When deciding what markets to enter and where to select sites, retailers cannot ignore the importance and influence of demographic analysis — and that’s relatively easy in the US. Retailers can turn to the Census Survey, the American Community Survey, as well as offerings from well-established private companies to secure quality demographic data at relatively low cost. However, this is simply not always the case in other markets. In fact, as they start to look beyond the US for growth opportunities, retailers start to run up against significant challenges related both to the quality and accessibility of the demographic data – and this is particularly true in emerging markets.

So, the saying “if you build it, they will come” does not hold true when looking for new markets. With over 10 years of GIS expertise in demographics and statistics, I can honestly say that to achieve success in these markets, retailers require the same high quality demographics they have come to expect in the US. Without this key information, there is no practical way to compare and evaluate locations based on the surrounding factors that drive store sales.

So, if the information is mission-critical but not readily available, and retailers generally lack the resources to access this data – or assess its efficacy – how can they navigate these challenges?

For retailers to ensure that decisions are being made with the best information available, they need to understand the dynamics that impact the quality of demographics, and there are two factors that need to be properly addressed: the geographic unit and demographic variables. Both are equally important. The geographic unit helps you determine whether the analysis is accurate and reasonable, and the demographic variable informs your level of accuracy.

What is a Geographic Unit?

Let’s delve a bit deeper into the concept of the geographic unit.

Census survey data is usually collected and released at a certain geographic level. As an example, it could be a micro geography unit, an administrative geography unit or at a postal code level. All of these levels have applications to retail, so let’s explore them in a bit more depth.

The micro unit is ideal for retail site selection, for everything from convenience stores to fast food concepts, as well as big box retailers. While different countries refer to micro units by different names – in the US it is often called Block and Block Group, in Mexico it is Manzana, in the UK it is Output Area, and in Japan it is Small Area – they all represent the smallest census geography, and usually have a population of less than 1,000.

In most developed countries, such as those of Western Europe, micro level datasets are readily available. However, this is not the case in emerging markets, and is a true shortcoming, because these datasets are more trustworthy and consistent due to less government intervention in their development. And while some open markets, like Latin America, may also have this level of data, it is generally maintained by local companies with government ties. The information tends to be updated infrequently and the quality of data is questionable. Additionally, without local connections, it is very difficult to get your hands on this data. As an example, before engaging with Tango Analytics, some of our global clients could not get access to any demographics in these countries.

The next level of data – administrative level – is common in some emerging nations. For example, China reports demographics at the Jiedao/Township level, South Korea collects data at Dong level and UAE publishes data at the Emirate polygon level. The administrative level is delineated for the purpose of administration i.e. local and regional government, such as counties and municipalities — and you can expect to also see this level of data in some Eastern European, Middle East and Asian countries.

There are real challenges for US retailers in dealing with the administrative level of data. In China, for example, the lowest administrative geographic unit in urban areas may be as small as one square km, which is equivalent to US census tract. The issue is that retailers in US rarely use census tract level data in their demographic analysis. Instead, they use block groups which are only a portion of a census track. Also, one square km in urban area of China could represent a level of a population of over 20,000 – a much higher density then found in the US – and this compounds the issue. For retailers, the trade area demographics retrieved at this level is generally inaccurate, and can be misleading.

That said, geographic analysis is scale dependent. So, if you are considering a market entry analysis rather than site selection, administrative level data may provide sufficient depth. If, however, you are performing site selection, you will need to reduce your scope and hone in on a more specific geographic level.

There are also technological advancements that have made administrative data more valuable to retailers. With the help of high spatial resolution satellite imageries, night time satellite imagery, Light Detection and Ranging (LiDAR), OpenStreetMap (OSM), Point of Interest (POI), Landuse and Landcover (LULC) and other available resources, as well as the improvement of areal interpolation methodologies, the use of administrative level data has transitioned from simply being available for scientific research purposes to true commercial applications.

The final level – postal code – is applicable to very few countries. The size of this geographic unit lies somewhere between micro level and administrative level and does have applications for retailers, in particular related to direct mail marketing campaigns. One serious disadvantage is that postal code boundaries can change after a very short period of time – so relying on them can present some challenges.

The ability to leverage not only the latest technology – but understand local differences – and access and analyze available demographic data, is critical to a successful entrance into a new market. Tango Analytics is working increasing with retailers and restaurants chains that are looking to other markets to expand. As part of our experience with over 50 brands across 6 continents, we have come to learn a great deal about these geographic levels and how to apply state of the art technology to ensure the quality of this data – and its successful application.

In part two of this blog, I’ll explore the challenges that arise with the second of the factors that need to be addressed when assessing the quality of demographics – demographic variables. If you’d like to learn more, download Tango’s Predictive Analytics datasheet below.