Real time geodemographics: New services and business opportunities (and risks) from analysing people in time and space
- 532 Downloads
Traditionally, geodemographics has been the description of people according to where they live, derived from the study of spatial information. New technologies such as GPS tracking and virtual worlds, however, provide an opportunity to describe people in much greater detail in terms of space and time. From the moment we wake up, our digital footprints provide a rich source of data for ‘real time geodemographics’, which can support some exciting new service and business opportunities — from pay-as-you-drive motor insurance to location-based social networking. This paper surveys the enabling technologies and illustrates what can be achieved with a series of case studies. We also examine the downside risks, especially the data protection and privacy issues that will impact public acceptance. Finally, we make a few predictions for how real time geodemographics will develop over the next few years.
Keywordsreal time geodemographics GPS tracking virtual worlds surveillance
Over the past quarter century, geodemographics has proved to be a valuable tool for customer analysis and market planning — in everything from predictive behaviour modelling to branch and store location assessment. 1
Traditionally, the knowledge of where people were spatially located has been limited to a home address, possibly also a work address, and perhaps to some retail sites used, based on postcode analysis. This situation is changing rapidly as new technologies make it possible to track people spatially in real time or near real time to a high degree of precision.
These technologies also carry risks, especially in relation to the protection of personal data and privacy. We will examine what needs to be done in this area in order to win public acceptance. Finally, we will make a few predictions about how real time geodemographics will evolve over the next few years and the players most likely to seize the initiative in this exciting new field.
There are three main technologies that are relevant.
Perhaps the best-known technology is the global positioning system (GPS), which is currently the only operational example of a global navigation satellite system (GNSS). 4 Operated by the US Air Force, GPS enables users to find their location (three dimensions of space and one of time) by referencing the position of several (normally four) satellites, which are within view from any point on, or close to, the earth's surface. A device that combines GPS referencing with a communication facility to transmit locational data to a central computer enables the device (and the person or object to which it is attached) to be tracked.
GPS is usually accurate to within a few metres, but reliability depends on a clear view of enough satellites (there are 24 satellites in near-earth orbit in the GPS constellation) and this can be a problem inside or near buildings, in tunnels, in hilly terrain or in woodland. Over the next few years, several new GNSS systems will come onstream, including Galileo (EU and ESA), GLONASS (Russia), COMPASS (China) and IRNSS (India) — and these may improve coverage and reliability, as well as security of service.
Vehicle satellite navigation (SatNav) systems make use of GPS. They overcome the reliability problems of GPS by using the so-called ‘dead reckoning’ — so that if, for example, a car goes into a tunnel, the SatNav device can calculate, using data from accelerometers and from the car's drive system, where the car has got to from the point it lost the GPS signals.5, 6 Some systems will also take into account the fact that the car can be assumed to be somewhere on the road network, in order to improve the positioning calculation.
A fourth important tracking technology is the radio frequency identification (RFID) tag. 9 An RFID tag is a low-cost device, essentially a computer chip with a built-in radio receiver and transmitter, which can be attached to or incorporated into a product, animal or person for the purpose of identification using radio waves. The tags can be read using remote sensors to identify where the tagged object is and at what time.
All of the above tracking technologies have limitations and we can expect devices to be developed in the future that combine the technologies in a way that will call upon the most accurate at any particular point in space. This is beginning to happen with the embedding of GPS devices in some high-end mobile phones. Cost is an issue currently, but we confidently expect technical advances that will dramatically lower costs and enable the production of cheap and reliable tracking devices that will function anywhere and at any time.
A more speculative surveillance technology is the so-called ‘smart dust’. 13 This consists of hypothetical, very low-cost nano-scale devices that would be capable of sensing objects in their vicinity and communicating the sensory data wirelessly among themselves in a way that can be picked up by a few conventional fixed receivers. The idea is that smart dust would be scattered over a wide area and provide a low-cost alternative to conventional surveillance technologies such as CCTV.
