If the conspiracy theorists are to be believed, there are satellites orbiting the Earth that can spy on individual consumers. They can even make out a car’s number plate. Why they should want to do this is not entirely clear. Nor is it especially believable – remote sensors could not tell the difference between a munitions factory and a hospital in recent missile strikes on Iraq, for example.
What is true is there are plenty of marketers who do want to get a satellite-style view of the terrain, and who are very interested in where consumers are driving to. Their interest is in deciding where to site new retail locations, or what the catchment areas are of existing sites. And with the growing regionalisation of retail and manufacturing, these companies want to make such decisions on a pan-European basis.
The most useful tool for business planning of this kind is the geographical information system (GIS). Until recently, users were generally limited to single-country systems, with the UK offering the most well-developed data sets. Recent developments have changed this, however, allowing users to operate a single, multi-territory system that allows common data sets and descriptors to be applied.
The Data Consultancy is one example of a supplier that has entered this market, with its Cartique product. The system allows for business mapping, routing and road network analysis across the whole of western and most of eastern Europe. With the inclusion of drive-time isochrones (a polygon applied to a map which covers all locations within a given time distance from the centre of the shape), it means users can plot sales territories, distribution networks and branches.
“Blue-chip multinationals are increasingly looking to plan regionally or globally. They don’t want to have different sorts of data and a different feel to the system in each country. They want the same software and approach and the same data in each,” says Brian Wade, managing director of The Data Consultancy.
His company had not previously been able to meet this demand because of the cost of sourcing all the relevant data. Since it was acquired by mapping software developer MapInfo it has become commercially feasible to do so.
In the short- to mid-term the company will still concentrate on supporting multinationals which have UK headquarters, but it will gradually roll out support and services in other European markets.
In data gathering terms, the difficult part – putting together a consistent set of cartographic features for every country at 1:300,000 scale – has now been done. “The work on our side has been about sourcing data of the right quality and creating a product that looks good and meets the needs of the market,” adds Wade.
While Cartique supports drive-time and mapping functions, it does not include any underlying demographic data. Although there are several European Union-funded initiatives to collate such data from member states and make it available in a common format, aggregated to similar levels in each country, it is still not possible to obtain everything from one buying point.
CACI launched its own pan-European GIS last year, called InSite Europe. But senior director John Rae notes that for a year beforehand, the company bought every data set it was offered. This allowed it to carry out its own cleaning and valid- © ation to assess just what was contained in the data.
One of the main difficulties to emerge is that most data is held at administrative boundary level, such as electoral districts in the UK, but this is not refined enough for marketers. Then again, in countries such as Germany, Sweden and the Netherlands, geodemographic data is held at postcode level, which is even less refined than the administrative version.
In Germany, for example, there are 14,000 Gemeinde or districts, but only 8,000 postcodes. In France, the whole of Marseille is included as a single, homogeneous postcode. CACI had to seek out population counts and the like which could be broken down to the level of resolution needed for marketing.
“There are issues. In Portugal and Spain, for example, there is a lot of data available, but the age bands used in each is different,” says Rae.
This has lead to some important design decisions on the InSite Europe system. These emerge most evidently where borders come into play, because the drivetimes in a catchment area which straddles a border will differ, often for simple reasons such as the speed limits which apply in each country. While the system will show such a cross-border catchment, the underlying report on the population counts it covers will remain split by country.
Experian has been developing geodemographic classifications in 17 countries around the world for the past 12 years to support its Global Mosaic system. These can be accessed within its Micromarketing GIS, with varying levels of cartographic detail in each. But all analyses can ultimately be reported back at a common top level for every country.
Paul Watts, director and general manager of Mosaic International, says an important differentiation of its system is that Experian offers local support in each country. “We’ve approached it on the basis that there is a market opportunity in those countries themselves. In developing a strong business locally, we can use that data and those systems to support multinational organisations who want to do analyses across countries,” he says.
That has important implications as to how the GIS is scaled in European applications. Many marketing ideas begin in one territory and are then rolled out across borders. If geographic and demographic targeting is important to the way the ideas are marketed, then clients do not want to have to change systems in each country. “We’re saying you can go to the Netherlands and find a full-blown Micromarketing service, just as in the UK. You can even find it in Finland,” says Watts.
Conversely, this multi-local approach means common definitions can be used to describe each territory within pan-European planning. “Although there are differences in Mosaic in each country, they work in the same manner. We have built them so they fit together within Global Mosaic,” he says.
What ultimately matters to marketers is likely to be less the technical issue of how a European GIS is constructed and more how it can be applied to practical problems. Where a product launch needs to be rolled out in different territories, having achieved success in one market, similar targeting propositions have to be applied.
That is how Circular Distributors (CD) has made use of Mosaic on behalf of a number of its clients who wanted to carry out sampling exercises in Europe. From its own research, CD had already discovered a substantial demand for samples in Italy and Germany, the very markets where L’OrÃ©al wanted to test market Elvive shampoo (Italy) and where Kimberly-Clarke looked to sample Kleenex Balsam (Germany).
In Italy, CD generated a ranking of target communes (deliverable electoral districts) based on age and the presence of three or more adults in households. This was overlaid with a store list to select the best targets, while major cities were broken down further into smaller unit areas.
The resulting 2.5 million item door drop led to 57 per cent of consumers using the sample, with 50 per cent having already purchased Elvive and 78 per cent definitely or probably buying it in the future.
According to Jim Dinsey, client services director at CD: “If you looked at Mosaic for the same basic approach, you would find it is a common platform. But it is wrong to think you can take a half-dozen countries and rank them on the same basis.” Understanding that it is the similar methodology that matters, rather than cross-border comparisons, means learning from one country can be taken into another.
In the UK, most population information is based on the postcode, of which there are 1.6 million, but in France there are just 3,000. Marketers may well have just one strategy which they want to apply in several countries. But the data on which they base their decisions has a bewildering range of variations. Fortunately, the key suppliers have done the difficult work of resolving these into single systems already.