Posted by 
“We optimize against over 1,000 variables,” boasted the ad tech CEO just before drinks were served at yet another digital media summit. This was the grandest of many “bigger is better” assertions that were peppered into content throughout the day. “50 attention measures,” “33 engagement metrics,” “517 data sources integrated” — the claims clattered from the stage like heavy rain on a tin roof.
Spoiled for choice
Don’t get me wrong. I love data as much as the next digital media and advertising guy, but I am growing increasingly wary of the information arms race.
The proliferation of options takes us farther away from landing on digital metrics that matter. It certainly adds complexity to an already confused market. It also provides the perfect impetus for laziness; why narrow your focus to a few choice metrics when you can hide behind an inglorious pile of stats?
I can understand why many people adopt the “bigger is better” mindset. And if you measure everything, there is bound to be at least one metric that glamorizes your campaign’s performance. A shotgun approach more often hits its target, doesn’t it?
In the end, we’re all salespeople peddling our advertising to clients, and it is often harder to sell “less is more.”
Of course, more is not better in every case, and definitely not with digital metrics.
On its own, the applied focus necessary to determine the right campaign metrics has enormous benefit, as it requires coordinated, concentrated effort by all stakeholders — the marketing lead and their consumer insights partner, ad agency and research provider. In many cases, this key group will not even meet before a program is launched, let alone meld minds to agree on what success looks like ahead of time.
Metrics that are clear to stakeholders and have been validated allow for more impactful optimization during a program and more effective evaluation afterward. Trying to bake in too many metrics weakens your ability to optimize digital’s iterative approach and can fuel a virtual cycle of superficially enhanced performance.
There are three key questions to ask yourself before adopting a metric:
Are you crystal clear on what it means?What specific event or consumer behavior is being counted? If you don’t know or can’t explain it, then it should not be on the list.
Is this a number you trust, and is it trusted by the industry?
Can you show, or even argue plausibly, how it is related to concrete objectives or business outcomes?
In truth, it almost does not matter which measures you choose; the process of narrowing one’s focus, learning and refocusing will lead the diligent practitioner to the metrics that matter for their brand.
That said, there is a proven core list from which to start — and you can count its numbers with two hands:
Keep it simple
1. Viewable impressions: This is the lowest of the bars that effective advertising needs to get over, and developments in recent years have made it a hot topic in digital advertising. A viewable impression is one that the audience has a proven opportunity-to-see.
In traditional media, opportunity-to-see (OTS) is the subject of little debate. For the first 20 odd years of digital advertising, however, there was no accepted definition. The industry solved this challenge over time, now publishing viewable impression measurement guidelines.
2. In-view time: The length of time an ad has been active and in view. For some ads on high-quality editorial sites, this could be minutes, while on other quick-hit pages (e.g., checking the weather), average in-view time could be closer to the one-second viewability threshold.
There is a growing body of evidence suggesting that in-view time has a strong correlation with brand effectiveness, so the longer an ad is in the viewport, the stronger the brand lift will be.
Yahoo Research’s Daniel G. Goldstein, R. Preston McAfee, and Siddharth Suri make a case for this in their research, saying that “… for exposure times of up to one minute, there is a strong, causal influence of exposure time on ad recognition and recall, with the marginal effects diminishing at durations beyond this level.” Digital attention analytics firm Chartbeat reported similar results, saying that when visitors “read more … they’re more apt to recall the brand that advertises next to that content.”
These first two metrics reveal how many chances an ad has to do its job. Marketers must also measure if an ad can break through the digital advertising clutter, if the viewer can recall it and if it’s associated with the sponsor brand with a properly branded impression.
3. Visibility: We live in an attention economy. You can measure visibility by asking people in a survey if they recall seeing an unbranded visual representation of the ad (in other words, any brand mention is removed from the visual stimulus).
4. Brand linkage: Standing out has no value unless viewers identify the ad with the brand. You can measure this by asking respondents in a survey if they can name the brand that’s sponsoring the unbranded visual representation of the ad. It seems so simple, but it is also incredible how often this fails to happen.
According to the Ipsos (my employer) database, “Half of all TV ads fail to achieve objectives due to weak branding,” i.e., people are unable to recall which brand the ads were for.
5. Universal interaction rate: The percentage of impressions in a campaign where a user entered the ad frame and remained active (not parked) for an agreed period (the IAB suggests 0.5 seconds). This is a basic measure of physical attention.
6. Universal interaction time: Average length of time the user interacted with the ad.
Visibility, brand linkage and interaction rate and time can be viewed together to get a picture of an ad’s capacity to stand out from the digital clutter, but not if the ad elicited the desired response — a change in intention, behavior or attitudes toward the brand.
7. Persuasion: This is about behavior change. Are consumers more likely to buy the brand more often or use the brand more frequently after being exposed to the advertising? This is often characterized as a likelihood to purchase, consider, test drive or sign up. The frequency of purchase and usage will give you a better understanding of the persuasion of an ad.
8. Brand favorability: a measure of relationship change. Do people feel more positive toward a brand after they experience the advertising? Often will often look at two components: affinity (brand closeness) and relevance (how well the brand meets a consumer’s personal needs).
9. Retransmission: Does the viewer feel strongly enough about the message and brand to share their feelings? There is ample evidence that recommendations from family and friends are the most credible form of advertising.
Leading marketers work to ensure that people who see their messages become part of their media team by creating ads that get discussed, shared or passed along online or off, and they measure this effect.
10. Sales: I would be remiss if I didn’t include proof of tangible sales on this list, as this IS the whole point. The CTR is all but dead, and, while attribution modeling has advanced significantly in recent years, the industry is still pursuing the holy grail of being able to reliably and universally tie brand ad impressions directly to sales at scale.
Less IS better
After choosing the handful of metrics that matter to you, I urge you to resist those more-is-more folks who will certainly come calling with must-have measures to incorporate into your analyses.
Just say no — or, better yet, challenge the yaysayers to focus on the one, two or three that actually mean something to your campaign.
Some opinions expressed in this article may be those of a guest author and not necessarily Marketing Land. Staff authors are listed here.
