ING’s Infrastructure Index: Methodology
Data sets used to compile the research
ING’s research team used two data sets:
1. Our existing data set, which counted all conversations about each of 50 UK cities between January and December 2024, and formed the basis of ING’s annual UK’s Most Talked About Cities report, published in May 2025.
The 50 cities in that index include the UK's Core Cities; cities within the Key Cities (i) network; cities in the Scottish Cities Alliance (ii); and other urban areas which rank in the top 30 in the UK for population size.
2. Contemporary data (1 August 2024 – 31 July 2025) collated by ING on the total number of conversations about infrastructure about each of the cities included in the UK’s Most Talked About Cities report.
The Infrastructure Index was created using these two data sets to calculate the percentage of conversations about each city that are about infrastructure and ranking them from highest to lowest.
How we measured volume of online conversations
The data includes mentions on news and social media platforms on which people talk about cities: online news, Facebook, Instagram, LinkedIn, X, TikTok, Reddit, blogs, forums, YouTube, comments, Pinterest and Twitch.
To find the number of infrastructure-related conversations about each city, we used a series of search term ‘proxies’.
Search terms
For each theme, we based the proxy search terms on our combined 10+ years’ experience undertaking digital visibility benchmarking work for cities and places. This experience has enabled us to understand the nouns and adjectives people use most frequently when they talk about cities in relation to each theme. The proxies were tight enough to include only results that directly related to that theme, but broad enough to not exclude any of the most important elements.
How we measured how much different infrastructure topics are talked about
We restricted this part of the analysis to the top 20 cities for total mentions of infrastructure. These 20 cities collectively account for almost 90% of all conversations about infrastructure across the full set of cities.
How we compared how much infrastructure is talked about vs other themes
We took the same proxy-based approach to measure conversations about other themes so we could compare their volume with the volume of conversations about infrastructure.
For this comparison, we looked only at five of the UK’s biggest cities: London, Manchester, Liverpool, Birmingham and Bristol. Given that these cities represent nearly 50% of the total population – and more than 70% of all conversations about infrastructure – of all cities in the index, this group provides a useful representative snapshot of wider trends. The total number of conversations for each theme is the number of conversations summed across these five cities.
Exclusions from the data
Five cities in the UK’s Most Talked About Cities Index – Bath, Derby, Lancaster, Lincoln and York – had to be excluded from the Infrastructure Index because the proxy terms used to measure the number of conversations about infrastructure (see below) created spurious results for them. Usually this was because other cities outside the UK have the same name (e.g. Lancaster), they feature in the names of other infrastructure projects (e.g. Lincoln Tunnel), or they are common surnames of prominent individuals (e.g. Lincoln, York).
As is standard for ING research examining digital visibility, we have excluded mentions of sport (local football teams, major sporting events, other sport-related terms such as score, league, match, etc). A significant number of conversations about football – especially from outside the UK – still use city names as proxies for football teams (e.g. “Liverpool” for Liverpool FC. Excluding these enables fairer comparison across cities which do not have outsized or internationally recognised teams, and also helps restrict results to conversations directly about cities.
This research also excluded any local or national phenomena that happen in or to cities that are linked to infrastructure but are otherwise out of most people's control (e.g. train cancellations/delays as a result of weather, violent incidents on trains and buses, arson attacks on local mobile masts, etc). Given that these types of incident tend to be more frequently picked up in hyperlocal news and conversation, and that this research is designed to illustrate what the world is saying about infrastructure, this helps make sure we are not unfairly biasing cities with more of this type of hyperlocal coverage.
Notes
(i) Key Cities - a national cross-party network of 23 cities and urban areas that represents 12% of the UK’s urban population outside of London. The network works with other cities, towns, and organisations within and beyond local government to influence Government policy, maximise the benefits of devolution, and work towards a productive, balanced economy for all parts of the UK.
(ii) Scottish Cities Alliance - a network of Scotland's eight cities working in partnership with the Scottish Government to promote the country's economic potential.