Thanks, Jeff. Geocoding and reverse geocoding use cases do present privacy challenges. That said, the ability to aggregate and anonymize results with location context provides ample protection from disclosure of Personally Identifiable information (PII) in many of these cases. For a bricks and mortar retailer, understanding location at a zip code level of positional accuracy (i.e. which zip codes are most prominently represented among those who enter a particular store location) can be useful to marketing use cases, but to determine whether a market is underserved, and whether a new store location would substantially cannibalize existing store traffic requires a much more precise geocode of customer points of origin. Better yet, is some of that traffic based upon the daytime location of that consumer (i.e. work address) as opposed to a home address?
Results can be aggregated and anonymized while still requiring the most precise geocode possible in order to provide informed insights. Thus, bringing the developers, IT, and business stakeholders together is really critical in elevating the importance of positional accuracy to the use case. The same department may have 3-4 use cases for location intelligence, and even if only one requires high-confidence positional accuracy, that may warrant the additional investment (of course, this is ROI dependent). This same use case can apply to an e-Commerce retailer who is looking to complement their online success by introducing a bricks-and-mortar channel.
Pitney Bowes Software
PITNEY BOWES SOFTWARE, INC
Sent: 07-31-2019 08:03
From: Jeff Rueger
Subject: Geocoding Trends and Challenges
I would like to share a web link to a list of podcasts related to Location Intelligence, Spatial Analytics and Geocoding. Podcasts can be a great way to stay in touch with the latest news and trends in the industry. https://geoawesomeness.com/geospatial-podcasts/
The Mapscaping podcast series by Daniel ODonohue is one of my favorite series. In particular, I would to point out a podcast related to geocoding called "Geocoding- translating between machines and humans". The podcast describes the purpose of a geocoder is to translate between human and machine languages in regards to describing a location or an address. The podcast interviews the co-founder of OpenCage data, Ed Freyfogle. You can listen to the podcast by following the web link: https://mapscaping.com/blogs/the-mapscaping-podcast/geocoding-translating-between-machines-and-humans
The interview covers a variety of geocoding related trends and challenges such as:
- There is an explosion of spatial data collection and all this new data brings about major challenges and how we aggregate data sources and maintain them. Adopting modern dev\opts techniques is critical to keeping up with the challenge.
- Forward and reverse geocoding is not always about returning the best hyper-accurate result, like a roof top geocode. For forward and reverse geocoding there are also use cases that require a less precise answer. These types of use cases rely on the geocoder to aggregate address and location data. Aggregated geocoded results to a postcode or city can protect privacy by not reveling the exact location of individuals. These use cases may only need to know about country, state, city or postcode hierarchical information.
- Location based technology is a critical part of our life now and many software developers who write applications do not have a background in spatial data or the geocoding industry. They need simple to use location tools and spatial data that is reliable, scales and just works well.
These challenges are something my engineering team and I are spending time tackling. We want to meet these trends head-on. Here are some examples of what we are doing about them:
What are we doing about it?
There is an explosion of spatial data collection and all this new data brings about major challenges and how we aggregate data sources and maintain them
We continue to modernize and improve our geocoding dev\opt data processing pipelines. Today, our geocoding team is able to process nearly 1 billion global addressing features every quarter. Three years ago global quarterly data releases using a private cloud to process and aggregate vast amounts of spatial data was not an option.
Sometimes a geocoder will need to return a less precise answer. Aggregated geocoded results to a postcode or city can protect privacy by not reveling the exact location of individuals
Improved postal and geographic geocoding For the July 2019 release. We will continue to enhance postal and geographic geocoding to include more region, city and postcode hierarchical information. These changes allow us to choose a better match when postal and geographic geocoding are required
Many software developers who write applications do not have a background in spatial data or the geocoding industry. Software developers need simple to use location tools and spatial data that is reliable, scales and just works well
Development has adopted a design center thinking approach where we want to focus much more on ease of use by listening to our clients and better understanding their use cases
PITNEY BOWES SOFTWARE, INC