Electoral Geomarketing: Definition, Methodology and Strategic Applications

Introduction

Election campaigns across Europe and beyond are undergoing a quiet revolution. While national opinion polls continue to capture media attention, a little-known discipline is profoundly transforming how candidates organise their ground strategy: electoral geomarketing. Born at the intersection of geography, statistics and political science, this approach analyses voting behaviour at a very fine spatial scale — polling station, precinct, ward, neighbourhood, street and building — to optimise the allocation of campaign resources on the ground.

Just as commercial geomarketing helps retail chains locate their outlets based on consumer flows, electoral geomarketing addresses a fundamentally different question: where, precisely, should a candidate concentrate door-to-door canvassing, leafleting and street-level campaigning to maximise their result on election day? This seemingly simple question demands considerable technical expertise and an intimate understanding of local social geography.

The Origins of Electoral Geomarketing

From Electoral Geography to Operational Voter Targeting

Electoral geography is an old discipline, rooted in early twentieth-century France. In 1913, André Siegfried published his Tableau politique de la France de l’Ouest, widely regarded as the founding text of electoral sociology in France and worldwide. In this landmark 511-page work illustrated with 102 maps, Siegfried demonstrated that voting behaviour is embedded in enduring territorial structures, shaped by deep geographical, social and anthropological determinants.

Siegfried famously linked electoral preferences to soil types: granite areas tended to favour conservative parties, while limestone areas leaned progressive — giving rise to the celebrated formula « granite votes right, limestone votes left. » Though sometimes caricatured, this analysis rested on a subtle interplay between geology, land ownership patterns, the influence of the Catholic Church and the weight of traditional social structures.

This tradition was renewed in the 1980s by Emmanuel Todd and Hervé Le Bras. In L’Invention de la France (1981) and later Le Mystère français (2013), these researchers combined anthropological, demographic and electoral data to explain the political fractures of French territory. Their cartographic analysis revealed the enduring influence of pre-industrial anthropological and religious structures on contemporary voting patterns.

The work of geographer Michel Bussi, notably in Éléments de géographie électorale (1998), extended this tradition by analysing electoral configurations at the intra-urban scale with remarkable methodological rigour. In the English-speaking world, Ron Johnston and Charles Pattie’s Putting Voters in their Place (2006) similarly demonstrated the powerful effect of local context on voting behaviour, showing that where people live shapes how they vote through neighbourhood effects and social interaction.

In a complementary perspective, geographer Christophe Guilluy renewed the analysis of French territorial fractures by highlighting the concept of « peripheral France » — areas disconnected from the economic dynamism of major cities. This framework has profoundly influenced French political debate and resonates directly with electoral geomarketing: precisely identifying territories where working-class populations are cut off from metropolitan dynamics helps understand the new geographies of protest voting and adapt ground-level campaign strategies accordingly. Similar dynamics have been documented across Europe, from the « left-behind » areas studied by Andrés Rodríguez-Pose (2018) to the urban-rural divides shaping elections in the Netherlands, Germany, Italy and Scandinavia.

Electoral geomarketing differs from academic electoral geography in its purpose: the aim is no longer merely to understand why a territory votes in a particular way, but to draw immediate operational consequences. The goal is to move from analysis to decision, from territorial diagnosis to a campaign action plan. This is where the « marketing » prefix takes on its full meaning: applying the optimisation and targeting logic of the commercial world to the electoral field.

The American Legacy and the European Adaptation

In the United States, electoral micro-targeting has become a campaign pillar since the early 2000s. Barack Obama’s 2008 and 2012 campaigns popularised the massive use of individual voter data to identify persuadable and mobilisable voters, block by block, door by door. Journalist Sasha Issenberg, in The Victory Lab (2012), documented this transformation: American campaigns now assign each voter predictive scores on their probable behaviours, turning the electoral field into a large-scale behavioural analysis laboratory. The American political consulting industry is vast, with some estimates suggesting it accounts for up to 7% of US GDP.

In Europe, the context is radically different. The electoral market is vastly smaller, campaign budgets are capped by law, elections are relatively infrequent, and individual voter data is not freely accessible under GDPR and equivalent regulations. European electoral geomarketing has therefore had to invent its own methods, based not on individual voter files but on the spatial analysis of aggregated data: election results by polling station (precinct, ward, bureau de vote), national census data, land registry records and housing typology. It is within this constraint that the richness of the approach paradoxically lies.