It is one thing to track people or objects through time and space, but how is this information to be marshalled and applied for business value? One way is through the use of a ‘virtual world’ (or, rather, ‘parallel world’). This is a web-based representation of the real world populated by people, places and things whose location in time and space has been identified through tracking or surveillance. The pioneer in this area was HP Labs through its Cooltown Project, 14 but this and other, more recent initiatives have morphed into the so-called Internet of Things. 15
Of course, once we dip into virtual worlds, there is no need to restrict ourselves to representations of the real world. Applications such as Second Life® 16 and ActiveWorlds 17 and online gaming enable representations of people as ‘avatars’ inhabiting an entirely artificial world, but where real business activity and social interaction, even market research, 18 can take place. While these artificial virtual worlds are fascinating, and offer many business opportunities, we shall restrict ourselves in this paper to representations of the real world.
Case Study 1 — Pay-as-you-drive motor insurance
The basic idea behind pay-as-you-drive motor insurance is that the risk of accident or damage depends on the roads travelled, speed and the time of day, as well as on the traditional risk-related characteristics of the driver (age, gender, previous history, etc). By factoring this additional risk data into the insurance pricing, and providing add-on services enabled by the technology, it is possible to offer a more competitive motor insurance product to some drivers.
Importantly, additional services are offered that can take advantage of the tracking technology. These include an emergency safety button (so that emergency services can be informed of the precise location and time of an accident), stolen vehicle recovery and an optional satellite navigation service.
Pay-as-you-drive motor insurance is one example of a broader class of applications for in-vehicle tracking and related technologies, generically known as ‘Telematics’. The recent report by the Association of British Insurers 22 sets the scene for how telematics will shape the future of motoring, both domestic and business. Other applications include road pricing, journey planning, the tracing of stolen vehicles and emergency service call-out.
Case Study 2 — Traffic forecasting and congestion management
Most current services provide information, often in real time, about traffic congestion by integrating data sampled from static roadside sensors, from sampling the current location and speed of a large number of commercial vehicles, or from a combination of both. Route-finding tools such as Smartnav 26 will track the user vehicle's location and integrate this with the congestion data in order to make route recommendations that steer the user around problem areas.
Inevitably, many current services limit congestion monitoring to major routes, leaving a large gap in information about minor roads and byways. An application that uses mathematical modelling to predict the congestion in all categories of road is ClearFlow from Microsoft Research, which has been trialled in Seattle. 27 This takes the form of a smart electronic map, which, for any particular road, models the status of that road by taking into account information on congestion on nearby roads as well as the day of the week, time and weather conditions. Crucially, ClearFlow can use road configurations it already knows about to predict how traffic on unfamiliar configurations will behave.
Case Study 3 — Virtual worlds
One of the first initiatives was the Cooltown project, initiated by HP Labs 14 in 2000. Cooltown seeks to develop an infrastructure to support web presence for people, places and things. By providing a bridge between the real world and a web-based virtual world, it will enable services to become more personalised, spontaneous and responsive to the wide variety of contexts in which people live their lives. Early applications include the ‘Cooltown Museum’, where a visitor to a real-world museum is tracked in a virtual representation of the museum and is served up with supplementary information and commentary according to their location. At the same time, the user can post electronic comments and messages into the virtual world, which are then accessible to other visitors as they move around the museum, or to others wanting remote access.
A recent example of a web-based infrastructure initiative is the so-called Internet of Things, which spawned a major international conference in Zurich in the spring of 2008. 15 Quoting from the conference flyer, ‘The term “Internet of Things” has come to describe a number of technologies and research disciplines that enable the Internet to reach out into the real world of physical objects. Technologies like RFID, short-range wireless communications, real time localisation and sensor networks are now becoming increasingly common, bringing the Internet of Things into commercial use. They foreshadow an exciting future that closely interlinks the physical world and cyberspace — a development that is not only relevant to researchers, but to corporations and individuals alike’. Major sponsors of the conference included Siemens, Google and IBM.
