Audience Analytics – A new revenue opportunity for telcos!
By : Flytxt R&D Team
Audience analytics combines the power of GPS and mobile data intelligence to gain insights about the trips and movement habits of groups of people. It opens up huge opportunities for many enterprises and public agencies, who are focusing on areas like smart city planning, infrastructure planning, safety planning and so on. Telcos sitting on enormous amount of rich information could monetize their data assets through efficient and smart use of audience analytics. Out-of-Home (OOH) advertising is yet another important application area whose success depends upon the advertising impact and reach of messages, a critical differentiator in the world of digital marketing. For example, visibility of a billboard (or hoarding) is an important factor to consider. It enables agencies to come up with appropriate pricing strategies for the billboards. Apart from these, through audience analytics, customized billboards can be setup on desired locations based on the profiles of users (which could be extracted from the mobile operator data), such as sports lover, high data users, etc.
However, determining the visibility of a billboard requires estimating the travel routes of users. Flytxt has developed a novel ‘location stitching model’ which estimates the route that the user would have travelled with a fair degree of accuracy, given the latching information of the user’s mobile device onto different mobile network towers. This latching information comprises of a sequence of latitude-longitude pairs of the cell ids, which the user’s mobile device was latched onto, at different instances of time in a day. The travel route is estimated by taking into consideration various factors, such as, the real time traffic information, distance of the travel route, giving preference to the shortest route and whether the route involves a highway or not, and so on. Once the route is estimated, the visibility of a billboard is obtained using a proximity computation formula, which measures the closeness of the billboard from the estimated route. The model also takes into account the direction of movement of the user as well as the direction of the billboard, to come up with the set of users who might have seen a particular billboard. The accuracy of the model can increase significantly if augmented with GPS or Wi-Fi hotspot data, which traditionally captures more granular and accurate location information.