Methodology: From Municipality to Building

Data Sources

The first layer consists of election results published by national authorities: the Ministry of the Interior in France, the SPF Interior in Belgium, the ISIE in Tunisia, Elections Ontario or Elections Quebec in Canada, the Kiesraad in the Netherlands, the Bundeswahlleiter in Germany, and equivalent bodies across democratic nations. These results are available at the level of the polling station (France), polling division or section de vote (Quebec), voting centre (Tunisia), or municipality — gemeente in the Netherlands and Belgium, Gemeinde in Germany and Switzerland, comune in Italy, municipio in Spain. A polling station typically encompasses between 500 and 1,500 registered voters depending on the country. This level of analysis already offers far greater granularity than the municipality or constituency, but remains insufficient for precise voter targeting in small and medium-sized towns.

The second layer comes from census and socio-demographic data produced by national statistical institutes: IRIS areas in France, dissemination areas in Canada, statistische sectoren in Belgium, Statistisches Gebiet in Switzerland, Output Areas in the United Kingdom, and imada in Tunisia. These datasets describe the socio-demographic composition of territories ranging from a few hundred residents in Canada and Belgium to roughly a thousand in France and the UK. This heterogeneity demands methodological adaptation from country to country.

The third layer, often overlooked but decisive, is the land registry and urban morphology. The dominant housing type on each street reveals powerful indicators of electoral behaviour: in France and Belgium, terraced houses (rijwoningen), detached suburban estates (lotissements, verkavelingen), four-façade houses (villas quatre façades, vrijstaande woningen), and social housing blocks (HLM, sociale woningen) profoundly structure the geographies of the vote. In Quebec, the contrast between condominiums, suburban bungalows and urban triplexes plays an analogous role. In the United Kingdom, the distinction between council estates, semi-detached suburbs and inner-city terraces produces similar patterns. In the Netherlands, the contrast between vinex-wijken, grachtenpanden and flatgebouwen carries comparable electoral significance. In all cases, housing type constitutes an explanatory variable of voting behaviour that is often more robust than aggregated socio-professional categories.

Spatial Analysis and Sub-Polling Station Disaggregation

The central methodological challenge of electoral geomarketing is to go below the polling station level. Precinct-level or ward-level analysis may suffice in very large cities (Paris, Lyon, Brussels, London, Montreal, Geneva, Amsterdam), but it is too coarse in towns and constituencies with fewer than 100,000 inhabitants. Fine-grained analysis makes it possible to extrapolate voting behaviour by zooming in by a factor of 15 to 30 within each polling station, reaching the granularity of the street, the housing estate (lotissement, woonwijk, Siedlung), the city block (îlot, huizenblok, Häuserblock) and the individual building.

This disaggregation relies on spatial analysis techniques implemented in geographic information systems (GIS) such as QGIS, combined with statistical processing in advanced spreadsheets. The principle consists of modelling, for each address in a municipality, an estimate of the probable electoral behaviour of its residents by cross-referencing four variables: precise geographic location, housing type (identified through the land registry and field observation), the socio-demographic characteristics of the immediate neighbourhood (drawn from national census data), and past election results from the relevant polling station.

This methodology, tested and refined in France and Belgium over fifteen years, has been successfully adapted in Tunisia during presidential and legislative campaigns, where sub-voting-centre disaggregation made it possible to identify pockets of mobilisable voters in constituencies where winning a parliamentary seat under a largest-remainder system came down to a handful of votes. It is currently being adapted for the Quebec, Swiss and Luxembourg markets, where the specificities of electoral systems and linguistic dynamics require methodological adjustments while preserving the core approach: descending to the finest possible spatial scale to identify precisely where to concentrate campaign effort.

Operational Concepts: A Bespoke, Artisanal Approach

The electoral geomarketing I practise differs radically from standardised approaches imported from the United States, which apply a one-size-fits-all formula regardless of local context. My artisanal method is built on complete adaptation to the political, electoral and social realities of each territory. Every engagement begins with an in-depth diagnostic with the candidate to identify their specific challenge and define the most relevant territorial strategy.

Adapting Strategy to the Candidate’s Context

Electoral geomarketing does not offer a single solution. Depending on the candidate’s political situation, several territorial strategies may be deployed.

Consolidating threatened strongholds. When a candidate senses that their traditional bastions are eroding, or that the election will not be fought on their historical themes but on an issue that does not favour them, the analysis identifies sectors where mobilising loyal but disengaged voters should be the priority.

Reconquering decisive areas of weakness. In some constituencies, a candidate may discover that they lost their seat not in their home town but in distant territories where they never campaigned. Electoral geomarketing pinpoints precisely those neglected municipalities, suburbs or neighbourhoods (quartiers, wijken, Stadtteile, barrios, rioni) that tipped the result, allowing a focused recovery effort.