In parallel with the above developments, the web itself is also evolving into a form, the ‘Semantic Web’, in which the semantics of information and services on the web is defined, making it possible for the web to understand and satisfy the requests of people and machines who use the web content. 31 The DBMedia project, led by the Universities of Berlin and Leipzig, is seeking to develop a semantic version of Wikipedia. Within the project is Mobile DBpedia, a tracking application (requiring a GPS-enabled mobile phone) that takes a user's GPS position and displays Wikipedia articles on places in the vicinity, as well as showing them on a map. 32
Case Study 4 — Retail planning
The Wharton Management School at the University of Pennsylvania, in association with Sorensen Associates, has been using RFID tags attached to shopping trolleys to follow grocery shoppers around stores. 33 Sorensen's PathTracker® technology employs RFID tags in conjunction with video surveillance to map out the patterns followed by shoppers. Wharton then analysed the resultant data using a range of methods including cluster analysis.
More recently, a dozen FMCG companies, including Proctor and Gamble, Coca Cola and Kellogg, have teamed up with 18 major retailers, including Walmart, Walgreens and Target, on ‘Project Prism’, an initiative led by AC Nielsen. 34 The objective is to understand how shoppers move around in a superstore in terms of their paths around the aisles, speed and time of day — and use this understanding to improve displays and, ultimately, sell more goods.
Prism (Pioneering Research for an In-Store Metric) uses infrared surveillance technology to track shoppers’ movements and correlate this with sales data. The project is described as ‘the first truly scientific measurement of the effectiveness of in-store sales tools such as shelf location and promotional displays’.
Case Study 5 — Location-based services
— Location-based social networking where user interactions can be triggered by physical proximity: for example, friends or business colleagues in a social networking group can be alerted if another friend enters the same building or moves to within a short distance. The recent acquisition of the Helsinki-based company Jaiku.com by Google underlines the importance that the industry places on this area. 38
— Proximity-based marketing — where promotional offers are sent by SMS or voicemail to prospects in the vicinity of the service provider:39, 40 the movie by Steven Spielberg, Minority Report, showed examples of how individual-specific messages and advertising might, in the future, be displayed in a shopping mall as surveillance devices recognise that individual. 41
Privacy and data protection
Subsequent qualitative research commissioned by the Information Commissioner's Office revealed the extent of public awareness and the concerns felt. 43 The research found that, as far as commercial organisations were concerned, there was general acceptance by the public that data have to be exchanged in order to receive innovative and cost-effective products and services. It also revealed anxieties about the potential for identity fraud and commercial data abuse. Another key concern was the need for greater transparency about commercial data sharing and the rules and penalties on holding data.
To underline concerns in this area, a recent report by the UK's House of Commons Home Affairs Committee, ‘A Surveillance Society?’, 44 examines the implications of the enabling technologies in great depth. It makes a number of recommendations for the tightening of data protection and additional overseeing of personal privacy protection, especially in the public sector.
One thing is clear — the technologies that enable real time geodemographics are evolving rapidly. Falling costs and increasing miniaturisation and sophistication are driving progress towards cheap tracking and surveillance devices that can be embedded in many everyday objects including mobile phones, motor vehicles, laptops, shopping trolleys, luggage, clothes, packaging and many others.
Underpinning real time geodemographics will be a vast new ocean of data. Making sense of these data will require the novel use of existing analytical tools as well as the development of new statistical and mathematical modelling methods. We expect that geodemographic classification systems will be developed that use the spatial footprint data as well as personal, neighbourhood and transactional data. We also anticipate the development of other forms of multidimensional analysis that will more effectively exploit the dynamic nature of the data involved.
There will be problems along the way. Disasters with privacy and data protection are bound to occur, hopefully isolated and redeemable — as they have been with almost every other technology in its early days, from horse-drawn vehicles to aviation, or folk remedies to modern pharmaceuticals. The commercial and public sector organisations that exploit real time geodemographics must demonstrate to consumers that the benefits in terms of new and valuable services outweigh the potential risks.
This paper is based on a presentation given by the author at the seminar ‘The Future of Geodemographics — 21st Century Data Sets and Dynamic Segmentation: New Methods of Classifying Areas and Individuals’ organised by the Census and Geodemographics Group of the Market Research Society and held on Thursday 6 March 2008 at The Society of Chemical Industry, London SW1. 3 Additional information on this and related topics may be found at the Geodemographics Knowledge Base. 48
References and Notes
- Harris, R., Sleight, P. and Webber, R. (2005) Geodemographics, GIS and Neighbourhood Targeting, John Wiley & Sons, Ltd, New York.Google Scholar
- The Census and Geodemographics Group of the Market Research Society, Conference. (2008) ‘The future of geodemographics — 21st Century datasets and dynamic segmentation’, http://www.mrs.org.uk/networking/cgg/cggmar08_prog.htm.
