We are more than six years into the programmatic buying revolution, and digital marketers have wholly embraced audience targeting data as the key driver to successfully reaching consumers online with display advertising.
In fact, it’s predicted that in 2016, 2 out of every 3 display ads will be programmatic, with audience targeting data at the core of virtually every impression.
There are many colors and flavors of audience data, with demographic, psychographic, and third-party behavioral data leading the pack. Yet, the keyword — one of the first digital marketing intent signals — still remains the most valuable and underutilized data point a marketer can use to target any single consumer.
The keyword that one searches on (whether on an external search engine or within the search function of your website) is a direct window into the intent of that consumer’s online behavior at that time. In fact, even the way different consumers search for the same content or page can reveal a lot about their current intent and state of mind. The keyword data trigger is at the heart of why paid search marketing has worked so well and dominated digital marketing budgets for more than a decade.
Many years ago, I ran the search program of a lighting retailer who was trying to generate more business from large orders by contractors and electricians, and get away from low-profit, highly time-intensive requests from consumers looking to replace hard-to-find single light bulbs in their homes. By simply removing a handful of the account’s most searched single-bulb terms (such as “light bulb” and “buy light bulb”) and reinvesting those budgets into higher volume intent terms (such as “bulk cases of light bulbs,” “wholesale light bulb orders,” etc.), the low-value traffic and calls stopped and the bigger orders increased.
I would imagine each of you reading this post has similar examples of how carefully getting into the mind of the consumers via their keyword choices has been the key to optimizing search accounts.
Amazingly to me, in recent study by Forbes, search data ranked last as the type of audience targeting most important to branding campaigns — behind many other audience data types that, in my experience, have less effective results than when using a consumer’s previous search history as a targeting data point.
Of course, this study was asking about branding campaigns, not direct response campaigns where search has proven itself to be a very highly effective and profitable digital marketing channel. Additionally, the marketers surveyed may not be clued into how to access their company’s search data, so their preference for alternative, easier-to-access audience data makes sense.
To argue the other side for a moment, a common knock on search data for audience targeting is that it requires you to be tracking a consumer who has already searched and arrived to your site. Because of that, it tends not to scale as much as some of these other data types that seemingly have an infinite amount of cookies and segments that can be purchased off the shelves.
Sure, big paid search advertisers that reach millions of consumers each month have plenty of data for keyword retargeting, but many others might have to over rely on over-the-counter, third-party audience segments because they don’t have the scale needed to fuel their programs.
So, it’s a scale challenge. I get it. But I would still use all of my keyword targeting first and then supplement with other data types to get the volume I need.
Having seen the power of the keyword for the last twelve years in search engine marketing (SEM), I wanted to do my own investigation. The following is an examination of seven common audience targeting data types that have proven to be valuable to marketers, along with how they compare to the power of targeting via keyword intent signals.
Other than the scale issue mentioned above, I cannot think of any single data type better to target a consumer than with their searched keyword history.
Tracking users as they navigate through your desktop and mobile websites can yield tremendous insight into the mindset of a visitor. However, are they really shopping or just browsing? Are they price comparing or ready to buy?
Although the consumer’s path can be highly directional, it still requires more context to be truly understood at high confidence.
Think of one of the top keywords in your own account. Imagine how the same customer visit might look different when that keyword is coupled with “buy” or “deal” or “information” or “competitor.” Each visit looks different, right?
Knowing the keyword or keyword phrase that drove the visit provides you much more insight into that customer’s intent than knowing the path alone.
Of course, the more you know about your consumers, the better you can target them. For example, if you know that a group of site visitors often browses your “eco-friendly” information, you can assume some commonalities and target your audience buying campaigns to them accordingly.
However, the keyword trumps psychographics because of its in the moment nature. Even the most environmentally friendly consumer isn’t a single-minded person at all times. At the precise moment they search today or tomorrow, they may be more interested in price, color, free shipping, what’s in stock, or other factors that can be extrapolated from the keywords searched.
