Targeting: How Much is Too Much? Blog Feature
Molly D Protosow

By: Molly D Protosow on May 24th, 2013

Print/Save as PDF

Targeting: How Much is Too Much?

media buying | media planning | media | Digital Media Landscape | Sales Tips | Sales Training | consumer behavior | marketing | marketing strategy | over-targeting | buying

 
Calling all media buyers who are tempted to apply a severe demo-targeting requirement to their next buy, is described by media sellers as "targeting to death." Can you achieve your goals without micro-managing the targeting requirements? From a high level, targeting on a publisher site may appear to be the safest choice. But, in actual practice, it may work against your own best interest.
Are you an over-targeter? Do you find that you shop endlessly for just the right target –audience profile and content and the both together - only to find that no one can hit your targeted impression goals on quality content? Publishers and networks call this targeting to death. A horrible way to go. Often the publisher misses the goal for the first 3 weeks of the campaign month and then a network comes in for a scaled AND targeted BIG week 4. Was this really necessary? More importantly, is it in the advertiser's best interest?
two figures holding up a bulls-eye One of the best things about digital advertising is the ability to target. Apart from contextually targeting users on relevant content, we can reach any demographic through behavioral targeting and past shoppers with retargeting. Let's not leave out geo, day-part as well as targeting by OS, device, browser and even Geo-dem. There are 62 pieces of information that can be known about each user. There are many PC operating systems that we can target and 150 mobile ones. There are more than 30 tablets even if you only heard of 1 or 2 of them.
Given that, how valuable is targeting too tightly in advance of testing? What are we really trying to accomplish – reach a particular target? Or, achieve a particular (and valuable) marketing effect on either branding or performance? That said, over-targeting may put too much focus on the front end of the process and encourage guessing. End result: buyers wanting no-waste in reaching a very specific person --even if there are only 7 of those people and it'd be simpler to call them to let them know you are targeting them for a marketing campaign.
Before locking in on a target, we should consider the opportunity-cost of too narrowly selecting a target …the right audience on exactly the right content…before testing a new campaign on a new site. Below are 4 things to consider before putting a specific demo target in the cross-hairs; today we will focus on #1.
1. Grouping strategy
2. Referring traffic
3. Taxonomy
4. Cost of Scaling
Grouping Strategy – This has an impact on both paid search and display; the clearest way to understand grouping strategy is to start by envisioning a Paid Search campaign. On SERPs (search engine results pages) there is a war going on as marketers fight, not so much for the top of the page but for overall brand supremacy.
All brands know that each paid position has a predictable CTR (click through rate). Their bid cost, in general, will be higher for top positions but can be lowered through factors like QS (quality score). So, improving QS is a never-ending practice. Apart from position and cost-per-click for that position, the marketer has to decide which matching strategy to apply to each keyword: Broad Match, Phrase Match or Exact Match. By the way, the more targeted the advertising, the more expensive yet smaller the inventory will be. Marketers test each keyword with each matching strategy at each position to learn which combination converts most efficiently as well as which one scales best. Why not simply always bid for positions # 1-3? inventory_word_boxes
Because the possible inventory usually exceeds the marketer's budget; the game is about maximizing the impact of their budget. It's about optimizing and scaling. It's possible, given the cost imposed on marketers for being in those top positions, that they will not have enough budget to cover all of the desired hours of the day as well as additional keywords. All of their funds were used up in securing the most coveted (and expensive) keywords, which their peers are also competing for. But the trickiest part of the planning is the grouping of Keyword>Matching Strategy>Creative>Landing Page>Conversion (any desired action or engagement). Each part of the grouping strategy has to be constantly tested to ensure optimal optimization. The first part of this chain is hidden here. It's actually what comes before the keyword that provides the insight. What is the intention of a user who searches with that keyword? Can a different intention be inferred from users using that same keyword across various search engines? Should we show a person with a different intention a differently worded ad? Use a different matching strategy? Bring them to a different landing page?

keywords_word_cloud_square

Now, let's consider how grouping impacts display. We think we are simplifying the game by pre-determining the exact content and demo target and now all we have to do is worry about which creative to serve them. But, let's look at the entire group:
The user's Intention > Choice of Creative > Served on a particular device > at that time > With that geo-target > landing page > Conversion path to desired action.
With all of those moving parts, how sure can we be that the correct target is M 34—39; HHI over $150k, interested in finance, etc…with the ad served when the content is relevant (contextual targeting) or irrelevant (BT)? Even if we worked out the perfect grouping strategy recipe on a different site or network, can we be so sure it'd convert that way again on a different platform we are testing? If we guess wrong on the target, we miss learning what is just outside of our bulls-eye. Maybe the next outer ring would convert better? Or have a higher AOV (average order value) or higher LTV (lifetime value)?
Maybe we should spend more time creating the right landing pages and less time guessing the perfect target? Think about that the next time you are tempted to over-target when testing a new campaign or a new site or network.
Next time, we will look at #2 – how Referring Traffic data may be more important than over-targeting.
 

 

About the Author:

Steve_Bookbinder

Steve Bookbinder is Co-founder and CEO of Digital Media Training, a training partner to some of the most successful sales organizations around the world.  DMT delivers training which treats sales as a competitive sport and changes behavior needed to help sellers consistently win.  DMT is a leader in M-learning training reinforcement with a proven track record of improving sales through training. Steve has delivered more than 500 keynote speeches at national sales meetings, conducted more than 3,000 training workshops and trained, coached and managed more than 35,000 sellers and managers from leading companies around the world for more than 20 years.

 

 

About Molly D Protosow

Molly Protosow is the COO and Training Strategist for DMTraining. She manages the day-to-day business and training operations while helping research and develop new training programs as well as refreshing signature programs to reflect the newest sales trends, technology, and best practices. Molly utilizes her wide-range of skills to create sales and marketing assets focused on delivering value to DMT’s clients. Molly has a passion for learning and leveraging new knowledge and experiences. Outside of DMTraining, Molly is a hard core Pittsburgh sports fan, enjoys staying active by running and golfing, and unwinds by reading and playing the piano.

  • Connect with Molly D Protosow