For much of the digital advertising era, the industry has raced to the bottom on price and quality.
But in the past five years, more advertisers have been explicitly incorporating media quality into their measurement and buying practices. As buyers increasingly prove that high-quality media delivers better value, even at premium prices, the market for quality-based programmatic solutions grows.
Perhaps the most notable success of this shift to quality has been the rise of attention metrics, which even Google and YouTube are now embracing. But the new focus on quality also includes other forms of media, as well as creative and audience/data quality assessment. It also encompasses custom algorithms and effectiveness/incrementality measurement that extend beyond last click or multi-touch attribution (MTA).
Incorporating measures of quality into our existing practices is just the first step. A bigger challenge looms: assessing media quality against a marketer’s short-term and long-term goals.
Predigital paradigms fall short in a postdigital world
In the predigital era, most channels were distinctly better for brand impact (upper funnel, longer term), while others were more suited for direct-response impact (lower funnel, short term).
The digitization of all channels inherently broke that paradigm. Now, every pixel on every screen can serve different needs based on a multitude of confounding variables, from location and time of day to device type and identity or audience signals that indicate intent and affinity.
Predigital ads could be planned, bought and measured primarily according to their anticipated impact in the short or long term. But today, nearly every digital ad’s value must be measured in both time horizons.
This challenge is not new to marketing, as Les Binet and Peter Field argued in their 2013 book, “The Long and the Short of It.” But it wasn’t top of mind for the vast majority of media-buying decisions in the past.
For example, the impact of TV ads was mostly measured in the medium and long term (such as increased brand awareness or market share) but occasionally assessed in the short term using proxies like spikes in offline/online traffic or hotline calls to infer future sales.
Today, while it’s possible to measure the short-term impact of most digital ads (though proving true incrementality remains elusive), much of an ad’s value is often realized after the short term. Even if people don’t immediately convert, those ad impressions might sway them eventually.
Defining short- and long-term impact
Let’s examine how buying and measurement that accounts for both short- and long-term effects could play out in a traditional brand awareness channel: TV, now digitized as CTV.
Picture a midrange auto brand releasing a new electric SUV. It’s running a national CTV campaign. But it’s also intensifying efforts in geographies that index highly for EVs and SUVs, as well as on device/household IDs that correlate with greater intent/affinity. This campaign should be measured for its immediate impact on in-market consumers, as well as through proxies for long-term impact on those not currently in-market.
For near-term impact, the marketer will track an uplift in leads, test drives and sales using attribution methods that connect ad exposures to these (ideally incremental) outcomes. Probabilistic methods and geo-based experiments will likely play a role, too.
The marketer may not know how much the ad served to IP 123.456.789 actually contributed to that household’s car purchase within 30 days, but they can evaluate whether campaigns with certain attributes outperformed others across different geographies.
For longer-term impact, the marketer can evaluate proxies that indicate whether the ads improved mental availability or the consumer’s ability to recall the brand when making a purchase. This is especially important for those not in-market during the campaign. Such proxies include share of search, brand lift of awareness and consideration, etc. The marketer would also rely on traditional proxies like reach and frequency, but adjusted to reflect media quality across dimensions such as time of day, day of week and channel.
Ideally, campaigns are structured to allow experiments across geos with different media, audience or creative treatments, enabling measurement of various short- and long-term effects.
Over time, those findings should be baked into that brand’s best practices to drive the right balance of uplift based on the business needs & campaign strategy. This results in a CTV campaign aiming to get <5% of exposed, in-market users to take immediate action by prioritizing media proven to drive that action. The other 95% should see ads in placements where the price aligns with value based on predicted attention and receptiveness (aka quality) levels. This increases the probability that they’ll remember the brand when they enter the market later.
Automating the solution
That’s just one example of how to weigh short- and long-term impacts when assessing media quality. But the real challenge is building systems that do this at scale across the vast majority of buying and measurement decisions.
As the industry shifts toward more automated systems, we must develop technologies, processes and best practices that avoid the same old mistakes. We must do all this in a way that appropriately considers the unique circumstances of each brand, campaign, objective and media mix. And we can’t allow the ecosystem to continue to prioritize short-term impact at the expense of long-term impact, such as with “AI solutions” that optimize toward ROAS (which is a flawed metric for many reasons and is also often measured in a single time frame).
Reaching this point – eschewing vanity metrics and non-incremental attribution models in favor of quality and incremental effectiveness – was a formidable challenge. But it felt inevitable.
Convincing marketers to value long-term results in the same way they do short-term impacts, and to do so concurrently? That’s arguably the bigger challenge. But it’s the only way to mitigate the risks of short-termism and build lasting success.
“Data-Driven Thinking” is written by members of the media community and contains fresh ideas on the digital revolution in media.
Follow Erez Levin and AdExchanger on LinkedIn.
For more articles featuring Erez Levin, click here.