- Hofmann-Wellenhof, B., Lichtenegger, H. and Wasle, E. (2007) GNSS — Global Navigation Satellite Systems: GPS, GLONAS, Galileo and More, Springer-Verlag, Austria.Google Scholar
- Wikipedia. (2008) ‘Automotive navigation system’, http://en.wikipedia.org/wiki/Automotive_navigation_system.
- SearchMobileComputing. (2007) ‘Cell of origin’, http://searchmobilecomputing.techtarget.com/sDefinition/0,,sid40_gci509920,00.html.
- GPS World. (2006) ‘Tracking first responders: Integrated Nav with software-defined radio’, http://uc.gpsworld.com/gpsuc/article/articleDetail.jsp?id=385393.
- Wikipedia. (2008) http://en.wikipedia.org/wiki/RFID.
- Connor, S. (2005) ‘Surveillance UK: Why this revolution is only the start’, The Independent, http://www.independent.co.uk/news/science/surveillance-uk-why-this-revolution-is-only-the-start-520396.html.
- Wikipedia. (2008) ‘Automatic number plate recognition’, http://en.wikipedia.org/wiki/Automatic_number_plate_recognition.
- Privacy International. (2007) ‘Satellite surveillance’, http://www.privacyinternational.org/article.shtml?cmd%5B347%5D=x-347-559095.
- Wikipedia. (2008) ‘Smart dust’, http://en.wikipedia.org/wiki/Smartdust.
- Kindberg, T., Barton, J., Morgan, J., Becker, G., Caswell, D., Debaty, P., Gopal, G., Frid, M., Krishnan, V., Morris, H., Schettino, J., Serra, B. and Spasojevic, M. (2001) ‘Cooltown — People, places, things: web presence for the real world’, HP Labs, Technical Reports, http://www.hpl.hp.com/techreports/2001/HPL-2001-279.html.
- International Conference for Industry and Academia, Zurich. (2008) ‘The internet of things 2008’, Organised by ETH Zurich, University of St Gallen, Massachusetts Institute of Technology, http://www.the-internet-of-things.org.
- Second Life®. http://secondlife.com.
- Activeworlds. http://www.activeworlds.com.
- Tarran, B. (2007) ‘MR firms publish first reports from Second Life’, Research Magazine, 26 January 2007, http://www.research-live.com/news_story.aspx?pageid=30&r=y&newsid=2738.
- Norwich Union. (2007) ‘Pay as you drive insurance: Data protection — Your information’, http://www.norwichunion.com/library/pdfs/payd-data-protection.pdf.
- Hughes, K. (2008) ‘Surveillance fears force Norwich to scrap “pay as you drive” car policies’, The Independent, http://www.independent.co.uk/news/business/news/surveillance-fears-force-norwich-to-scrap-pay-as-you-drive-car-policies-848562.html.
- Murray, K. (2006) ‘Pay-as-you-drive motor insurance’, Money Extra, 20 September 2006, http://www.moneyextra.com/features/023927.features.html.
- Rapp Trans (UK) Ltd. (2006) ‘Motor insurance and new technology: Shaping the future’, for the Association of British Insurers, http://www.abi.org.uk/BookShop/ResearchReports/RAPP%20Report.pdf.
- Trafficmaster. http://www.trafficmaster.co.uk.
- ITIS Holdings plc. http://www.itisholdings.com.
- Press Release. (2006) ‘Real time traffic routing from the comfort of your car’, National Science Foundation, http://www.nsf.gov/news/news_summ.jsp?cntn_id=107972.
- Smartnav. http://www.smartnav.com.
- Biever, C. (2006) ‘Electronic map keeps drivers away from jams’, New Scientist, October 2006, http://technology.newscientist.com/channel/tech/dn10345-electronic-map-keeps-drivers-away-from-jams.html.
- Feiner, S. K. (2002) ‘Augmented reality: A new way of seeing’, Scientific American, April 2002, http://www.sciam.com/article.cfm?id=0006378C-CDE1-1CC6-B4A8809EC588EEDF.
