How do Facebook’s ads drive search traffic?

According to a new study released by Facebook this past December, Cross-Channel Planning: Making Search Work Harder, Facebook’s ads do drive search traffic – sometimes.

This study summarized 23 “conversion lift” studies – which are Facebook’s bid to provide the digital ad business with a more holistic alternative to traditional last-click attribution – from July to September 2015.

Participants in the study came from a variety of consumer sectors – including automotive, financial, and travel – with spend above $10,000 a month that were also willing to share traffic and conversion data with Facebook.

Here are highlights, plus my own take on what the data means.

Only one in four campaigns sees any Facebook-search effect

A mere 25 percent of campaigns studied actually saw any statistically significant increase in search referral traffic.

Across the entire group, search lift in search referral traffic was 1.8 percent with a standard deviation of six percent.


Does this mean that three out of four search marketers are wasting their time buying Facebook clicks? Possibly. Unfortunately, the study doesn’t say much about this high failure rate.

Was the lack of search referral traffic due to the nature of the business, the campaigns, the ads themselves, or some other unknown variable?

Obviously, some brands are more likely to drive search referral traffic than others, so it would have been more helpful if Facebook had revealed additional information about the profile of each business.

Also, if other marketing mix models are any guide to interpreting results, there may be a requirement of a very high investment in social advertising in order to overcome the background noise from other marketing and drive measurable search lift.

My teams see a much tighter correlation between paid media outside search and lift in search volume when the display/social/video budget is (Read more...) in relation to search.

Facebook can depress paid search CPCs

Even with the foregoing question unanswered, it’s clear that some brands can use Facebook to achieve some efficiency in paid search. Facebook’s data shows that some marketers can use Facebook to drive down paid search cost per clicks (CPCs).

“When Facebook paid media was introduced into the marketing mix, consumers were more likely to search for branded keywords and in some cases less likely to search for unbranded keywords.

For (one advertiser) 97 percent of the incremental paid search visits were in the form of branded keywords. As a result, we saw that Facebook ads make search ads 1.8 percent more efficient, largely as a result of this change in consumer search behavior.”

My guess is that to a lesser extent the brand created through social (and other forms of paid media) also increases the click-through rate (CTR) and therefore the quality score of generic keywords, resulting in a cost savings.

“For brands in Automotive, Financial Services, and Retail, we found that Facebook paid media caused statistically significant lift in lower-funnel KPIs these advertisers care about, including online and offline sales. Lift in these KPIs varied widely, ranging from flat to 79 percent. For one large retailer, the majority of this sales lift came from increased basket size.”

If the study data is valid, it will likely be celebrated by search marketers, who are always looking for a way to reduce click costs.

Why this matters

Facebook’s study is intended to put the company in a better position to argue that any increased share of budget it receives will be offset by lower CPCs – at least among those industry verticals buying lots of unbranded keywords. Only time will tell if this argument is successful in terms of prying loose more budget for Facebook campaigns.


What is clear – at least in my mind – is the idea that many marketers who’ve not yet experimented with Facebook paid media should be testing the effects of paid social spend and determining for themselves what the effects are on CPC. And more importantly, return on investment and return on advertising spend.

As you likely know from reading my ClickZ columns, I’m a big believer in the power of Facebook Custom Audiences. I have seen it used to great effect and believe it to be the wave of the future, along with similar products now offered by Twitter and Google as well.

It’s also critical for traditional search marketers to begin to regard paid social with more respect than it’s been traditionally accorded and coordinate appropriately. Copy, creative, campaign timing, and messaging should be aligned across Google, Facebook, and any other platforms required for reaching one’s target audience.

For some, measuring the direct effects of Facebook on search marketing KPIs may prove a challenge, but that’s only an argument for making sure that the marketing models are sophisticated enough to take full account of the real relationship between social and search that Facebook’s study points to.

There are also really powerful ways of targeting in Facebook using both custom audiences and other targeting parameters in conjunction using  the Audience Overlap feature, which I’ll cover in the future column.

Homepage and article images via Pixabay and Flickr.

Related reading

This entry passed through the Full-Text RSS service - if this is your content and you're reading it on someone else's site, please read the FAQ at fivefilters.org/content-only/faq.php#publishers.