role in indicating intent.
  • An increase in the use of voice search in the future might compromise the data over time.
  • Google Trends for Search data is based on available samples that can shift from day-to-day, therefore multiple data pulls over a period should be used to establish a fair average and a reliable read.
  • Don’t forget Goodhart’s law – ‘when a measure becomes a target, it ceases to be a good measure’ – keep that data objective.

If you’d like to talk to the 1000heads Insight Team about measuring the effectiveness of your brand’s communications, drop us a line at