Digital advertising has, and will continue to be, one of the most significant and impactful revolutions in campaign strategy since the invention of broadcast television. By giving campaigns the ability to target highly tailored messages directly and exclusively at target voters, online adverting platforms have forever changed how campaigns are (or should be) run.
But even in 2016, campaigns are still failing to execute effective digital targeting on a daily basis.
I’m not talking about understaffed, down-ballot campaigns. I’m talking about top-tier US Senate, US House, and statewide campaigns that are paying big money on digital ads as a central part of their strategy. The result is potentially millions of dollars in wasted or mis-spent ad money this cycle that could have but likely didn’t make a meaningful, measurable impact on the election.
Below are a few specific examples—just from this week—from my personal Facebook news feed.
Example 1: The general brand awareness ad
Maryland Delegate Kathy Szeliga is perhaps the strongest Republican candidate for US Senate in recent Maryland history. She’s sharp, she’s hardworking, she’s rallying the party around her campaign, and she’s running at a time of unprecedented opportunity for a Maryland Republican. I plan to vote for her in Maryland’s April 26th primary election.
Unfortunately, her digital advertising has been consistently un-targeted—causing her campaign to hemorrhage resources on inefficiency while she fights a tense primary against Chrys Kefalas.
For the past several weeks, Kathy’s campaign has been running this very sharp and very engaging Facebook Canvas ad. It’s cutting edge stuff, and her team should be proud of the way it’s put together. But unfortunately, the targeting for the ad is causing it to under-perform.
Unlike some ad platforms, Facebook will tell you why you’re seeing an ad. Just tap in the down arrow in the top-right corner of a post, and then on “Why am I seeing this ad?” Facebook will give you a breakdown of the criteria the advertiser used when placing the ad, and allow you to opt out, etc.
Note: Per a Facebook rep, the “why am I seeing this ad” prompt will show what Facebook thinks is the most important reason why you saw the ad. There may be other factors involved, but what it shows is the primary one.
Kathy’s ad, as you can see in the screenshot above, is being targeted shown to me because they are targeting “people 18 years and older who live in Maryland.” That’s it. Not specific issue or interest criteria. Not because I am in a custom audience they created. Not because they’re targeting specific voters. Just all Maryland adults, everywhere.
Lets break down the numbers:
- The Census estimated that as of July 2015, there were 6,006,401 Marylanders, of which 22.6% were under 18 years old. That means there are approximately 4,648,954 Marylanders eligible to see that ad on Facebook — assuming every adult in Maryland has access to Facebook.
- By contrast, the Maryland Board of Elections says that as of 04/22/15, there were only 631,632 registered Republicans in Maryland statewide.
- If you do the math, that means that only ~10.5% of the people who could see that ad are eligible to vote in the primary. Put another way, 89.5% of the money being spent on that ad is being wasted on audiences who cannot support her in the election when it counts.
Unless her primary goal is to increase her statewide name ID across the board, rather than to win votes in the April 26th Primary Election, this is an inefficient and wasteful way to target ads.
A few alternative ad universes may have included:
- People who like popular incumbent Republican governor Larry Hogan on Facebook.
- A “lookalike audience” of Marylanders who are similar to her current supporter, donor, or volunteer lists.
- A custom audience of targeted voters based upon past primary election turnout modeled on the voter file.
Example 2: The fundraising ask
US Congressman Dr. Joe Heck is the frontrunner for the Republican nomination for US Senate in Nevada. For more than 35 years he has worked as a community volunteer, physician, firefighter, search and rescue team member, ambulance attendant, SWAT physician, and member of the Army Reserve. He’s a fantastic candidate that conservatives can be proud of, and if I were in Nevada I’d be voting for him this year.
Earlier this week, one of Dr. Heck’s Facebook ads appeared in my timeline, informing me that a generous donor had agreed to match all contributions and asking me to make a donation. Again curious why I was being targeted, I checked with Facebook and saw this:
So lets break this down:
- According to the US Census Bureau, there are 244,278,303 adult persons in the United States of America.
- According to the Nevada Secretary of State there are 1,459,304 registered voters in Nevada as of last December. Of those, just 488,608 are registered Republicans.
- So assuming all adult Americans and all registered voters in Nevada are on Facebook (which is obviously not true, but work with me), just 0.200021% of this ad’s audience was comprised of Nevada Republicans.
I was certainly not in Nevada when I saw the ad. I was in California, and my Facebook profile has my address listed as living in Maryland. I’m also not an engaged Heck supporter, past donor, or otherwise someone common sense would dictate a likely donor. Why would anyone assume that I might be a good target for this ad?
Now, I’m willing to concede that in competitive statewide or federal elections donors and supporters often come from out of state, so advertising out of state can — if done smartly—make sense. But this doesn’t make sense, and for two reasons.
- For the vast majority of online donors, donating is not their first engagement with the campaign. The best practice is to move supporters up a ladder of engagement whereby you incrementally work to get them more emotionally invested in the campaign before ultimately making a donation or volunteer ask.
- To facilitate that “engagement ladder” approach, you’d want to target a donation ask at known supporters who you believe are ready and likely willing to make a donation. This ad makes no attempt at that.
Instead of blindly targeting all American adults, the campaign may have considered a few alternative ad universes:
- People who already like / follow Dr. Heck on Facebook.
- Known supporters on the Congressman’s email list — particularly those who’ve regularly opened email blasts.
- A remarketing audience targeted as people who’ve visited the campaign’s website—particularly those who have visited the donation page but not donated.
The first full sentence on Facebook’s own marketing page about ad targeting says: “Get specific about who you’d like to reach with your Facebook Ads.” Get specific. Hyper-target your content to audiences and communities who will be most likely to engage with it. You’ll have a greater chance of success and spend less money.
Again, this isn’t a criticism of Del. Szeliga or Rep. Heck, both of whom are running great campaigns in general and for whom I’d be proud to vote. Perhaps these ad examples were exceptions to the rule; perhaps they’re regularly doing exactly the sort of targeting I’m advocating, and this was merely a test. I hope so. I’m sharing them because I think that — tests or not—they’re Class A examples of what not to do when running digital ads.
Next week, I’ll publish a few counter-examples of great digital ad targeting (none of them from our own ads) to show the alternative perspective. Great digital ads zero in on quality audiences based upon their location, demographics, interests, behaviors, and connections. If your ad universe doesn’t factor in at least three of these angles, you’re doing it wrong.
Digital & Field is published by Hines Digital — the mission-driven digital agency committed to helping conservatives around the world organize grassroots communities, raise money online, and run smarter campaigns. Check us out at www.hines.digital.