Certainly, social is data is highly valuable; you will get no argument from me there. If a consumer “likes” or retweets or displays other clear engagement signals on a social site, there’s some valuable information there that can be used to target them later.
To me, social data may one day be able to rival the keyword; for now, however, the keyword is still king.
Social engagement is still rather passive compared to a user opening a search engine and actively querying specific keywords and phrases. Sure, they may check out a movie clip on a social site or read some content from a brand, but they might not be interested in taking action anytime soon (or ever).
Maybe the consumer is not ready to convert the moment they search, either — but they are executing a specific task via the search that can define them as being in a certain funnel step, which the marketer can use to retarget appropriately with relevant content/ads for that point in the funnel.
This type of data has become widely used by audience buyers simple because it is easy to access and buy. Generally, third-party data vendors track site behavior from anonymous users and resell it in categories such as “sports enthusiasts,” “luxury car buyers,” and my personal favorite, “rocker dads.” (Yes, that’s a real segment offered by a third-party vendor.)
However, third-party data has certainly been under fire since its inception. In many cases, it may perform beautifully… but ask any audience buyer how confident they are in third-party behavioral data, and there are questions of recency, relevance, and how segments are defined. A lot of it works, but there are plenty of segments that provide zero value.
Search data coming either from paid search visits to your sites or internal searches on your sites is considered first-party data and is inherently more valuable. Instead of the somewhat hazy definitions and groupings that some third-party data can offer, advertisers can track individual consumers based on their personal keyword choices.
Sure, some people in the same regions share similar traits. I would imagine what’s important to Alaskans in December is different than what’s important to Argentinians at the same time. And if your business is only located in certain areas, it makes perfect sense to target only those areas.
But this is mass targeting. The best practice in digital marketing is granularity, with the ideal of 1:1 individual consumer targeting at scale. Certainly, geotargeting should be considered as a parameter setting that can help the other targeting data on this list be more effective rather than a single-point campaign driver.
Demographic data has some of the same shortcomings as geographic data. You’re making assumptions about groups of people simply because they share an age range, salary range, gender, race, etc. This is also mass targeting, and even two people that share multiple demographic traits have limited value when one searches for “new apple desktop” and the other searches for “new windows laptop.”
Like geographical data, demographic can be useful. However, it’s just not specific enough to individual consumers to rely on as the sole data point driving your audience buys.
It’s hard to argue with conversion data as a solid audience targeting data type. When a consumer converts — whether it’s a sale, a lead form filled out, or another common conversion type — that’s a clear indication of need and a strong predictor of future behavior. It’s an important signal that marketers can use to define targeting parameters.
The scale issue here is worse than with keywords. If your average conversion rate is 1%, that means you have 100 times more keyword data than conversion data.
The other obvious downside here is that these visitors have already converted. They’ve already completed the funnel and are engaged with the advertiser. While current clients are almost always a good source of future revenue, there are many other retention programs (such as email campaigns, loyalty programs, etc.) that focus on this important segment.
Certainly, combining multiple data types together with the consumer’s keyword history makes a lot of sense (site visitation path + the keyword data, geographic or demographic data + keyword data, etc.).
There are virtually infinite potential audience targeting opportunities where the keyword history of an individual consumer can really help audience buyers zero in on the right bid and right ad copy that should be used to drive conversion activity with their efforts.
Most search marketers have either run or experimented with Google’s remarketing lists for search ads (RLSAs) and have found strong results for some of the obvious reasons covered in this post.
And remember, there’s a lot more data that the average paid search account can provide to audience buying teams than just the keyword. The ad copy that the consumer clicked can be very revealing. The bid price or Quality Score of the keyword at the time of the click can also help audience buyers understand the value of a visitor. There’s dozens of other variables that can be used by these teams.
The real question here isn’t how valuable is search data to audience buyers — it’s why audience buyers aren’t using the keyword data more often!
The post 7 Audience Targeting Data Types Versus Search Data (The Keyword Is Still The Undefeated Champ) appeared first on Search Engine Land.