Achieve Media Goals Without Reviled Ad Techniques

I offered an alternative opinion to Jakob Nielsen’s “The Most Hated Ad Techniques” last column. Feedback ranged from support to strong objections. It definitely hit a nerve with some readers.

Yet if a site publisher offers an effective ad placement, and the ad creative is well executed, even ads that tend to offend can do a very good job. When my agency is responsible for achieving an advertiser’s goals, we’d be remiss if we didn’t at least consider proven techniques. Doing so doesn’t mean I personally like or endorse them; I’m acting in my client’s best interest.

What if an advertiser tasks its agency with building a media plan devoid of these hated ad techniques, yet still requires the agency to meet campaign objectives within the same ad budget? We’ve got a challenge. Though branding campaigns have quite different objectives and more flexibility than direct-response campaigns, in both cases the media buyer must achieve measurable objectives. Even the world’s largest brand advertiser, Procter & Gamble, favors a media-buying-led strategy over an ad-creative-led one.

Can we buy for a direct-response campaign without offensive ad techniques? Let’s find out.

Direct-response media campaigns have a lot of variables. Sometimes, it feels more like algebraic problem-solving than media buying:

• A = budget
• B = campaign deliverable objective (e.g., visitors, leads, sales, etc.)
• C = anticipated ad performance per placement type
• D = anticipated ad performance per site
• E = type of placement cost (cost per thousand, click, or action)
• F = quantity of ad placement purchased per site
• G = cost per ad placement per site

Just to drive traffic to a site, a formula might look like this:

Traffic = A/[Placement 1: (E1 x F1 = G1)(C1 x D1)] + [Placement 2: (E2 x F2 = G2)(C2 x D2)] + [Placement 3: (E3 x F3 = G3)(C3 x D3)] + [Placement 4: (E4 x F4 = G4)(C4 x D4)]

For the sake of this analysis, we’ll ignore conversion rate. (Though it’s an important variable, I can’t control for it.) Here’s the scenario: My campaign objective is to get 100,000 visitors with a total budget of $60,000. Let’s say Placement 1 would’ve been pop-ads had the advertiser not prohibited them.

At an average click rate of 2.85 percent for a CPC (define) buy of $10,000 with a $5.00 average CPM (define), I would’ve bought 2 million impressions and gotten 57,000 clicks. Suddenly, I have to change gears because of the advertiser’s ban on pop-ups. What are my alternatives, using the same dollar allocation and visibility on the same kind of site?

• Banners: With an average click rate of 0.25 percent and a $22.50 CPM, $10,000 buys me 444,444 impressions, or 1,111 clicks at $9 CPC. Way too expensive, and not nearly enough traffic for the cost.
• Targeted opt-in solo email: With an average click rate of 1.5 percent and a $125 CPM, $10,000 buys me 80,000 deliveries, or 1,200 clicks at $8.33 CPC. Fuggedaboutit.
• Newsletters: With an average click rate of 1.0 percent and a $50 CPM, $10,000 buys 200,000 deliveries, or 2,000 clicks at $5 CPC. We’re coming down, but we’re nowhere near as cheap as those pop-ups.
• E-mail classified ads: With an average click rate of 0.75 percent and a $20 CPM, $10,000 buys me 500,000 subscribers, or 3,750 clicks at a $2.66 CPC. Wow! Our lowest yet.
• Search: With an average click rate of 4 percent, CPCs are all over the place (and, yes, some do exceed the $9 banner CPC). But these visitors tend to be big-time qualified visitors. That’s why the return on investment (ROI) is so good with search campaigns.

Media buyers must achieve certain objectives. Without certain placement types in our arsenal, we have to get creative, dig for other placements, negotiate, and play hardball. This takes more time. More time means less profit, particularly on smaller buys. Yes, it’s possible to achieve campaign objectives without hated ads. But with campaign profitability and ad placement performance driving media-buying decisions, particularly for direct-response marketers, agencies certainly face a challenge.