- Wikipedia. (2008) http://en.wikipedia.org/wiki/Augmented_reality.
- MIT SENSEable City Lab. (2006) Real Time Rome, Massachusetts Institute of Technology, http://senseable.mit.edu/realtimerome.
- W3C. (2008) ‘The world wide web consortium’, http://www.w3.org.
- Becker, C. and Bizer, C. (2008) ‘DBpedia mobile: A location-enabled linked data browser’, LDOW 2008, Beijing, China, http://events.linkeddata.org/ldow2008/papers/13-becker-bizer-dbpedia-mobile.pdf.
- Knowledge@Wharton. (2005) ‘Tag team: Tracking the patterns of supermarket shoppers’, Wharton School Publishing, Wharton School of the University of Pennsylvania, http://knowledge.wharton.upenn.edu/articlepdf/1208.pdf?CFID=69501531&CFTOKEN=16866761&jsessionid=9a30ee1a3ddd5216324f.
- Rigby, R. (2008) ‘Superstore search for new styles in the aisles’, Financial Times, 28 May 2008, http://www.ft.com/cms/s/0/d6490a94-2c4f-11dd-9861-000077b07658.html?nclick_check=1.
- Wikipedia. (2008) http://en.wikipedia.org/wiki/Location-based_service.
- Bennett, V. and Capella, A. (2002) ‘Location based services: Wherever you are, wherever you go, get the information you want to know’, IBM DeveloperWorks Sample IT Projects, http://www.ibm.com/developerworks/ibm/library/i-lbs.
- Infinian Corporation. ‘Mobile commerce solutions’, http://www.infinian.com/solutions.htm.
- Engeström, J. and Koponen, P. (2007) ‘Google acquires Jaiku’, Google Q&A, http://www.jaiku.com/help/google.
- Munson, J. P. and Gupta, V. K. (2002) ‘Location based notification as a general-purpose service’, Proceedings of the 2nd international workshop on Mobile commerce, Atlanta, Georgia, USA, p.40-44, Pub. Association for Computing Machinery, New York, http://portal.acm.org/citation.cfm?id=570705.570713&coll=ACM&dl=ACM&type=series&idx=SERIES11257&part=series&WantType=Proceedings&title=WMC.
- Wikipedia. (2008) http://en.wikipedia.org/wiki/Location-based_media.
- Mathieson, R. (2002) ‘The future according to Spielberg: minority report and the world of ubiquitous computing’, mpulse (a Cooltown magazine), http://www.rickmathieson.com/articles/0802-minorityreport.html.
- Surveillance Studies Network. (2006) ‘A report on the surveillance society’, For the Information Commissioner, http://www.ico.gov.uk/upload/documents/library/data_protection/practical_application/surveillance_society_full_report_2006.pdf.
- Murphy, O. (2007) Prepared for COI, on behalf of the Information Commissioner's Office, http://www.ico.gov.uk/upload/documents/library/corporate/research_and_reports/surveillance_report_v6_final.pdf.
- House of Commons, Home Affairs Committee, Fifth Report of Session 2007–2008. (2008) A Surveillance Society? The Stationery Office Limited, London, http://www.publications.parliament.uk/pa/cm200708/cmselect/cmhaff/58/58i.pdf.
- Greenfield, S. (2003) Tomorrow's People: How 21st Century Technology is Changing the Way We Think and Feel, Penguin Books, Harmondsworth.Google Scholar
- Counts, S., Ter Hofte, H. and Smith, I. (2006) ‘Mobile social software: Realizing potential, managing risks’, Conference on Human Factors in Computing Systems, Pub. Association for Computing Machinery, New York, http://portal.acm.org/citation.cfm?id=1125451.1125767.
- Nelson, S. D. (2007) ‘Gartner CRM Summit 2007 explores a critical phase in CRM’, Gartner Inc., Links via, http://searchcrm.techtarget.com/news/article/0,289142,sid11_gci1272764,00.html.
- The Census and Geodemographics Group of the Market Research Society. (2008) ‘The geodemographics knowledge base’, http://geodemographics.org.uk.