In recent months, keen followers of the marketing media saw an apparently new term emerging almost everywhere: Share of Search. SoS was also mentioned by strategist Tom Roach in the article he wrote in 2020, arguably the best marketing article of the year. It lifted the veil on this fresh concept. Renowned researcher Les Binet is also said to have discovered hopeful results in his research on the metric. He published his research results in the past few months and they look positive indeed. High time we cast our eye on Share of Search. But before we dive into this promising metric, let us take a moment to consider the complexities of brand metrics.
As digital native marketers, we have become accustomed and addicted to real-time metrics. We want to know how many clicks and how many conversions are being generated by a resource right now, so we can immediately optimise to improve the results. We will discuss the damaging effects this has on brands some other time, but for now we can all agree that performance marketing is now all about measurability.
So what is the situation like for brand metrics? Whereas performance marketing is mainly used for sales oriented results in the short term, branding focuses on the long-term growth of aspects like brand awareness. Les Binet and Peter Field’s paper “The Long and the Short of It” showed us that both performance and branding are necessary for sustainable long-term brand growth.
However, those damned brand metrics… they are hard to measure. For starters, you need panels and getting any movement at all in those metrics requires a large dose of patience and consistency. Sometimes this patience is really put to the test because brands have gone through 3 CMOs before anything can happen, in which case consistency is the first thing to go out the window…
Consequently, optimisation based on brand metrics in real time is extremely difficult (or impossible). Possible proxy KPIs for these metrics have been researched for several years now. One of the things that can be optimised in real time is “attention”, which we have discussed previously. One of the KPIs also generating high expectations is SoS. Will this metric finally give us real-time insight into how our brand is doing?
The answer to the question “What is SoS?” is pretty simple, as illustrated by Tom Roach’s comparison below.
Share of Search is said to be a good predictor of a brand’s so-called mental availability. The more “available” you are as a brand, the more likely a consumer will think of you when they want to buy a product or service in your category and the more likely that person will search for your brand. SoS can be an important metric as mental availability is a significant predictor of growth. And growth equals market share.
Les Binet’s recent research has shown that in the Automotive and Energy markets, SoS can indeed be predictive of market share developments, so much so that SoS is actually ahead of Share of Market. See the Les Binet chart below:
Firstly, SoS can offer insight into brand power and the cross-channel effect of brand deployment. This is potentially more efficient than costly brand surveys that require panels.
SoS probably also provides more insight than brand lift studies (BLS, as seen on YouTube and Facebook) of a random channel, as they only say something about the effectiveness of the particular channel the BLS focused on.
Secondly, SoS can be predictive of the served available market (SAM), i.e. the market share. It can therefore be very valuable to monitor this metric, despite the delayed effect between a brand campaign and a rising or falling market share. Note that initial research results are showing that the delayed effect varies by category:
However, there has been criticism too, primarily by Professor Byron Sharp. The metric should not become a goal in itself, as it is easy to manipulate and may lead marketers to stop focusing on what really matters.
Share of Search:
1/ Competition: Start by identifying all the players/competitors in your category:
Tip: View the Google Ads auction data on generic campaigns.
2/ Search volume: Use Google Trends or Google Keyword Planner to find out the search volume per competitor per month (also own brand name).
Tip: Google Keyword Planner provides up to 24 months of historical absolute data, which is a great dataset to start with.
Tip: Use social media for sentiment underpinning. If the intent behind the branded search changes, volume has less predictive value.
Tip: Exclude keywords with a high impact on search volume, such as ‘scandal’ / ‘dieselgate’ in the case of Volkswagen.
3/ Share of Search: Next, easily calculate the share of search based on the previously shared formula: advertiser brand volume / total brand volume in category.
1/ Advertiser: Determine the number of sales and the turnover per month.
2/ Competition: Determine the total revenue/sales per month in the entire category.
Tip: Take a look at the CBS data, for example, or request this data from financial parties such as ABN and ING.
To find out how Share of Search could work for you, get in touch.