Targeting
Difficult to Find Consumers
The Challenge
A leading fashion retailer was looking to engage a very
specific target:
- Hispanic Females
- Age 18 to 49
- Income below $40,000
- Spanish speaking dominate
- Living in a major metropolitan area within 12 zip codes
The Strategy
- Geographic targeting to isolate the desired zip codes
- Demographic targeting to identify Hispanic Females, age 18 to 49, with income of $40,000 or less
- Target web sites primarily with Spanish Content
- Overlay behavioral targeting focusing on shopping interests
- Use Keywords and Phrases to dynamically insert ads into articles that contained content matching the Keywords Utilize Mobile Advertising
A/B Testing
Two sets of creative were used:
- One set of creative was focused on branding with a special price on a fashion item
- The second creative promoted a Register to Win a Valentine’s Giveaway
- The campaign was launched with 50% of the impressions targeted to desktop, 50% targeted to mobile
- The creative message between the two platforms were identical
Results
The first objective
was to out perform a national average CTR of 0.05% - 0.08%, then optimize
weekly towards better results.
After the first week
we learned:
- The Mobile response far outperformed Behavioral, Content and Keyword Targeting
- Average CTR for the overall campaign was 0.40%
- Mobile 0.46%
- Keyword 0.16%
- Content 0.11%
Creative
Across both platforms
the “Register to Win” creative performed well:
•Mobile 0.38% CTR
•Desktop 0.37% CTR
However, the Mobile platform (0.37%) clearly
outperformed the desktop (0.12%) with
the branding creative
Findings
The mobile results clearly validate the research showing the propensity of Hispanics to use mobile access for the internet over desktop
- However, the creative using “Register to Win” resonated with all consumers
- The branding creative works best on the mobile device
The PNS Digital optimization reallocates impressions
based on predetermined results. In this case CTR. While the initial plan was to
allocate the impressions 50% / 50% between desktop and mobile, the short
campaign timeline (7 days) required us to push impressions to the mobile
platform due to superior CTRs.
From a research prospective it would have been
interesting to see if the results held up if we forced the 50% / 50%
distribution plan. But campaigns are measured by results, not research!
More to follow as the campaign unfolds.
No comments:
Post a Comment