Managing the Frequency Curve

• Tweet
• Buffer

During 2005 planning, I've been surprised by the number of questions surrounding online ad frequency.

Now that online allows us mad-scientist media types to adjust the ad frequency dial on a whim, I thought we'd moved on to sexier frontiers, such as behavioral targeting, audience clustering, and search's branding impact.

Alas, good media topics often return for more dialogue and exploration. Some clients see this as fertile ground for further exploration and optimization.

How can frequency be utilized to maximize an online ad program? That, of course, depends on the ultimate campaign goal. Some possible reach goals include:

• Deliver message to desired percentage of intended target audience (reach) by average number of times (frequency).
  
  
• Establish the right number of ad exposures based on the optimal number of user ad interactions, click-throughs, sales, or site transactions.
  
  
• Base results on optimal post-view events or transactions (post-view measurement often counts for over 60 percent of an online ad campaign's total impact).
  
  
• Determine the optimal exposure level to increase brand awareness, message association, consideration, preference, or purchase intent.

Depending on your goal, you can take one or more of the following approaches:

• Cap incremental impressions served to unique visitors to maximize the frequency level within individual sites or properties and minimize waste.
  
  
• Monitor frequency by ad placement, unit size, or content section within individual sites.
  
  
• Cap impressions per user session.
  
  
• Establish an effective frequency level, then deliver another relevant ad message to the same user.
  
  
• Measure effective frequency across parts of an ad buy, such as content categories or site types (portals versus news versus blogs, etc.).
  
  
• Measure effective frequency across the entire campaign.

A Cautionary Note

If you want to pull the plug on a campaign because it didn't reach the optimal frequency level in the first two weeks, drop the cord. You're monitoring the wrong metric to achieve a different result.

Online media require a reasonable time to develop an optimal reach curve. Remember, we're talking about a media plan based on the consumption of page requests. Users can often access dozens of unique sites. Don't be quick to pull the plug unless you only want a quick measurement timeframe, especially with smaller-scale efforts.

Will Patience Pay Off?

We've been amazed at the number of campaigns we've run where optimal frequency occurs after 8-10 impressions, as well as how many times brand impact scores have achieved high percentage levels from just one exposure. Take the time to evaluate several scenarios before you draw definitive conclusions.

As you measure and compare campaigns, capture and benchmark data (even for brand impact measurements) at several interval levels, such as 1, 2-3, 3-5, 5-8, 8-10, and so on. Though it may seem like extra work, the insights are well worth the effort. You'll more fully optimize future flights and campaigns.

Next time a client asks, "How many times should we expose the ad to a user?" you can offer a sophisticated answer, as with all those other hot topics in the world of online advertising.