Targeting undecided and mobilisable voters. Unlike symbolic campaigning — which consists of showing up at markets, organising rallies and passively waiting for voters to come to the candidate — electoral geomarketing encourages proactive outreach. In a context of widespread distrust towards the political class, only already politicised citizens spontaneously seek out a candidate. Yet it is precisely the non-politicised, undecided, volatile or habitual abstainers who swing elections. Electoral geomarketing enables campaigns to go and find them where they live, building by building, street by street, through door-to-door canvassing (porte-à-porte, huis-aan-huis, Haustürwahlkampf), doorstep conversations and direct voter contact.

Swing Polls & Streets

Swing polls designate the polling stations (precincts, wards) that are electorally the most divided, where the balance of power between the main political currents is tightest. In the current French context of a three-way electoral split, a typical swing poll shows an approximate distribution of one-third left-wing alliance, one-third centrist bloc and one-third Rassemblement National. These territories are where every vote weighs heaviest, where neighbourhood effects can shift a few dozen voters and influence the overall result. Similar tripartite or multipartite splits are observable in the Netherlands (PVV, left-green, VVD-centre), in Belgium (N-VA, Vlaams Belang, left-green in Flanders), in Germany (AfD, CDU, SPD-Greens) and in many other European democracies where fragmented electorates produce razor-thin margins.

Swing streets are, within these swing polls, the specific streets and buildings where a candidate must campaign intensively so that word-of-mouth in the neighbourhood operates in their favour. Where polling-station-level analysis shows a near-perfect three-way split, zooming in reveals that social housing blocks (HLM, council estates, sociale woningen) tend to divide between left and populist right, that gated residences and private estates often favour the centre and centre-right, while surrounding housing developments — depending on whether they consist of terraced houses, semi-detached homes or detached villas — lean towards the left, the presidential majority or the populist right depending on the territory. It is precisely this granularity — the ability to distinguish between a cul-de-sac of terraced houses and a gated estate 200 metres away — that separates an effective ground campaign from random leafleting across the territory.

Optimising Strategic Campaign Decisions

Beyond voter targeting, electoral geomarketing informs other essential operational decisions.

Territorial composition of candidate lists. In municipal elections in France, Belgium (gemeente- or Gemeinderatswahlen), the Netherlands (gemeenteraadsverkiezingen) or Switzerland, where several dozen names must appear on a list, electoral geomarketing ensures balanced territorial representation. In proportional systems — such as Belgian or Swiss municipal elections where, for example, 9 names are needed for 9 seats — spatial analysis helps select candidates who are representative of the different neighbourhoods and electoral basins of the constituency.

Location of campaign offices and meeting points. Electoral geomarketing identifies optimal locations for pop-up offices, neighbourhood meetings (réunions de quartier, wijkvergaderingen), local rallies, town halls and small-group gatherings, maximising accessibility for targeted voters and candidate visibility in decisive areas.

The Digital Tools Ecosystem

Electoral geomarketing sits within a broader ecosystem of digital tools designed for political campaigns. In France, several players structure this market — Qomon, NationBuilder, or formerly 50+1 — offering campaign management solutions: activist CRMs, integrated platforms, electoral dashboards and canvassing apps. In the United Kingdom, tools like Ecanvasser and Campaign Lab serve similar functions. These tools generally work at the polling station or ward level.

In the United States, companies such as NGP VAN (on the Democratic side), i360 (Republican) and Catalist dominate the political micro-targeting market, drawing on national databases containing hundreds of millions of voter profiles enriched with thousands of behavioural and demographic data points.

High-precision electoral geomarketing occupies a complementary niche: where these platforms remain at the polling station or ward level, the artisanal geomarketeer produces a zoom of factor 15 to 30 within each polling station, distinguishing streets, cul-de-sacs, housing estates and individual buildings. This work, by nature more precise, more time-consuming and therefore more costly, does not replace these tools but complements them — providing the fine-grained spatial intelligence that no off-the-shelf platform can deliver.

Practical Applications: From Municipal to Legislative Elections

Municipal Elections: The Ideal Terrain

Municipal elections (élections municipales, gemeenteraadsverkiezingen, Kommunalwahlen, elezioni comunali, elecciones municipales) are the ideal terrain for electoral geomarketing. In a municipality, canton or constituency of 5,000 to 100,000 inhabitants, the number of voters is small enough for a well-targeted ground campaign to genuinely influence the result. Victory margins are often razor-thin: a few dozen to a few hundred votes frequently separate the leading lists in a runoff. In this context, knowing precisely which streets to prioritise for door-knocking (canvassing, démarchage, huis-aan-huisbezoek, Haustürwahlkampf) is not a luxury but a strategic necessity.

In practice, an electoral geomarketing analysis produces an electoral potential map tailored to the candidate’s socio-electoral positioning and the specificities of their territory, and sometimes a socio-electoral typology of different target electorates (young left-leaning professionals, centre-right retirees, working-class swing voters). A street-and-building prioritisation map is then produced, enabling the candidate and their volunteers to know exactly where to canvass first, which streets to cover if time permits, where to concentrate efforts between rounds, and which neighbourhoods to avoid wasting time in.

