We sat down with Dr. Stephanie Gutnik to get her views on the current state of DOOH advertising.
Stephanie is probably the first and so far, the only marketing practitioner and academic on earth who successfully defended her doctoral thesis in connected TV and out-of-home advertising.
DSPulse: What is Billups?
SG: Billups is an OOH managed services agency (the largest independent OOH agency in North America). We have built a proprietary tech platform that allows us to seamlessly plan OOH/DOOH media campaigns. We can also buy DOOH programmatically through our own DSP and partner DSPs. Importantly, our tech platform also includes patented attribution capabilities to measure upper through lower funnel KPIs.
What holds the DOOH industry back from gaining a bigger share of ad budgets?
DOOH is an emerging medium that is part of a traditional channel, meaning, despite its novelty, it is often treated how OOH has long been treated by marketers. Which means that pre-existing perceptions of OOH being inflexible or unmeasurable need to be changed. This is no small feat.
I do believe that the more we continue these conversations and demonstrate D/OOH’s effectiveness across a plethora of KPIs, we will start to see a bigger share. And perhaps even a fair share – I have reviewed many studies that show how hard D/OOH works for the money spent.
These studies indicate that an optimal level of D/OOH spend in a campaign is typically at least double of what had been originally budgeted.
How would you assess the current state of measurement in DOOH?
It has come a long way over the past five years but there are still hurdles.
In some cases, a higher minimum spend is required to generate the kind of scale necessary to ensure proper incremental measurement via device ID.
This doesn’t help change the opinion of marketers who are approaching DOOH with a test budget not big enough to be measured. This approach reinforces their perceptions of the channel being immeasurable.
A big reason why I joined Billups is the work being done by the Sciences team.
Patented, machine-learning-based technology has been created to understand viewability of the D/OOH faces, the elements most likely to be seen by viewers, what might be blocking the faces and for how long the faces will be viewable to a consumer.
We pair this with a scientific understanding of viewers’ journeys – and attention metrics are born, which is a big topic in the rest of the media landscape. Attention studies have demonstrated a positive correlation with valued KPIs.
This is also the foundation for privacy-compliant, cookieless measurement from upper to lower funnel. So, I was impressed by Billups’ measurement methodology before I joined. And the more I learn about our roadmap and methodology, the more excited I become to share these capabilities with marketers.
Are audience measurement and campaign effectiveness measurement in DOOH possible without tracking people’s phones?
Yes! The OOH industry has understandably become highly reliant on mobile data to measure outcomes – but Billups’ measurement methodology is designed to function without device IDs. Again, a reason why I was so excited to join the team.
We use fully consented and anonymized data, whereby each device is assigned a unique encryption key, and the data is encrypted within this key. The key is fully deleted if someone opts out.
Our measurement system reconstructs each device’s daily journey but only stores interactions between the OOH unit(s) seen and a small set of locations being measured for visitation, none of the intermediate is stored. This way, we understand how OOH affects a device’s behaviour at a unit level but know nothing else aside from basic demographic information.
Here are some of the methodologies we are using:
– Cookieless hashed email measurement: we match a client list of hashed email addresses with pseudonymized IDs and run our standard control vs test groups to measure the difference in behaviour between those unexposed and those exposed to the OOH unit(s).
– Causal modeling/counterfactual analysis (aka our CTO and the Sciences team are geniuses): the first attribution study designed at Billups many years ago can be run anywhere in the world. It is based on using predictive Bayesian time series analysis and measuring the difference between what happened in the market with OOH present and what would have likely happened in the same market if OOH was not present.
– Media Mix Modeling: we can help clients look at the effectiveness of their entire media budget across channels, not just limited to OOH.
Should DOOH remain a one-to-many (mass medium) or become a one-to-few medium? These two trends seem to be mutually exclusive.
It’s up to the advertiser to decide how the medium will best serve their campaign KPIs.
For example, a brand-building campaign for a new product or service will greatly benefit from DOOH locations and creative that reaches target audiences and other consumers the brand does not yet know would be interested.
At the same time, there are verticals where DOOH does function as a one-to-few medium, such as in the back of a taxi/rideshare or even at a restaurant or bar. In these instances, the viewer dwell-time is significant, and the opportunity arises to add a QR code in the creative to generate lower funnel actions.
As CTV starts to explore co-viewership metrics, it begins to mimic DOOH’s impression multiplier in a one-to-few situation. The opportunities for budget share shift to enable more scale across premium video supply and reach across incremental audiences cannot be understated.
Why didn’t bots, fraud and the cookieless future in online advertising result in more dollars for DOOH, as was widely expected?
A study was recently published about the billions of dollars being wasted on invalid and fraudulent traffic in the advertising industry.
It is tempting to assume that this “mismanaged budget” should be moved into a channel like D/OOH.
I have similarly had questions from folks in the industry when social media platforms face scrutiny over various issues, who reason that any paused budget will be moved to D/OOH. I wish it worked this way!
It’s the same reasoning as in my first question – we need to be working closely with marketers in-house and at agencies not only to educate them, but to help reduce the risk of moving budget plans from what is familiar to what is unknown, and thus add more risk for an outcome that has not yet been tested for certain marketers.
It is one thing to talk about taking risks and promoting innovation, but it takes a lot of time and hand holding to make it happen.
This is human nature, and it’s a motivating challenge to have when you believe in the D/OOH channel as I do. I am confident the risk will be worth it for marketers who make that leap.
Why is programmatic DOOH not growing as fast as predicted?
Currently, programmatic DOOH is just that: it’s an automated form of buying digital OOH inventory. Digital OOH, while in growth mode, is still not as popular as traditional OOH. So we are looking at a fraction of a fraction.
It’s fantastic that we can buy DOOH programmatically, as it levels the playing field with other channels. However, there are still aspects of programmatic DOOH that are different from other channels.
For instance, creative needs to be manually approved (this is an awesome feature in support of brand safety that other channels can sometimes struggle with).
Buys can still be location-based (in addition to audience- and data-driven), and to a buyer unfamiliar with the medium, the availability of many publishers and verticals can sometimes lead to decision paralysis.
In this respect, specialist OOH agencies and platforms bring a lot of value to buyers because they can guide them through all the above nuances. And a buyer may still prefer to transact via insertion order based on the parameters and needs of their campaign.
It comes down to what is easiest for the buyer and what transaction method makes the most sense for the campaign.
Because OOH is still so different from other media, expecting sales generalists to understand all its intricacies is a tall order.
Specialist agencies and platforms take this all on and can easily/effectively work with the other agencies and platforms involved in the marketing plan. It just makes more strategic (and thus economic) sense.
Just five years back many thought that specialist agencies would disappear. Today we see that specialized knowledge will only increase in utility for marketers.
Does Billups measure campaign effectiveness? If so, how? Could you give a few examples from real campaigns?
Yes! Please find more information at Billups Analytics. Brand performance, website conversions, app downloads, location visitation…
Some case studies: Varo Bank, Postmates (I loved this one when I was living in NY and not yet at Billups) and Salonpas) (a programmatic example).