- Can you dive deeper into how your technology detects the turned-on devices, and if it’s passively collecting device IDs, do/will opt-in laws affect you (even if no “user data” is collected – how do you have location permission to view this device at all)?
Our product detects wireless device communications over the air which are open for reception. Unlike wired technology where you can send a signal to a specific device, wireless technology broadcasts messages which every device listens for but only processes those meant for its reception. It is somewhat similar to TV broadcasts: All stations broadcast simultaneously but the TV only processes the selected channel for display. Our product listens to all broadcasts tracking its "source" which is used by the destination in order to reply.
As the technology does not access the contents of the broadcast, we are not violating communications privacy so opt-in is neither a requirement nor issue.
- Do you report on total impressions and unique impressions?
Yes, our device tracks both total impressions and unique impressions. We also track frequency of impressions, how many times a particular impression viewed the advertisement, the duration of an impression, and how long that impression viewed that advertisement. Last but not least, we are also able to track attribution and conversions of impressions into the brick and mortar location (car lot, model home, voting booth, mall store-front, etc.).
- If you are collecting unique device IDs, and if we could theoretically take those device IDs and place them in an ad segment, then how can opt-in not be an issue? (Even if that’s not your practice to do so.) And what if location services are turned off on the device, can you can still detect it?
We do not share specific MAC address information with clients. We collect only the impressions, frequency of impression, duration of impression, and conversion of impression of those MAC addresses. No information on the direct customer is provided or captured. Only the device itself. In sum, our internal database does not hold any client specific data about the person holding the phone so opt-in is not something we need.
- Do you have any way to tell the story to clients of a person with two phones?
Unfortunately not. Because Abraxas cannot legally tie a mobile device to a specific individual, we cannot tell a story of that type of person.
However, we can pull data that shows the identical patterns (impression, duration, frequency and conversion) of two phones. With those analytics, we can then start to paint a picture of several scenarios; e.g. are your children always with you and does one of your children have a cell phone, etc. It's highly unlikely, however, that we'd see the same exact pattern from two different MAC addresses every single time coming from two different people. Our internal AI system uses machine learning and can piece together a likely scenario of this kind of identical pattern from multiple devices.
- You state that you pick up cellular signals with nothing on the device. But then you state that you can pick up signals from all WiFi devices – does the WiFi need to be on?
No. Wifi does not need to be on. Only the device needs to be on.
- What if the device is roaming or in a poor reception area e.g. at a conference?
The only requirement for our technology is that the device has power. A phone does not need to have reception for our device to pick it up. Phones are constantly sending out signals to find themselves in the universe whether they have reception or not.
- Can you discuss your accuracy in terms of how you know the exact number of devices that are within 1/8-mile radius?
We are able to pick up signals ("pings") from all WiFi devices in the device's radius because all devices send out these signals (unless they are turned off). Through testing, we have concluded that our device in its current configuration is able to detect these signals up to 1/8 of a mile.
Based on how many of these "pings" our device picks up, we can then determine the number of devices in that 1/8 mile radius. The range and accuracy can vary widely depending upon device placement, type of antennas, and traffic flow.
The amount of traffic in one location does not negatively impact our collection.
- I’m personally concerned that saying that some device that was within 660 feet of my mobile billboard qualifies as an impression. That’s over two football fields away – our units are huge, but I’m not sure that they’re visible from that far away, especially in urban blocks. It makes me question the validity a bit.
Believe it or not, our devices can capture quite the distance. Even bigger than 660 feet. That's because we not only work in the OOH space, but we also work with smart city initiatives. However, we can bring this distance up or down depending on your needs. Every client is different so we work with you to figure out what your sweet spot is.
- Can you discuss the options you have to increase or decrease that radius around the unit?
The range of the device is mainly influenced by the type of antenna being used. We can adjust the antennas we're using in order to increase/decrease the range of the device. For example, the 3 dB antenna is the standard one that we typically use which captures at 1/8 of a mile. But we also have 9 dB antennas to increase range (1/4 of a mile) and a smaller 1 dB antennas to decrease range (1/16 of a mile).
- What antenna size do you use for attribution? If the mobile billboard with a device is 1/8 mile from the store that also has a device, could the same phone could be “seen” by both devices?
Our antennas are customized per location. If the above scenario is presented, we would use smaller reaches for conversions. It really just depends on the location.
- What about traffic flow? Does the antenna have a line-of-sight frequency that will pose a problem in congested high traffic areas?
There are no issues with high traffic. The Abraxas Scout can handle any volume of traffic or pedestrians at an urban setting, event, or any other congested area. In fact, one of our mobile billboard clients was at an event that had over 150,000 attendees. We were able to capture impression data at that volume and even when cell data was down (mentioned in more detail above).