Legislative, Cantonal, Provincial and Regional Elections: The Battle of Swing Polls

In these intermediate-scale elections, the geographic scope changes but the logic remains identical. While national context and the social composition of a constituency’s electorate account for roughly 75% of a candidate’s score, the effects of local campaigning represent the remaining quarter: grassroots activism (militantisme de terrain, lokale campagnevoering), public meetings, relationships with local institutions and the candidate’s personal profile constitute a real margin of manoeuvre. In constituencies where the electorate is fragmented and multi-candidate runoffs multiply, this margin becomes decisive.

Ethical Issues and Limitations

Electoral geomarketing raises legitimate questions. If a candidate only targets potentially favourable voters, do they risk abandoning entire sections of their municipality? The answer is nuanced: the aim is not to abandon neighbourhoods but to optimise the intensity of presence. The entire municipality is covered, but decisive sectors receive multiple canvassing passes where already-won or definitively-lost areas receive only one.

This approach demands strict ethical standards. First, I never work for more than one candidate per municipality or constituency, to avoid any conflict of interest. Second, I guarantee absolute voter anonymity: analyses are based on spatial aggregates of a minimum of 10 voters (streets, buildings, blocks), never on identifiable individuals. Electoral geomarketing does not aim to build voter files but to inform the strategic allocation of campaign resources — resources that are by nature limited in time and space. In Europe, full compliance with GDPR and national data protection regulations is not optional but foundational to the methodology.

Conclusion: A Discipline with a Future

Electoral geomarketing stands at the crossroads of a rich intellectual tradition and a technological revolution that makes analyses of unprecedented precision possible. Siegfried wrote in 1913 that « experienced politicians are well aware of these solid factors of the electoral game. Activists observe without delving deeper that such-and-such a canton is ‘good’, another ‘bad’, and they act accordingly. » More than a century later, electoral geomarketing extends this intuition by providing it with a rigorous methodology and operational tools.

In political landscapes across Europe marked by electoral fragmentation, the erosion of traditional strongholds and the multiplication of tight races, the ability to identify decisive micro-territories is no longer a marginal advantage but a potentially determining factor.

The era when clinging to historical strongholds and hoping for the best was sufficient is over. Tomorrow’s victories, in a context of deep electoral, social and spatial fractures, will be built by micro-targeting swing streets within swing polls.

Electoral geomarketing is neither a technological gimmick nor a campaign luxury. In the hands of an experienced practitioner, it transforms data into strategic advantage and hundreds of hours of groundwork into votes won.

The future of this discipline will depend on its capacity to integrate methodological advances from academic research while maintaining its operational grounding. Progress in spatial analysis, growing access to fine-grained data and the emergence of new technical tools open promising perspectives. But the heart of the approach will always remain the same: understanding a territory in depth to act effectively on the ground — extending the tradition inaugurated over a century ago by André Siegfried in the countryside of western France.

Read more:

→ Services: Consulting, Mapping, Training and Conferences
→ About Geoffrey Pion
→ FAQ n°1 — Understanding Election Results in Your Territory
→ FAQ n°2 — Voter Targeting in Election Campaigns
→ FAQ n°5 — Using Past Election Results to Prepare Your Next Campaign

References

Bussi, M. (1998). Éléments de géographie électorale : à travers l’exemple de la France de l’Ouest. Rouen: Publications de l’Université de Rouen, 399 p.

Bussi, M., Le Digol, C. and Voilliot, C. (eds.) (2016). Le Tableau politique de la France de l’Ouest d’André Siegfried. 100 ans après. Héritages et postérités. Rennes: Presses universitaires de Rennes.

Guilluy, C. (2014). La France périphérique : comment on a sacrifié les classes populaires. Paris: Flammarion.

Issenberg, S. (2012). The Victory Lab: The Secret Science of Winning Campaigns. New York: Crown.

Johnston, R. and Pattie, C. (2006). Putting Voters in their Place: Geography and Elections in Great Britain. Oxford: Oxford University Press.

Le Bras, H. and Todd, E. (1981). L’Invention de la France. Atlas anthropologique et politique. Paris: Gallimard.

Le Bras, H. and Todd, E. (2013). Le Mystère français. Paris: Seuil.

Rodríguez-Pose, A. (2018). « The revenge of the places that don’t matter (and what to do about it). » Cambridge Journal of Regions, Economy and Society, 11(1), 189-209.

Siegfried, A. (1913). Tableau politique de la France de l’Ouest sous la Troisième République. Paris: Armand Colin. [Reissue: Brussels, Éditions de l’Université de Bruxelles, 2010.]