Doors, not people: the real shape of door-to-door demand in Ireland

01 · The unit problem

Population tells you the size of the prize. It tells you nothing about the cost of collecting it.

Ireland had 5,149,139 people on census night 2022 and 1,836,728 occupied private households. A door-to-door campaign cannot sell to a person. It sells to a household, through a door, at a location. So the planning unit is the door, and the constraint is how far apart the doors are.

We define the metric as doors per square kilometre: occupied private households divided by area. Because CSO publishes population density directly, the same figure can be derived as density × 0.98 ÷ average household size, where 0.98 is the share of the population living in private households nationally (5,046,681 of 5,149,139) and household size is the county average. Average household size is 2.74 nationally, but it ranges from 2.48 in Dublin City to 3.02 in Fingal and Meath — a 22 per cent swing that quietly moves door counts in the opposite direction to population in the commuter belt.

Dublin City has roughly 1,825 doors per km². The state average is 26. Leitrim is 8. That is a 230-fold range inside one small country.

Two counties can have the same population and completely different field economics. Two counties can have the same population density and different door density, because their households are different sizes. Neither of those facts shows up in a plan built on headcount.

03 · What density does to a rep-day

Geometry sets your cost per acquisition before anyone knocks.

Average distance between doors scales with the inverse square root of density. Quadruple the doors per km² and you halve the walk. That single relationship explains most of the productivity gap between an urban and a rural field team, and it is entirely predictable from census data.

The model below assumes a five-hour productive canvassing window, 2.5 minutes of fixed time per door (approach, knock, wait, conversation or no-answer) and a 5 km/h walking pace. It is illustrative — your own contact and conversion rates sit on top of it — but the shape holds.

Illustrative model. Doors attempted, not contacts made. Rural figures are optimistic: they exclude driveways, gates, and the drive time between clusters of houses, all of which worsen with distance.

Two things stand out. First, the productivity gap in attempted doors is roughly 2.4× between Dublin City and Leitrim — meaningful, but smaller than people expect, because fixed door-time dominates until spacing gets very wide. Second, and far more important, the territory gap is 230×. The same weekly door target that fits inside 0.3 km² in Dublin City needs 63 km² in Leitrim. That is not a canvassing problem, it is a routing, vehicle and supervision problem — and it is where rural D2D budgets actually break.

05 · Who actually opens the door

Density decides how many doors you reach. Demography decides how many open.

Three census variables move contact rate and conversion more than anything else in a field campaign, and all three are published per county.

People aged 65 and over

776,315 people, 15.1 per cent of the population, and the fastest-growing cohort in every single county. Growth since 2016 runs from 18 per cent in Kerry to 35 per cent in Kildare and 30 per cent in Meath. This group has the highest daytime presence at home of any cohort, which makes daytime canvassing viable in territories where it otherwise would not be — and it makes conduct, pace and consent handling a compliance issue, not a soft skill.

Households of one person

425,974 people live alone — roughly 23 per cent of all occupied private households. Dublin City alone has 61,525, Cork 49,705, Donegal 16,436, Tipperary 15,933, Galway County 15,615. Operationally these are single-decision-maker doors: no deferral to an absent partner, shorter sales cycle, higher same-visit close rate. In a territory plan they are worth weighting up.

Recent movers

265,098 usual residents moved home in the year to April 2022, in 112,548 households — about one household in sixteen. Seventy-one per cent moved within the same county. A move is the moment every household utility, waste, broadband and subscription decision is reopened at once, which is why new-build and recently-turned-over streets convert far above territory average. Dublin City recorded 32,910 within-county moves. Roscommon and Leitrim have the highest out-of-county outflow, at 46 and 45 per cent of movers respectively — a churn signal for anyone selling a recurring service in those counties.

06 · Tenure and product fit

Who owns the door determines what you can sell through it.

Nationally, of 1,541,557 households with a stated tenure: 44.1 per cent own outright, 34.5 per cent own with a mortgage, and 21.4 per cent rent privately. The county spread is wide enough to change which proposition you deploy where.

CSO Census 2022 Profile 2. Excludes local authority and voluntary/co-operative rented households, so shares are of stated private tenure rather than all households.

The practical read: solar, insulation, retrofit, alarms and anything fixed to the building belong in outright-owner territory — Kerry, Roscommon, Kilkenny, Carlow. Broadband, waste, energy switching, media subscriptions and insurance travel best in high-churn, high-rental territory — Dublin above all, then Cork and the regional cities. Deploying the wrong one into the wrong tenure profile produces a poor result that gets blamed on the channel.

07 · Data integrity

Four traps that will make your territory model wrong.

Most published Irish density figures are not comparable with each other. If you are building a commercial model on them, these four issues matter more than any analysis sitting on top.

1. Water-inclusive versus land-only area

CSO areas include water bodies. Third-party sites frequently do not. Dublin City is 4,618 persons/km² on the CSO basis and roughly 5,032 on a land-only basis — a 9 per cent difference in the same city. Both are defensible. Mixing them is not. Every figure in this report uses the CSO water-inclusive basis. If your field product needs true land-only door density, recompute all 31 rows from a single land-area source and label it as such.

2. De facto versus usually resident population

Wexford is 163,919 people on the de facto (present on census night) measure and 163,527 usually resident. Neither is wrong; they count different things. This report is de facto throughout, matching the CSO Summary Results and county press statements.

3. The Cork boundary change

The 2019 boundary extension moved roughly 85,000 people from Cork County into Cork City. Any 2016–2022 comparison for either area individually is invalid. Only the combined Cork total of 584,156 is like-for-like, and CSO itself frames comparisons at county-of-Cork level. Cork is footnoted or suppressed in every change calculation we publish.

4. Two national household totals

Occupied households appear as 1,841,152 in the Summary Results and 1,836,728 in Profile 2. The gap is 4,424 households, or 0.24 per cent — immaterial at county scale, material in a contract. This report uses the Profile 2 figure and states so.

08 · Applying it

How we turn this into a territory plan.

This is the sequence we run before a single rep is deployed on a client campaign, and it is the same sequence whether the market is Irish counties or English districts.

1. Fix one area basis and write it on every page. Water-inclusive or land-only, de facto or usually resident. Decide once, apply to all rows, disclose it. Most territory disputes are basis disputes.
2. Convert population to doors. Occupied private households, not people, not addresses, not postcodes. Use county household size, not the national average, wherever it is published.
3. Tier the market by doors per km² and assign a field model per tier — walk-only, estate-based, evening-weighted, cluster-routed or appointment-led. One national model applied to five different geographies is how D2D campaigns lose money.
4. Overlay the demographics that move contact rate — 65+ share for daytime viability, one-person households for decision speed, recent movers for switching windows.
5. Filter by tenure for proposition fit before assigning territories, not after the first disappointing week.
6. Drop to town level below 25 doors per km². County averages describe rural counties badly. Built-up area statistics, not county totals, are the planning unit once density falls.
7. Set rep targets from geometry, not from the best territory. A 60-door day in Mayo is a better performance than a 100-door day in Dublin City. Targets that ignore this punish your rural teams for their postcode.