"On TV And Video" is a column exploring opportunities and challenges in programmatic TV and video.

Today’s column is written by Michael Goefron, managing director of account management and chief brand safety officer at GlassView Media.

The banner ad has been decried and even its creator seems dubious about his invention, but let’s give this much-derided ad unit its due: It knows when to quit.

Or, more accurately, those who buy banners ads know that there’s certain frequency to which the law of diminishing returns sets in, and such calculations are baked into display ad campaigns. Yet, oddly enough, there are no such safeguards for video advertising, as most online video campaigns are not optimizing for frequency.

It has long been known that optimizing for frequency is a key to successful advertising. In the late 1800s, Thomas Smith, a London ad man, wrote a book called “Successful Advertising,” which advocated 20 times. Later, Herbert Krugman, a GE exec, decided that three was the magic number. TV campaigns have long recognized and optimized for the fact that a viewer must see an ad multiple times before it makes an impact.

Conversely, serving an ad too frequently can have a reverse effect and create viewer fatigue. This means that that funny “Clash of Clans” ad with Liam Neeson, in which the acclaimed actor utilizes his signature dramatic snarl to threaten a competitor while playing the game, might not be so amusing on the 23rd viewing. It can also account for why it can become annoying when Hulu, which recently introduced an ad-free paid subscription version, serves the same commercial two or even three times in a row during a single break – and then again during the next break.

Perhaps because the online video segment is comparatively new, frequency optimization of ads isn’t normally set. Just because the industry hasn’t yet embraced frequency controls for online video ads doesn’t mean it shouldn’t.

Cap It, Set Goals

From the initial planning stages, advertisers and video platforms alike should work to put a cap on video delivery from the start. Even if the cap is set at five times per user per day, brands paying on a cost-per-view basis will avoid excessive waste, as well as the awkward realization that a video campaign could be having the opposite of the intended effect. If frequency caps are not set, advertisers run the risk of driving annoyance rather than affinity.

It’s also essential to set frequency goals based on ad classification. There are two primary classifications of online video advertising: direct response and branding. However, within each there are many sub-classifications.

In direct response, for instance, advertisers might be trying to get someone to merely sign up for promotional materials, which is common with B2B pitches, or to contact a sales rep or click through and buy. Advertisers should be realistic about their expectations for each. For instance, the average click-through rate (CTR) for banner ads is 0.08%, meaning one person in 250 will click through, according to Google.

It’s important to consider historical data. Monthly, quarterly and even yearly trends on frequency can help yield smarter optimization moves. For example, higher frequency for CPG products around the holidays or holiday charitable giving campaigns are almost expected. For those kinds of campaigns during those months of the year, looser frequency caps would be much more acceptable.

When the collective hangover from the holidays sets in January or February, however, people may become more averse to these types of messages. This make it more important for advertisers to set new frequency limits for messaging their target audience during those months.

How’s It Working?

Advertisers must monitor and assess frequency in real time to determine how their campaigns are doing. Chances are a video ad campaign will beat the banner ad CTR.

However, it’s valuable to learn when people are clicking. Are they clicking after the first exposure, for example, or the third? If marketers look at the data and see a substantial cluster around five exposures, that’s their number and they should adjust their frequency accordingly.

They should do this while the campaign is happening, not afterward. Third-party services such as comScore allow marketers to measure frequency levels as the campaign is taking place.

Since playing the same ad over and over may create viewer fatigue, why not expose the consumer to different but related or sequential ads? In a branding campaign, for instance, marketers might want to evolve the creative to tell a story arc that runs over several ads.

In direct response, advertisers might want to move the consumer further down the purchase funnel from, say, consideration to preference by showing how their product stacks up against the competition.

Frequency is one of the most underrated levers marketers can pull from an optimization standpoint. It’s time for the video ad segment to get serious about frequency, which should be a primary consideration for any online video ad buy.